NEUROTENSIN ENGAGES MESOLIMBIC DOPAMINE CIRCUITS TO REGULATE BODY WEIGHT By Hillary L. Woodworth A DISSERTATION Submitted to Michigan State University in partial fulfillment of the requirements for the degree of Physiology—Doctor of Philosophy 2017 ABSTRACT NEUROTENSIN ENGAGES MESOLIMBIC DOPAMINE CIRCUITS TO REGULATE BODY WEIGHT By Hillary L. Woodworth Body weight is determined by feeding and volitional physical activity behaviors that are regulated, in part, by dopamine (DA) neurons of the ventral tegmental area (VTA). Here, we sought to understand how the neuropeptide, neurotensin (Nts) engages VTA DA neurons to modify body weight. The rationale for this work is that pharmacologic application of Nts into the VTA suppresses food intake and promotes locomotor activity, yet the endogenous circuits by which Nts acts on the VTA to modify these behaviors and body weight remain unclear. First, we identified the endogenous sources of Nts input to the VTA; using retrograde tracing we found that the lateral hypothalamic area (LHA), a critical neural hub for coordinating energy balance, provides substantial Nts projections to the VTA. We next examined how Nts directly engages VTA DA neurons by identifying Nts receptor-expressing cells in the VTA. To do this, we generated mice expressing Cre-recombinase in Nts receptor 1 (NtsR1) or Nts receptor 2 (NtsR2) cells, which revealed that NtsR1 is expressed on many VTA DA neurons, whereas NtsR2 is predominantly restricted to glial cells. Furthermore, only the VTA NtsR1 neurons project to the nucleus accumbens (NA), where DA release is known to modify feeding and locomotor behavior. We therefore tested the physiologic necessity for Nts action via the VTA by genetically ablating VTA NtsR1 neurons. Mice lacking VTA NtsR1-DA neurons were hyperactive, failed to gain weight, and could not appropriately coordinate feeding behavior with peripheral energy cues, demonstrating that VTA NtsR1 neurons are essential for energy balance. Finally, we tested the hypothesis that endogenous Nts input from the LHA to the mesolimbic DA system would be sufficient to regulate body weight. Indeed, chemogenetic activation of LHA Nts neurons increased physical activity, restrained food intake, and promoted weight loss in lean mice. Interestingly, the anorectic effects of LHA Nts activation were mediated via NtsR1 and DA signaling, while the physical activity was NtsR1-independent. Furthermore, in hungry mice (a state in which increased appetitive drive can promote overeating and weight gain), activation of LHA Nts neurons suppressed intake of chow and palatable sucrose rewards. Collectively, this work defines an endogenous LHA Nts circuit that engages the mesolimbic DA system via NtsR1 to suppress food intake in both energy replete and energy depleted states. Enhancing action via this circuit may thus be useful to support dual weight loss behaviors in an obesogenic environment. To my family, labmates, and mentors iv ACKNOWLEDGEMENTS First, I would like to thank Dr. Gina Leinninger for her unremitting support, encouragement and enthusiasm over the last five years. Not only are you my mentor, but you are also an authentic role model, leader, and teacher whom I admire for your communication skills and respect for others. Thank you for showing me how much fun science can be, for being my advocate, and for always helping me feel like my work was valued. It’s been an honor and a privilege to be your graduate student. I also want to thank Dr. Steve Heidemann, a long-time mentor whom I first met as an undergraduate and who connected me to Gina Leinninger. Your kind, sage, and honest advice helped guide my career trajectory toward the right track on multiple occasions and I can’t express how much I appreciate it. I would like to thank Drs. Michelle Mazei-Robison and A.J. Robison for always having open doors and being willing to help me with experiments and give advice. I would also like to thank my thesis committee members, Drs. Michelle Mazei-Robison, Cheryl Sisk, Alex Johnson, and Bob Wiseman for helpful suggestions, discussion, and support. I am indebted to three talented undergraduate students who made substantial contributions to the project: Hannah Batchelor, Trevor Lewis and Bethany Beekly. Thank you for your willingness to learn, constant curiosity, and attention to detail. Your hard work made a huge difference and I couldn’t be more proud to have been your mentor. I would like to thank all the members of the Leinninger lab—Raluca Bugescu, Juliette Brown, Gizem Kurt, Patricia Perez-Bonilla, and Laura Schroeder, for being my support system and treating me to some of the best philosophical conversations with many laughs over food and drinks. Much of this work would not have been possible without Sandra O’Reilly—I truly appreciate how consistent and thorough you were with my metabolism experiments. Many members of the Mazei-Robison and Robison labs helped me on numerous occasions with behavioral experiments, particularly v Andrew Eagle and Ken Moon. I am so grateful for the hours you spent helping me set up equipment, answering my questions, and looking at my data. None of this work would have possible without the animal care staff—thank you for taking such good care of our mice. I’d also like to thank Susanne Mohr and Cindy Arvidson for providing me with support and guidance as a student in the Physiology PhD Program and MSU MD/PhD Program. Finally, I would to express my gratitude for the entire Physiology faculty and staff who have been continuously helpful, friendly, and interested. Thank you for making me feel at home for the past five years. This work was supported by the NIDDK: HLW (F30-DK107163) and GML (R01-DK103808). vi TABLE OF CONTENTS LIST OF FIGURES ...................................................................................................................ix KEY TO ABBREVIATIONS .......................................................................................................xi Chapter 1. A Weighty Introduction: Neurotensin and Dopamine Circuits that Regulate Energy Balance ........................................................................................................................1 1.1 Background and Significance ...........................................................................................1 1.2 Coordination of Energy Balance ...................................................................................... 3 1.2.1 What is Energy Balance and How Does it Determine Weight? .............................. 3 1.2.2 The Brain Coordinates Energy Balance .................................................................5 1.2.3 The Lateral Hypothalamic Area (LHA) is Essential for Energy Balance .................. 7 1.3 Dopamine Signaling and Body Weight ............................................................................ 9 1.3.1 Dopamine Circuitry, Signaling, and General Function ............................................ 9 1.3.2 Dopamine and Feeding .......................................................................................12 1.3.3 Dopamine and Energy Expenditure .....................................................................16 1.3.4 Neuropeptide Regulation of Dopamine Signaling .................................................17 1.4 Neurotensin Physiology ..................................................................................................19 1.4.1 Discovery, Receptors, and Expression Patterns ..................................................19 1.4.2 Neurotensin and Dopamine Signaling ..................................................................21 1.4.3 Neurotensin and Energy Balance ........................................................................22 1.5 Neurotensin and Dopamine Signaling in Energy Balance Disorders ........................... 24 1.5.1 Is Dopamine Disrupted in Obesity? ......................................................................24 1.5.2 Evidence for Neurotensin in Obesity ....................................................................28 1.6 Goals of the Dissertation .................................................................................................29 REFERENCES ........................................................................................................................33 Chapter 2. Determination of Neurotensin Projections to the Ventral Tegmental Area in Mice ..........................................................................................................58 2.1 Abstract ............................................................................................................................58 2.2 Introduction ......................................................................................................................59 2.3 Results ..............................................................................................................................61 2.4 Discussion ........................................................................................................................68 2.5 Methods ............................................................................................................................74 APPENDIX ...............................................................................................................................78 REFERENCES .........................................................................................................................91 Chapter 3. Identification of Neurotensin Receptor Expressing Cells in the Ventral Tegmental Area Across The Lifespan .................................................................................105 3.1 Abstract ..........................................................................................................................105 3.2 Introduction ....................................................................................................................106 3.3 Results ............................................................................................................................108 3.4 Discussion ......................................................................................................................113 3.5 Methods ..........................................................................................................................117 APPENDIX .............................................................................................................................121 REFERENCES .......................................................................................................................129 vii Chapter 4. Neurotensin Receptor-1 Identifies a Subset of Ventral Tegmental Dopamine Neurons that Coordinate Energy Balance ......................................................... 137 4.1 Abstract ..........................................................................................................................137 4.2 Introduction ....................................................................................................................137 4.3 Results ............................................................................................................................139 4.4 Discussion .....................................................................................................................145 4.5 Methods ..........................................................................................................................151 APPENDIX .............................................................................................................................159 REFERENCES .......................................................................................................................170 Chapter 5. A Central Neurotensin Circuit That Coordinates Weight Loss Behaviors ..... 176 5.1 Abstract ..........................................................................................................................176 5.2 Introduction ....................................................................................................................176 5.3 Results ............................................................................................................................178 5.4 Discussion ......................................................................................................................185 5.5 Methods ..........................................................................................................................189 APPENDIX .............................................................................................................................196 REFERENCES .......................................................................................................................208 Chapter 6. Summary, Discussion, and Translational Implications.................................... 213 6.1 Summary of Dissertation ...............................................................................................213 6.2 Discussion .......................................................................................................................214 6.2.1 Technical Considerations of Transgenic vs. Knock-in Models to Study VTA NtsR1 Neurons ...................................................................................................214 6.2.2 Phenotype of Mice Lacking VTA NtsR1 Neurons ............................................... 215 6.2.3 Chemogenetic Activation of LHA Nts Neurons: Limitations and Future Directions ........................................................................................................217 6.3 Translational Implications ..............................................................................................219 6.3.1 Neurotensin Circuits as a Translational Target .................................................. 219 6.3.2 Final Thoughts on the Challenges of Addressing Obesity .................................. 221 REFERENCES .......................................................................................................................223 viii LIST OF FIGURES Figure 1. Schematic of hypothesized neural circuit by which Nts acts to modify mesolimbic DA signaling and body weight ..........................................................................32 Figure 2. Local Nts expression in the VTA ...........................................................................79 Figure 3. Validation of VTA targeting in NtsCre;GFP mice injected unilaterally with FG .... 80 Figure 4. The NA shell contains clusters of Nts neurons that project to the VTA ............. 81 Figure 5. Cortical Nts inputs to the VTA originate in the CeA but not neocortex ............. 82 Figure. 6. Nts neurons in the IPAC, pallidum, and ST that project to the VTA ................... 83 Figure 7. Sub-regions of the hypothalamus provide Nts afferents to the VTA .................. 84 Figure 8. Brainstem Nts inputs to the VTA ...........................................................................86 Figure 9. Nts-ergic VTA inputs from the LHb and RMTg ....................................................88 Figure 10. Quantification of Nts-expressing afferents to the VTA by anatomical sub-region ..............................................................................................................................89 Figure 11. Schematic illustration of Nts afferents to the VTA ............................................ 90 Figure 12. Generation of mouse models to identify developmental vs. adult expression patterns of NtsR1 and NtsR2 ........................................................................... 122 Figure 13. Developmental vs. adult expression of NtsR1 and NtsR2 in the VTA ............ 123 Figure 14. NtsR1 and NtsR2 colocalization with TH in development compared to adulthood .............................................................................................................................124 Figure 15. Ad-syn-mCherry reveals projections of VTA NtsR1 neurons ......................... 126 Figure 16. Projections of VTA NtsR1 neurons .................................................................. 127 Figure 17. Generation of NtsR1Cre;GFP and NtsR2Cre;GFP reporter mice to visualize NtsR1 and NtsR2 neurons in the VTA .................................................................................160 Figure 18. Distribution and neurochemical phenotype of VTA NtsR1 and NtsR2 neurons .................................................................................................................................161 Figure 19. VTA NtsR1 neurons project to the ventral striatum ........................................ 162 Figure 20. Loss of VTA NtsR1 neurons disrupts energy balance .................................... 163 ix Figure 21. Water intake and additional metabolic parameters in mice lacking VTA NtsR1 neurons .....................................................................................................................165 Figure 22. VTA NtsR1 neurons modulate hedonic and motivated ingestive behavior ... 166 Figure 23. Loss of VTA NtsR1 neurons alters response to amphetamine ...................... 167 Figure 24. Anxiety-like behavior assessed via elevated-plus maze ................................ 168 Figure 25. Ablation of VTA NtsR1 neurons decreases markers of DA signaling ............ 168 Figure 26. Virally-induced vs. endogenous GFP reporter expression for identification of VTA NtsR1 neurons .........................................................................................................169 Figure 27. Examination of the LHA NtsVTA circuit in WT and NtsR1KO mice ............ 197 Figure 28. Acute activation of LHA Nts neurons promotes energy expenditure and suppresses feeding .............................................................................................................198 Figure 29. Additional acute metabolic data in WT and NtsR1KO mice ........................... 199 Figure 30. Acute NtsR1 or D1R blockade abolishes LHA Nts-induced suppression of feeding ..................................................................................................................................200 Figure 31. Food intake separated by circadian period with NtsR1 or D1R antagonists ............................................................................................................................201 Figure 32. Chronic activation of LHA Nts neurons induces mild weight loss in chow-fed lean mice ..............................................................................................................202 Figure 33. Chronic VEH or CNO injection in WTChR controls ........................................... 202 Figure 34. Repeated activation of LHA Nts neurons does not induce weight loss in obesity ..................................................................................................................................203 Figure 35. Chronic activation of LHA Nts neurons does not prevent weight gain in lean mice on HF diet ....................................................................................................................204 Figure 36. Activation of LHA Nts neurons suppresses fasting-induced ad lib and motivated food intake ..........................................................................................................205 Figure 37. Learning acquisition during operant training .................................................. 206 Figure 38. Assessment of reward, anxiety, and repetitive behaviors associated with LHA NtsNtsR1 circuit activation .............................................................................. 207 Figure 39. LHA Nts neurons restrain feeding and promote locomotor activity, in part via NtsR1 and the mesolimbic DA system .............................................................. 222 x KEY TO ABBREVIATIONS 3N oculomotor nucleus 3V third ventricle 4V fourth ventricle aca anterior commissure AHA anterior hypothalamic area APTN anterior pretectal nucleus aq cerebral aqueduct ARC arcuate nucleus ATg anterior tegmental nucleus BLA basolateral amygdala cc corpus callosum CeA central amygdala ChR channel rhodopsin CNO clozapine-N-oxide cp cerebral peduncle CPP conditioned place preference CPu caudate/putamen D1R dopamine receptor 1 D2R dopamine receptor 2 D3V dorsal third ventricle DA dopamine DAT dopamine transporter DMH dorsomedial hypothalamus DR dorsal raphe DREADD Designer Receptors Exclusively Activated by Designer Drugs xi DTg dorsal tegmental nucleus EA extended amygdala ec external capsule En endopiriform cortex f fornix FG fluorogold fr fasciculus retroflexus GABA gamma-aminobutyric acid GFAP glial fibrillary acid protein GFP green fluorescent protein GP globus pallidus ic internal capsule ICV intracerebroventricular IL infralimbic cortex ip interpeduncular nucleus i.p. intraperitoneal IPAC interstitial nucleus of posterior limb of anterior commissure ISH in situ hybridization LDTg laterodorsal tegmental nucleus LepRb leptin receptor LHA lateral hypothalamic area LHb lateral habenula LO lateral orbital cortex LPO lateral preoptic area LS lateral septum lv lateral ventricle xii MCH melanin-concentrating hormone MCR-4 melanocortin receptor-4 MHb medial habenula ml medial lemniscus MPN medial preoptic nucleus MPO medial preoptic area MS medial septum α-MSH alpha-melanocyte stimulating hormone mt mammillothalamic tract NAc nucleus accumbens core NAsh nucleus accumbens shell ns nigrostriatal tract Nts neurotensin NtsR1 neurotensin receptor 1 NtsR2 neurotensin receptor 2 OFT olfactory tubercle ot optic tract OX orexin/hypocretin PAG periaqueductal gray PBN parabrachial nucleus pc posterior commissure PFC prefrontal cortex POA preoptic area PPTg pedunculopontine tegmental nucleus PrL prelimbic cortex pSTN para subthalamic nucleus xiii PVH paraventricular hypothalamus RER resting energy expenditure RMTg rostromedial tegmental nucleus RLi raphe linear nucleus SC superior colliculus scp superior cerebral peduncle SN substantia nigra SPTg subpedunculuar tegmental nucleus ST stria terminalis STN subthalamic nucleus TEA thermic effect of activity TEF thermic effect of food TH tyrosine hydroxylase ts tectospinal tract VMH ventromedial hypothalamus VP ventral pallidum VTA ventral tegmental area ZI zona incerta xscp decussation of superior cerebellar peduncle xiv Chapter 1. A Weighty Introduction: Neurotensin and Dopamine Circuits that Regulate Energy Balance 1.1 Background and Significance Global rates of obesity have grown significantly over the past three decades with >1.9 billion adults qualifying as overweight or obese in 20151,2. Obesity is a particularly serious problem in the U.S., where almost 70% of adults have a BMI in the overweight or obese category (>25 kg/m2) and minority populations are disproportionately affected3. U.S. childhood obesity rates have also sky-rocketed, with 20% of children ages 6-19 qualifying as obese nation-wide4,5 and rates upwards of 32% are present in certain locations such as the South Bronx6. Obese children are 80% more likely to be obese as adults and have increased risk of early death7, and in fact, increased body mass has now surpassed undernutrition as a top cause of death and disability worldwide8. Furthermore, being overweight increases risk for developing chronic conditions such as type-2 diabetes, heart disease, stroke, cancer, fatty liver disease, and kidney disease, conditions that are expensive to treat and reduce quality of life9,10. The high incidence of obesity already exerts a considerable toll on healthcare systems worldwide, but the rising costs associated with treating obesity/overweight and concomitant complications will be unsustainable without intervention11. Curtailing the obesity pandemic is thus a global health concern of paramount importance. Despite the overwhelming prevalence of obesity and overweight, few medical strategies to date have proven effective in maintaining long-term weight loss. The first-line prescription for weight reduction is diet and exercise, which has been capitalized by the U.S. weight loss industry, a market worth $60 billion in 201512. While many individuals initially lose weight through dieting, most do not maintain the weight loss long-term13. A regimen of both diet and 1 exercise is more likely to provide long-term benefit13, however adherence to such lifestyle modification is notoriously challenging. Given the high likelihood of weight regain, many dieters enter a vicious cycle of weight cycling or “yo-yo dieting” which increases risk of heart disease, stroke, and diabetes14. Thus, not only are lifestyle modifications largely ineffective, but repeated dieting and weight cycling imposes adverse health effects. A handful of pharmacologic agents have been developed to aid weight loss, but most have side effects and not all are approved for long-term use15. The effectiveness of obesity medications is typically modest, with most agents achieving 5% loss of body weight at 1 year16,17. While 5% weight loss can reduce risk of obesity-related complications, it is not enough to achieve a healthy BMI for most patients. Additionally, few studies have examined outcomes including potential weight regain after patients stop taking the medication, thus the long-term effectiveness of anti-obesity drugs is largely undetermined. Currently, the most effective obesity treatment is bariatric surgery, with the Roux-en-Y gastric bypass procedure (RYGB) producing an average loss of 5 BMI points at 5 years18,19. In addition to substantial weight loss, bariatric surgery also alleviates obesityrelated type 2 diabetes, hypertension, and joint pain18, plus patients report increased quality of life after surgery20. The major drawbacks of bariatric surgery include the risk of complications and the cost which averages $15,000-$25,00021,22. Due to this, surgery is usually reserved as a last resort for the morbidly obese (BMI>40 kg/m2), which comprise less than 10% of overweight or obese adults. Thus, moderately overweight or obese individuals who are still at risk for developing obesity-related complications have limited options for weight control (diet and exercise or pharmacotherapy), which provide only modest benefit. Therefore, there is a clear need to develop better strategies to promote weight loss and prevent weight gain to improve health outcomes for overweight and obese individuals. It is also vital to understand why the surge in overweight and obesity has occurred, as this may inform design of interventions to manage body weight. Average caloric intake has 2 increased by around 240 kcals/day since 1970, with most of the increase attributed to carbohydrates23. Interestingly, fat intake decreased over the same time period as obesity rates continued to rise, suggesting that excess caloric intake, not high dietary fat consumption, potentiates weight gain. Further, occupational-associated energy expenditure has progressively declined since 196024 and only 1 in 5 adults fulfills the recommend amount of daily physical activity25. Thus, one would speculate that the obesity epidemic is fueled by excess caloric intake combined with reduced physical activity. While this certainly explains what causes obesity, it does not explain why individuals overeat and move less. The increasing availability of palatable, calorie-rich, and inexpensive food has fueled obesity rates26, but what permits caloric intake in excess of metabolic demands? Under normal circumstances, energy intake is exquisitely coordinated with energy expenditure in an effort to defend body weight from both loss and gain27,28. However, these mechanisms are not completely understood and are impacted by numerous variables (i.e. food palatability, genetics, sedentary lifestyle) that contribute to the development of obesity. The brain plays an essential role in coordinating motivated behaviors such as feeding and voluntary physical activity, but the neural circuits involved remain poorly understood. The purpose of this chapter is thus to provide an overview of energy balance physiology and describe how disruption of dopamine-mediated feeding and physical activity behaviors contributes to the development of obesity. We will also evaluate how the neuropeptide, neurotensin (Nts) may engage dopamine circuits, and hence the potential of the Nts signaling system as a novel target for weight control. 1.2 Coordination of Energy Balance 1.2.1 What is Energy Balance and How Does it Determine Weight? The term “energy balance” is used to describe the relationship between energy intake and expenditure that determines body weight. Energy intake consists of calories consumed 3 through food and liquids. Energy expenditure refers to the calories burned to support basal metabolism and voluntary physical activity. For most individuals, energy expenditure is the sum of resting metabolic rate (RMR), the thermic effect of feeding (TEF), and the thermic effect of activity (TEA)29. RMR comprises 60-75% of total energy expenditure and is the energy required by the body to perform basic physiologic functions, or more simply, “the number of calories an individual would use if he/she stayed in bed all day.” TEF accounts for 10% of energy expenditure and is the energy required for digesting food. TEA can account for 15-30% of energy expenditure and refers to the additional calories burned through volitional activity and exercise29. When energy intake exceeds expenditure it creates a caloric surfeit, or “positive energy balance”, which can be stored in the body as fat and lead to weight gain. Conversely, when energy expenditure exceeds caloric intake, the body experiences a caloric deficit or “negative energy balance”; as a result, calories required to support survival are obtained from adipose reserves, leading to weight loss. At face value, energy balance appears to be a simple math equation, but its coordination is complex, requiring continuous communication between the periphery (to sense energy status) and the brain (to appropriately modulate energy intake and expenditure). Energy balance is intimately tied to the idea of a body weight “set-point” wherein genetic and environmental factors determine an individual’s body weight, which is defended through homeostatic mechanisms that compensate for positive or negative energy balance28. For example, in controlled over-feeding studies, total energy expenditure increases and appetite decreases as the body attempts to deplete the caloric surplus30,31. Similarly, weight loss leads to increased appetite and reduced energy expenditure that drives recovery of lost body weight and indeed, most people who lose weight gain it back32-36. Cota et al. provide a salient example of just how tightly the body coordinates energy balance: the average adult male uses around 900,000 calories per year and only gains (on average) one pound or 3600 extra calories during 4 that time, which amazingly amounts to >99% accuracy in matching food intake with energy expenditure27. Although energy balance is exquisitely fine-tuned in the short-term, small, incremental weight gain over years is thought to contribute to obesity. As such, the body weight set-point slowly drifts upward with time28, and losing weight after being overweight or obese for several years is extremely challenging as the body strives to defend a heavier set-point35,37. A prime example of this physiology comes from study of obese participants on the TV show “The Biggest Loser”: contestants lost >120 pounds on average during the show, but 6 years later, individuals had regained approximately 2/3 of the weight and their energy-expenditure was significantly decreased compared to what would be expected for their body weight38. Thus overweight individuals who have lost weight are constantly battling increased hunger and reduced energy expenditure as their bodies defend their heavier weight set-point. 1.2.2 The Brain Coordinates Energy Balance Energy balance strongly relies on behavioral output, namely feeding and volitional activity that are controlled by the brain. Interestingly, most genes implicated in obesity have enriched expression in the nervous system39, supporting the role of the brain as a master regulator of body weight. In particular, sub-regions of the hypothalamus have been implicated in regulating energy balance, including the arcuate nucleus (ARC), paraventricular nucleus (PVN), ventromedial hypothalamus (VMH) and the lateral hypothalamic area (LHA). The hypothalamus is found at the base of the brain near the third ventricle and is well positioned to intercept circulating energy cues from the periphery. Important hormonal cues detected by the hypothalamus include the anorectic hormone leptin and the orexigenic hormone ghrelin. Leptin is secreted in proportion to adipose tissue as a signal of long-term energy storage and acts on hypothalamic nuclei to suppress food intake40. Ghrelin, by contrast, is secreted from the stomach as hunger increases and acts on the hypothalamus and other brain areas to promote food intake41. The ARC is a key site for energy integration and has been extensively studied. 5 The ARC contains two discrete neuronal populations expressing either agouti-related peptide (AgRP) or promelanocortin (POMC), which exert opposing actions on feeding and body weight42. AgRP neurons are activated by physiologic hunger and promote food intake while reducing energy expenditure43,44. AgRP neurons also express neuropeptide Y (NPY) and GABA, and the contribution of each individual neurotransmitter has been shown to increase feeding45. AgRP neurons are active in a fasted state, which is potentiated by ghrelin-mediated excitatory input46,47. After a meal, ghrelin levels fall and leptin becomes a dominant circulating signal of energy status, which inhibits AgRP neurons to reduce food intake48. By contrast, neighboring POMC neurons are activated during satiation and suppress food intake while increasing energy expenditure49-51. POMC is a precursor protein that is cleaved into alpha melanocyte stimulating hormone (α-MSH) and exerts anorectic effects by binding the melanocortin-4 (MCR-4) receptors at key brain sites. Rodents deficient in MCR-4 signaling are hyperphagic and obese52,53 and MCR-4 mutations are the most common monogenic cause of human obesity54,55, underscoring the importance of melanocortin signaling in energy balance. While the ARC is a critical center for direct sensing of peripheral energy cues, POMC and AgRP neurons likely feed into a variety of downstream circuits that fine-tune feeding behavior and energy expenditure such as the lateral hypothalamic area (LHA). Indeed, POMC neurons project heavily to the LHA which expresses MCR-451,56, but the functional role of melanocortin signaling in the LHA has not been characterized. For the purposes of this chapter, we will focus on the LHA as a hypothalamic center that plays an essential, yet incompletely understood role in energy balance by both receiving input from the ARC and directly sensing peripheral energy cues. 6 1.2.3 The Lateral Hypothalamic Area (LHA) is Essential for Energy Balance The LHA was initially described as a “feeding center” when investigators found that electrical stimulation induced voracious feeding behavior in sated animals57,58. Conversely, lesions of the LHA produced such profound self-induced starvation that animals had to be forcefed to maintain survival58-60. Subsequent neuroanatomical studies revealed that the LHA provides efferents to numerous brain regions, but it provides particularly dense projections to dopamine (DA) neurons in the ventral tegmental area (VTA) that modify the motivation to obtain pharmacological and natural rewards (e.g. food, physical activity, and sex)61. Indeed, electrical stimulation of the LHA is rewarding and increases DA release in nucleus accumbens (NA)57,62-64. Thus, the LHA directly accesses DA-ergic circuitry that controls motivated behaviors, and provides an explanation for why lesion of the LHA abolishes the motivation to eat. Additionally, the LHA receives information concerning energy status that may be important for appropriately coordinating feeding and other motivated behaviors. Some LHA neurons express receptors for leptin while others respond to ghrelin65,66, indicating that the LHA can directly intercept circulating anorectic and orexigenic cues. The LHA also receives dense input from the ARC51, and thereby receives indirect information regarding peripheral energy status. Taken together, this work suggests that the LHA is uniquely positioned to integrate peripheral energy cues with DA-mediated motivated behaviors that may promote or suppress feeding. Given that the LHA responds to both anorectic and orexigenic cues, there are likely distinct neural mechanisms by which the LHA can coordinate motivated behaviors and energy balance. Indeed, several populations of neurons have been described in the LHA that vary in neurotransmitter and neuropeptide content, projection targets, and function67,68. The contributions of the key neuropeptide-defined neuronal populations in the LHA with respect to energy balance will be briefly discussed below. 7 Orexin: Many neurons in the LHA express the orexigenic neuropeptide hypocretin/orexin (OX)69. As its name implies, OX stimulates food intake but also plays a critical role in wakefulness and arousal68,70. As such, DREADD-mediated activation of OX neurons increases feeding, locomotor activity, and energy expenditure71. OX neurons are activated by cues of energy depletion including fasting, ghrelin, and low glucose66,72,73 and thus act to promote food intake and arousal necessary to obtain food when peripheral energy stores are low. Although OX-expressing neurons are only found within in the LHA, they project widely throughout the brain, including to the VTA74,75. OX neurons also co-express markers of glutamatergic neurons76 and therefore can excite afferent projection targets such as VTA DA neurons75. Melanin-concentrating hormone: A separate subset of LHA neurons contains melaninconcentrating hormone (MCH), which similar to OX, promotes food intake and weight gain77,78. Instead of stimulating arousal however, MCH supports sleeping behavior79. MCH neurons express both glutamatergic and GABAergic markers, and thus may be divided into further functional subsets based on excitatory vs. inhibitory potential79-81. While MCH neurons do not project directly do the VTA, many instead project directly to GABA-ergic neurons in the NA that provide tonic inhibitory input on food intake82-84. Thus GABA-ergic MCH neurons may dis-inhibit the NA, removing the “brake” on feeding to promote food intake67. Neurotensin: A third molecularly distinct population of LHA neurons expresses the neuropeptide, neurotensin (Nts)66,85. Unlike the orexigenic effects of MCH and OX, pharmacologic Nts suppresses feeding and promotes weight loss86-90, however the functional role of LHA Nts neurons has not been completely defined. A subset of LHA Nts neurons expresses the long form of the leptin receptor, are activated by leptin, and co-express GABA65,85. Some LHA Nts neurons project locally to and inhibit neighboring OX neurons91 while 8 others project to the VTA66,85, where pharmacologic Nts administration produces anorexia and locomotor activity86,88,92. Consistent with this, developmental loss of leptin receptor signaling specifically from LHA Nts neurons results in reduced striatal DA action, increased adiposity, and disrupted feeding response to leptin66,85. Nts is an established modulator of DA signaling (see section 1.4.1) and the LHA may provide an endogenous source of Nts input to the VTA that impacts motivated behaviors. In support of this, it was recently demonstrated that chemogenetic activation of LHA Nts neurons increased mesolimbic DA release, locomotor activity, and reduced body weight93. Thus, Nts and DA action may be useful to promote weight loss behaviors, while disruption of Nts-mediated DA signaling might contribute to the development of obesity. Understanding how Nts engages the DA system to modify motivated behaviors, including the specific role of LHA Nts neurons, will be important for discerning the therapeutic potential for Nts in regulating energy balance. 1.3 Dopamine Signaling and Body Weight 1.3.1 Dopamine Circuitry, Signaling, and General Function Dopamine (DA) is a catecholamine neurotransmitter that is essential for movement and motivation94, and DA disruption is a hallmark of several neuropsychiatric diseases. Neurons that synthesize DA are found primarily in the VTA and substantia nigra pars compacta (SN) of the midbrain. DA synthesis begins with the amino acid, tyrosine, which is converted to L-DOPA by tyrosine hydroxylase (TH), the rate-limiting step in DA synthesis. L-DOPA is rapidly converted to DA by DOPA-decarboxylase, and DA is stored in highly concentrated vesicles until an action potential initiates their release95. DA binds to five receptor isoforms, however most work has focused on the DA receptor 1 and 2 isoforms (D1R and D2R), both of which are expressed on GABA-ergic medium spiny neurons (MSNs) in the NA, and all are coupled to G9 proteins. D1R is coupled to Gα which increases protein kinase A (PKA) activity through induction of adenyl cyclase and cyclic adenosine monophosphate (cAMP), while D2R is coupled to Gi which essentially has the opposite effect and inhibits PKA activity96-98. PKA regulates a variety of downstream targets including AMPA and NMDA receptors, CREB, and DARP-3299, and the outcome of DA receptor signaling therefore includes changes in glutamatergic regulation, ion channel phosphorylation, and gene transcription100. Interestingly, D2R is also expressed presynaptically on DA neuron terminals and dendrites and acts as an inhibitory autoreceptor that decreases DA synthesis and release101-103. After DA is released into the synaptic cleft and binds to DA receptors, it is taken back up into the pre-synaptic terminal by the DA transporter (DAT), where it can then be recycled back into vesicles or degraded (see104,105 for review). VTA DA neurons project to the ventral striatum and prefrontal cortex (the mesolimbic and mesocortical pathways, respectively), whereas SN DA neurons project to the caudate and putamen (CPu), collectively referred to as the dorsal striatum (the nigrostriatal DA pathway). Generally speaking, the mesolimbic circuit controls motivation, while the nigrostriatal circuit regulates movement, hence the degeneration of nigrostriatal DA neurons in Parkinson’s disease causes loss of motor function. While the VTA and SN may exert some overlapping contributions to motivation and movement106, we will focus on role of VTA mesolimbic signaling in energy balance because it is implicated in regulating motivated feeding and locomotor behaviors that alter body weight. DA has often been suggested to signify reward, however some have argued that this term is ambiguous and obfuscates the actual role of DA in behavior107-110. According to Berridge and Kringelbach, reward is a complex term that encompasses aspects of pleasure, reinforcement, and learning, all processes that are mediated by distinct neural circuits that may or may not involve DA. They suggest that procurement and consumption of rewards, such as food, sex or drugs, are mediated by at least two distinct neural mechanisms which are 10 described in terms of “liking” and “wanting”. “Liking” is analogous to the positive feeling or pleasure evoked during physical contact with the reward, while “wanting” refers to behavioral investment required to attain the reward. Although DA has been popularized as the “pleasure” neurotransmitter, pioneering work by Berridge et al. called this into question when they showed that DA-depleted rats have intact hedonic or “pleasure” responses to sweet taste111. This suggests that DA is not necessary for the “liking” of rewards, and has since been supported by further work showing that hedonic reactions are not impacted by interruption of DA signaling109,110. Instead, DA mediates the effort an animal is willing to exert to attain a reward, or “wanting.” Thus, DA depletion does not alter hedonic response to a reward, but it does reduce willingness to work for the reward or responsiveness to cues that predict the reward112,113. Salamone and Correa propose a similar model narrowed to mesolimbic DA release, whereby the behaviors necessary to approach the reward (analogous to “wanting”) require DA release in the NA, whereas the direct interaction with the reward (analogous to “liking”) is independent of NA DA61. Both the models of Berridge et al. and Salamone et al. align with the involvement of DA in reward prediction error, the observation that DA neurons fire based on the perceived expectations of how “rewarding” a particular stimulus will be. As described by Schultz, DA neurons fire in response to an unexpected reward and are inhibited when the reward is omitted112. In a sense, DA neurons code the “error” between perceived and actual expectations, thus explaining why disruption of DA signaling may interfere with the perceived value of a reward, not necessarily consumption of the reward itself112,113. A final note with respect to VTA DA neurons is that they are molecularly heterogeneous and can be divided into subsets based on their electrophysiological properties, projection sites, co-expression of classical neurotransmitters, and response to neuropeptide modulators114. While VTA DA neurons heavily project to the NA, they also project to the prefrontal cortex 11 (PFC), amygdala and hindbrain, and DA release in each of these areas plays unique roles in behavior115. For instance, VTA DA neurons projecting to the NA are activated by rewards and promote positive reinforcement, while VTA DA neurons projecting to the PFC are activated by aversive cues and promote conditioned-place aversion116-118. Thus, VTA DA signaling does not purely regulate pursuit of pleasant stimuli but is also processes aversive stimuli. 1.3.2 Dopamine and Feeding The idea that DA is essential for movement and motivation came from early work showing that destruction of midbrain DA neurons leads to hypotonia and hypophagia119,120. Consistent with these observations, mice genetically lacking DA fail to gain weight and starve to death by 3 weeks of age unless DA is restored pharmacologically94. Since feeding requires physical movement, it was difficult to ascertain whether DA-deficient mice were hypophagic due to a primary feeding deficit or because of generalized impairment of motor function necessary to procure food. The latter explanation is supported by observations that animals with impaired DA signaling, either brain-wide or specifically to the dorsal striatum, ate less due to defective forepaw usage121,122. Furthermore, selective disruption of DA signaling in the NA does not impair free food intake, but instead reduces lever-pressing for food, which becomes more pronounced as the amount of work necessary to attain food increases123-128. These data suggest that NA DA release is not necessary for ad libitum feeding, but instead regulates the effort (e.g. work) an animal will exert to obtain food. Salamone and colleagues examined this using a food choice test, wherein animals were given the option of either lever-pressing to obtain palatable rewards or freely accessing less-desirable chow. At baseline, the reward was highly preferred and animals chose to lever-press for rewards while eating very little chow. However, disruption of DA signaling in the NA shifted their preference to chow, suggesting that loss of NA DA signaling reduced willingness to work for the palatable reward. By contrast, disruption of DA signaling in the dorsal striatum equally reduced both lever-pressing and chow 12 intake, indicating generalized reduction of feeding regardless of effort129-131. Similar results were observed using other behavioral paradigms to parse motivated vs. free intake132,133, substantiating the hypothesis that NA DA is specifically necessary for “wanting” or the approach phase of motivated behavior. By contrast, DA deficient mice with select restoration of DA to the dorsal but not ventral striatum recover their ability to feed ad libitum134,135, suggesting that dorsal striatal DA is sufficient for food intake that does not require effort. However, this type of freefeeding behavior is not necessarily analogous to “liking”. The hedonic responses to food can be assessed in animals through orofacial responses during eating136, and Berridge argues that “liking” does not require DA at all. Complete destruction of mesolimbic and nigrostriatal DA neurons does not reduce hedonic responses to sweet taste111. Furthermore, DA-deficient mice have normal preference for rewards and enhancing NA DA fails to increase “liking” of palatable food, but does increase “wanting”137,138. Instead, “liking” may be mediated by mu opioid or GABA-ergic inhibition of anatomically restricted “hedonic hotspots” in the NA shell and ventral pallidum109. DA neurons also play important roles in the initial and continued response to cues that may impact feeding. For example, DA neurons are activated in response to unexpected presentation of food, but this response quickly habituates112,139. When animals are conditioned to associate a cue with a food reward, DA firing will initially occur during consumption of the reward, however with learning, a shift occurs and DA neurons fire instead during the cue112, hence DA serves as a signal that predicts delivery of the reward. Rats presented with a novel palatable meal have increased NA DA release during consumption, however when presented with the same meal a second time, DA release is absent during consumption due to habituation. By contrast, if a novel palatable food is presented instead at the second meal, NA DA release increases during consumption, suggesting that DA neurons fire in response to the unexpected palatability of the second meal140. Although NA DA release may occur during feeding due to 13 novelty or unexpected presentation of food, as discussed above, lesion studies have shown that NA DA is not required for free feeding behavior. In line with the idea of reward prediction error, NA DA serves more as a “teaching signal” that codes how valuable a reward is, and therefore, how much work an animal should put forth to attain the reward113. While DA release occurs in response to palatable taste transmitted by gustatory pathways 141-143, additional work has demonstrated that taste-independent post-ingestive signals can also trigger mesolimbic DA release and food reinforcement. For example, mice genetically blind to sweet taste still develop preference for sucrose and display DA release in response to sucrose but not saccharin144. Given that saccharin is devoid of calories, this suggests that postingestive signals, likely released in response to caloric delivery to the gut, are sufficient to induce DA release independently of taste perception. This is supported by recent work showing that striatal DA release also occurs in proportion to increasing concentrations of intra-gastric infusion of lipid emulsions145, thus taste perception is not necessary for DA release in response to caloric content. In sum, these studies support the view that DA release occurs in response to both 1) sensory perception of taste, especially when novel or more palatable than expected, and 2) delivery of calories to the gut via post-ingestive mechanisms and independently of taste. Mesolimbic DA neurons and food “wanting” are also modulated with respect to peripheral energy status. For example, food deprivation increases DA neuron excitability, decreases DA re-uptake, and potentiates DA release during re-feeding62,140,146-149. Thus energy deficit appears to heighten the sensitivity of the mesolimbic DA system to food, and indeed, food-restricted animals will work harder to obtain food compared to sated controls150,151. Interestingly, fasted animals will also work harder to obtain sucrose but not for saccharin, indicating that post-ingestive mechanisms signal the caloric value of ingested rewards to VTA DA neurons151. However, after unlimited access to food and subsequent satiety, they diminish 14 their effort to obtain the palatable rewards151. Thus, food “wanting” is modulated by peripheral energy status (e.g. deficiency or satiety) presumably to coordinate food intake needed to ensure survival. There has been considerable interest in understanding if hormones conveying energy status, such as leptin or ghrelin, directly or indirectly engage VTA DA neurons to modify energy balance. Leptin is a satiety cue secreted from adipose tissue and serves as a long-term signal of peripheral fat stores40. Leptin action via the long form of the leptin receptor (LepRb) is also important for suppressing feeding via the hypothalamus, but some part of leptin action may be mediated via the VTA. Indeed, exogenous leptin treatment alters signaling in VTA DA neurons152,153, but only 6% of VTA DA express LepRb and they project to the amygdala rather than the NA154. Furthermore, while peripheral or ICV leptin suppresses food intake, sucrose self-administration, and conditioned place preference for palatable food150,155-157, the evidence for direct behavioral effects of leptin signaling via the VTA are limited. One study reported that intra-VTA leptin decreased food intake while silencing of VTA LepRb receptors increased it152, but subsequent work has not supported a direct role for VTA LepRb neurons in regulating feeding158,159. Instead, the anorectic effects of leptin may be induced via LHA LepRb neurons that project to the VTA and modulate DA signaling65,160. In contrast to leptin, ghrelin is a gastricderived hormone that increases in the circulation to signify energy deficit and strongly promotes food intake41. Ghrelin acts on the growth hormone secretagogue receptor (GHSR), which is also present in the hypothalamus and VTA161. Intra-VTA ghrelin increases firing of DA neurons through enhanced excitatory input and increases NA DA release161-163. In line with this, ghrelin infusion into the VTA increases both ad libitum food intake and self-administration of palatable food161,164,165 and blocking GHSRs in the VTA or DA signaling in the NA abolishes these effects161,162,166. Taken together, there is some evidence for direct modulation of VTA DA 15 neurons by leptin and ghrelin, but these hormones may also exert control via hypothalamic targets that project to the VTA. 1.3.3 Dopamine and Energy Expenditure The role of DA signaling in energy balance has been studied extensively in the context of feeding, yet energy expenditure represents an equally important determinant of body weight. It is well-established that pharmacologic enhancement of DA release also induces locomotor activity167,168 and in fact, psychostimulant-induced locomotor activity is often used as a behavioral surrogate to assess striatal DA signaling. For example, reducing DA re-uptake through pharmacologic or genetic disruption of DAT activity is associated with increased locomotion169,170. Similarly, agonist-mediated stimulation of D1R also increases locomotor activity, while D2 agonists increase autoinhibition that tends to suppress locomotor activity104. It is interesting that altering mesolimbic DA signaling has a predictable and well-established direct impact on locomotor activity, while the same manipulations rarely alter food intake unless animals are forced to work for food61,171. Thus, locomotor activity in general is tightly linked to DA signaling, but certain aspects of food intake may not be. Furthermore, it is difficult to interpret what locomotor activity means in rodents since it varies based on the testing procedure and encompasses aspects of novelty, exploratory, and emotional behaviors171. A standard way to examine volitional activity in rodents is to provide access to running wheels. Animals will lever-press for access to wheels and in some cases find it more reinforcing than sucrose172,173. Additionally, animals prefer to spend time in environmental contexts that they have been conditioned to associate with running wheels, even when the wheel is not present174. These data indicate that volitional wheel running is both rewarding and reinforcing, and therefore may engage mesolimbic DA circuits. Indeed, rodents bred for high levels of physical activity have increased levels of striatal DA and increased expression of D1R175,176. 16 Long-term access to wheels also increases TH expression and ∆FosB in the mesolimbic system174. Interestingly, food-deprived rats work harder for wheel access and run more when they are on the wheel, suggesting that physiologic state regulates the motivation to run177. Recently, Fernandes et al. showed that deletion of a critical mediator of leptin signaling specifically in the VTA caused increased volitional activity and preference for running wheels compared to controls158. Together, these studies support the hypothesis that the hormonal signature of energy depletion (i.e. lack of leptin signaling or food restriction) may directly act via VTA DA neurons to increase locomotor activity. While it seems counterintuitive that animals would increase energy expenditure in an energy-depleted state, one explanation is that ambulation and exploratory behavior is advantageous in a food-restricted state, as it could increase the probability of finding food. Consistent with this theory, Beeler et al. argues that DA is not a feeding signal per se, but instead coordinates energy expenditure in terms of how much energy should be allocated towards specific behaviors, including, but not limited to feeding171. DA neurons become hypersensitive during food deprivation146,147,178 in part due to changes in insulin, leptin, and ghrelin levels, thus engaging in reinforcing behaviors such as feeding or running in a fasted state is associated with more DA release and increased effort to attain the reward. This is supported by numerous lines of evidence demonstrating that fasted animals work harder for not only food or running but also for abusive drugs179-181. In sum, locomotor behavior is tightly tied to DA release, and may indicate a general increase in attention, arousal, and exploratory behavior that can increase effort for rewards in a context-dependent manner. 1.3.4 Neuropeptide Regulation of Dopamine Signaling Recent interest has been invested in understanding how neuropeptides modulate midbrain DA neurons, and hence how they direct behaviors relevant to energy balance. For example, pharmacological and electrophysiological studies suggest that VTA DA neurons express receptors for the neuropeptides, orexin, corticotropin-releasing factor, and neurotensin, 17 which all have the capacity to regulate DA neuronal function and motivated behaviors (reviewed in182-184). The specific importance of neuropeptides in directing neuronal signaling is just beginning to be appreciated (as recently reviewed by van den Pol185). Neuropeptides are often co-expressed along with the classical amino acid neurotransmitters glutamate or GABA, however the properties of neuropeptides vary greatly from classical transmitters. First, glutamate and GABA are released from the pre-synaptic terminal and travel mere nanometers across the synapse, where they act rapidly on ion channels to activate or inhibit the postsynaptic neuron. By contrast, neuropeptides released form presynaptic terminals can diffuse microns outside the synapse and thus can bind to both the post-synaptic neuron and neighboring neurons that do not make direct synaptic contact. While classical transmitters can gate ligand-gated ion channels to rapidly alter membrane potential, neuropeptides bind to Gprotein coupled receptors that change gene transcription or intracellular Ca++, and hence mediate long-term signaling changes in target neurons. Furthermore, in addition to release at the presynaptic terminal, some neuropeptides can be released from the dendrites or anywhere along the axon, which further increases the number of neurons that are influenced by neuropeptide release. Thus, although classical transmitters and neuropeptides may be released from the same neuron, they mediate very distinct and temporally dissociable effects on target cells. Furthermore, the distribution of different neuropeptide receptors on target neurons allows for further refinement of signal transduction and physiologic regulation, and these may be selectively targeted via pharmacological means. Understanding the role of neuropeptide and neuropeptide receptor systems that engage the DA system thus holds promise for selective, long-term control of DA-mediated behaviors that impact body weight. 18 1.4. Neurotensin Physiology 1.4.1 Discovery, Receptors, and Expression Patterns Neurotensin (Nts) is a 13 amino-acid neuropeptide that was first isolated from bovine hypothalamus by Carraway and Leeman in 1973. Upon discovery, Nts was first appreciated for its hypotensive effects as the authors noted “visible dilation” when it was applied to blood vessels in anesthetized rats186,187. In addition to regulation of blood pressure, Nts has since been implicated in a diverse array of physiologic processes including feeding, locomotor activity, body temperature, and gastric motility188. Nts is synthesized as part of a larger pro-peptide that also contains neuromedin-N and is cleaved into biologically active Nts by pro-hormone convertases in the Golgi apparatus and endoplasmic reticulum189. In neurons, the processing, storage, and release of Nts has not been well characterized, but similar to other neuropeptides, Nts has been localized in dense core vesicles at pre-synaptic terminals190. The discovery of Nts in the brain suggested it exerts central control of physiology, and indeed ICV administration of Nts is sufficient to alter food intake, locomotion, and body temperature87,191,192. Nts is also highly expressed within the gastrointestinal tract and adrenal gland, and release of Nts from these tissues presumably accounts for the high levels of Nts in the circulation. However, circulating Nts has an extremely short half-life due to rapid degradation by proteolytic enzymes193 . Thus, the central effects of Nts are likely due to Nts that is produced by, and released within the brain, but not from peripheral tissues. In the brain, studies using in situ hybridization and immunohistochemistry reveal that Nts is highly expressed within the LHA, NA, lateral septum, stria terminalis, preoptic area, central amygdala, periaqueductal grey and parabrachial nucleus188,194. 19 Nts binds two G-protein coupled receptors, Nts receptor 1 and Nts receptor 2 (NtsR1 and NtsR2), and Nts has a 10-30 fold higher affinity for NtsR1 than NtsR2195-198. NtsR1 is coupled to Gq, which leads to induction of phospholipase C and increased intracellular Ca++199,200. The downstream signaling transduction of NtsR2, however, is not well characterized; it may be Gq or Gi-coupled and could vary across species197,201. Nts also binds to a third non-G-protein-coupled receptor known as Nts receptor 3 (NtsR3) or sortilin, which consists of a single transmembrane protein but little else is known about its role in Nts signaling202,203. In situ hybridization and autoradiography reveals that NtsR1 is robustly expressed within the VTA, SN, septum and suprachiasmatic nucleus in adult animals. Interestingly, Ntsr1 expression is high and extensively distributed throughout the brain during early embryonic stages but is rapidly downregulated within the first few weeks of life, suggesting that Nts signaling via NtsR1 may be involved in the development of neural circuits204,205. By contrast, the expression of NtsR2 has been difficult to localize because most in situ hybridization studies show low levels of diffuse expression throughout the entire brain, however some have reported notably high levels in the cerebellum, periaqueductal gray, hippocampus, and VTA and SN206,207. Furthermore, NtsR2 does not exhibit a transient peak in expression during early development like NtsR1, and instead, is produced at low amounts during gestation and slowly increases to adult levels after birth. Several authors have suggested that NtsR2 is expressed on astrocytes in vitro208-210, however controversy remains as one report found that NtsR2 binding sites do not colocalize in vivo with the astrocyte marker, GFAP211. Clearly, the differences between NtsR1 and NtsR2 in terms of their affinity for Nts, expression patterns, and ontogeny suggest that they are functionally distinct, and indeed this is supported by work using antagonists or mice with developmental deletion of NtsR1 and NtsR2. In general, NtsR1 is implicated in mediating the feeding, locomotor, reward, and body temperature effects of Nts, while NtsR2 is implicated in Nts-induced analgesia203,212-218. The contributions of different Nts 20 receptors with respect to Nts modulation of DA signaling and energy balance will be discussed in detail in the following sections. 1.4.2 Neurotensin and Dopamine Signaling Nts regulates DA signaling and has accordingly generated considerable interest as a potential therapeutic target in diseases characterized by disrupted DA action, especially schizophrenia and Parkinson’s disease. Generally-speaking, Nts or Nts analogues enhance the activation of DA neurons through three different mechanisms: 1) increased intracellular Ca++219, 2) decreased D2R-mediated autoinhibition220-223, and 3) increased presynaptic excitatory input217,220,224. While several studies suggest that NtsR1 is required to activate DA neurons, roles for NtsR2 have also been proposed217,224, hence the precise mechanism by which Nts promotes activation of DA neurons remains unclear. Pharmacologic studies indicate a specific role for Nts action via the VTA, where it increases NA DA release and locomotor activity similar to psychostimulants92,214,225-232. Nts administration to the VTA also supports self-administration217,233,234, conditioned place preference (CPP)216,235, and sensitization236,237, indicating that Nts signaling in the VTA modifies DA-dependent behavior and is rewarding. Loss of NtsR1 signaling abolishes Nts-mediated selfadministration and CPP216,217, suggesting that NtsR1 confers the reinforcing properties of Nts activity in the VTA. In support of this, transgenic NtsR1Cre reporter mice indicate that NtsR1 is expressed almost exclusively by DA neurons within the VTA238. However, the receptor mechanism by which Nts acts in the VTA remains controversial, since studies using receptor antagonists or receptor-null mouse models have yielded conflicting data on the importance of NtsR1 and/or NtsR2214,228. Indeed, the VTA contains mRNA and binding sites for both receptors206,207,239,240, so both could play roles in modulating DA signaling and behavior. Reagents to permit the visualization of NtsR1 and NtsR2-expressing cells in the VTA would be 21 useful to determine if the receptor isoforms directly or indirectly modify VTA DA neurons, and hence to understand how Nts exerts control over DA signaling. While pharmacologic administration of Nts in the VTA strongly influences mesolimbic DA signaling and reward behaviors, the endogenous circuits providing Nts to the VTA, and how they modulate it, remain undetermined. Some Nts-containing neurons must provide Nts to the VTA, as demonstrated by the density of Nts immunoreactive terminals in the midbrain241-244 that are found in close apposition to VTA DA neurons238. Interestingly, Nts itself is present in the VTA and colocalizes with DA markers in rats244-246, but not in mouse or human247,248. Thus in mice and humans, Nts input to the VTA does not originate locally and must therefore come from elsewhere in the brain. Retrograde tracing in rats demonstrated the many Nts afferents to the VTA reside in the preoptic (POA) area, rostral LHA, and NA182,249, but the functional significance of these Nts inputs remains undefined. Curiously, destruction of the Nts neurons in the POA and LHA, which provided the majority of input to the VTA, did not substantially reduce Nts reactivity in the VTA, suggesting that Nts regulation of DA signaling is highly conserved250. Methods to selectively activate Nts neurons that project to the VTA are necessary to reveal how endogenous Nts circuits contribute to behavior. For example, chemogenetic activation of LHA Nts neurons increases NA DA release and locomotor activity93, similar to the effects of pharmacologic Nts in the VTA. While it will be important to assess the roles of each population of Nts neurons that project to the VTA, these data suggest that LHA Nts neurons mediate at least some aspects of Nts action via the VTA, and deserve further exploration. 1.4.3 Neurotensin and Energy Balance A first glimpse into the idea that Nts may be involved in energy balance came from work in the early 1980’s showing that central administration of Nts suppresses food intake in hungry rats251. The central anorectic effects of Nts have since been replicated by multiple independent 22 groups87,252-254, and peripheral Nts treatment has also been demonstrated to suppress appetite and reduce body weight 90,252,255,256. Nts administration directly to the VTA potently suppresses food intake, spurring interest in whether this might be associated with the known roles for Nts in regulating mesolimbic DA signaling86,88. Consistent with this, intra-VTA Nts also reduces operant responding for food88,257 and in most of the studies cited above, animals were either fasted or trained to eat food within a single short time period, both procedures that increase motivation to eat. Nts also suppresses feeding when administered to hypothalamic sites258, but not the NA87, indicating that Nts-induced anorexia occurs within specified anatomical circuits. Importantly, repeated peripheral injection of brain-penetrating Nts analogues or NtsR1 agonists suppresses food intake and promotes sustained weight loss in both lean and obese mice89,90, indicating that the anorectic effects of Nts are sufficient to modify body weight. Nts treatment does not, however, suppress feeding in mice lacking NtsR1212,253, suggesting that Nts mediates anorectic effects via NtsR1. This finding prompted investigation of whether loss of NtsR1 altered body weight, but independent reports suggest that NtsR1 knockout mice are either overweight212, not different213,218, or slightly lighter238 compared to controls. This variation could be explained by differences in genetic background and environment, similar to how some, but not all individuals with genetic predisposition to obesity become obese. Mice genetically lacking NtsR1 are also more susceptible to weight gain when given high fat diet, implicating NtsR1 signaling in palatable food intake238. Interestingly, mounting evidence suggests some overlap in the mechanisms by which the anorectic hormone leptin and central Nts suppress feeding. Leptin treatment increases Nts expression in the hypothalamus259,260, but disruption of NtsR1 signaling abolishes the ability of leptin to suppress food intake238,253,261. Taken together, these data support a hypothesis whereby leptin acts via LepRb-expressing neurons to increase Nts signaling via NtsR1, leading to reduced food intake. Leinninger et al. began to investigate this hypothesis by showing that 23 many LepRb-expressing neurons in the LHA also express Nts, and that the LHA is the only place in the brain where LepRb and Nts are co-expressed65,85. Deletion of LepRb selectively from LHA Nts neurons leads to increased body fat and impaired coordination of feeding behavior with peripheral energy cues66,85. These LHA Nts neurons project to the VTA, where their terminals are in close apposition with DA neurons85,238. Furthermore, NtsR1 is expressed almost exclusively on DA neurons in the VTA238, thus LHA Nts neurons can directly access mesolimbic DA circuits and are poised to regulate DA signaling via NtsR1. Consistent with this, chemogenetic activation of LHA Nts neurons increases DA release to the NA, as well as locomotor activity and energy expenditure that causes weight loss over 12 hours, but no alteration in food intake was observed93. This acute study, however, does not rule out a role for LHA Nts neurons in regulating feeding, nor does it indicate whether action via LHA Nts neurons can maintain sustained weight loss. These issues merit further investigation to understand the contribution of LHA Nts neurons to physiology and energy balance. 1.5 Neurotensin and Dopamine Signaling in Energy Balance Disorders 1.5.1 Is Dopamine Disrupted in Obesity? The hallmarks of obesity include overconsumption of calorically dense food and lack of physical activity, leading to weight gain. Since DA circuits regulate both feeding and volitional activity, many efforts have attempted to establish a causal link between dysfunctional DA signaling and obesity. One explanation for why obese individuals overeat is that they experience less reward in response to palatable foods and overeat to compensate; this is termed the “reward deficiency” hypothesis262. An alternative explanation is that obese individuals overeat because they have heightened reward responses to food cues and hence experience more pleasure during ingestion of highly palatable food. Thus, hypersensitive 24 reward circuits could also drive hyperphagia and weight gain, and this is referred to as the “reward-surfeit” hypothesis263,264. Evidence for both of these hypotheses will be discussed below. Several studies indicate that obese individuals have reduced D2R availability in the striatum265-268, suggesting rationale for the “reward deficiency hypothesis”. Indeed, humans with polymorphisms in the taq1 gene, which reduces D2R binding by 30-40%, are overrepresented in obese human populations compared to lean controls269-272. Furthermore, individuals with these taq1 variants are more likely to have reduced striatal response to palatable food that is predictive of future weight gain271,273. Similar results have been noted in animal obesity models. For example, rodents fed high-fat diets have reduced DA release in the NA, and blunted expression of DA-associated markers274-277. Johnson and Kenny showed that rats fed a palatable “cafeteria diet” consisting of “bacon, sausage, cheesecake, pound cake, frosting, and chocolate” have reduced striatal D2R binding and engage in compulsive-like behavior to attain the calorie-dense foods. Experimental depletion of striatal D2R receptors in lean rats mimicked compulsive feeding behavior, but did not alter ad libitum cafeteria diet consumption or weight gain, suggesting that lack of D2R signaling may regulate specific facets of feeding behavior but is not sufficient to cause obesity on a calorie-rich diet278. Taken together, these data suggest that striatal D2R signaling is blunted in obesity, which could be interpreted as hypoactivity of motivational and reward circuits, lending support to the reward deficiency hypothesis. Despite a seemingly strong link between low D2R and obesity, the idea that reduced D2R signaling is a cause rather than consequence of obesity has been called into question. Conflicting with previous work, several more recent studies have found no link between obese populations and D2R binding279-283. Furthermore, whole-body or striatal knockdown of D2Rs decreases locomotor activity but does not predispose animals to weight gain on HF diet278,284,285, suggesting that reduced D2R signaling may cause lower activity levels in obesity, but is not 25 sufficient to drive overconsumption on palatable diet. This casts doubt on the reward deficiency hypothesis and suggests that reduced striatal D2R may be a consequence rather than a cause of weight gain. An alternate explanation is the reward-surfeit hypothesis, which argues that individuals with hyper-responsive reward circuits overconsume food that leads to weight gain and obesity263,264. Indeed, obese individuals show a greater response to anticipation and receipt of food in brain areas associated with sensory and hedonic aspects of food286 and they perceive palatable foods as more pleasant than lean controls287,288. Obese individuals also show greater striatal activation in response to visual food cues than normal weight subjects289,290. Interestingly, adolescents at risk for obesity show greater striatal activation during receipt of palatable food but not during anticipation, suggesting that initial vulnerability to obesity may explained by hyper-responsive reward circuits during food intake264. Thus, via the reward-surfeit hypothesis, increased hedonic and sensory response to palatable foods may initially fuel overconsumption, and repeated intake of palatable foods over time leads to conditioned associations with cues that predict food, which over-activate striatal regions to drive further overconsumption after the onset of obesity264. However, none of these neuroimaging studies have implicated DA dysfunction in hyperactive responses to food and food cues in obesity. In fact, increased hedonic response to food is likely mediated via non-DA-ergic mechanisms such as mu opiod receptor signaling or other neurotransmitter systems in both animals and humans280,281,291. Although evidence exists for both the reward deficit and reward surfeit hypotheses, like many human diseases, obesity may be a heterogeneous disease that can be caused by either hyperactivity or hypoactivity of the reward system, which could vary with genetics and environment. Consistent with this, humans with genetic reduction in D2R availability had increased risk of future weight gain if they showed reduced striatal response to food cues, whereas humans with genetically normal D2R signaling had increased risk of future 26 weight gain if they showed increased striatal response to food cues273. Thus, obesity cannot be defined by one mechanism but clearly involves disruption in reward circuitry that encompasses both DA-mediated “wanting” and DA-independent “liking.” A note on food addiction: Many behavioral and biological parallels have been drawn between obesity and substance abuse, leading some to ask whether overconsumption of food in obese individuals is caused by “food addiction.” For example, similar to obesity, taq1 allele polymorphisms are associated with substance abuse disorders and drug addicted subjects show reduced striatal D2R binding292-295. Although the neural circuitry disrupted in substance use disorders and obesity may overlap to some degree, Stice et al. caution against using the term food addiction296. As the authors point out, the substance use disorder often has elements of both dependence (i.e. tolerance and withdrawal) and abuse (use of substance despite negative consequence) and they argue it is difficult to apply these criteria to palatable food. In particular, few studies have shown that animals can develop tolerance and withdrawal symptoms to sugar297,298. Furthermore, few people who try addictive substances go on to develop substance use disorder, yet a much higher proportion of individuals who eat palatable foods gain weight and become obese. In the words of Stice et al., “very few humans are driven to violent criminal behavior due to craving for chocolate”296. While this may be true, it is important to remember that unlike most addictive drugs, possession and consumption of palatable food is not against the law, thus the immediate negative consequences for obtaining high calorie foods are comparatively minimal. In fact, consumption of calorically-rich food is socially acceptable and often expected (i.e. cake and ice cream at birthday parties), plus food intake in general is necessary for survival. Thus, it is difficult to directly compare food to addictive drugs as “substances” that have abusive potential. In sum, although obesity and substance abuse may share some common features, care should be taken in considering 27 whether obese individuals are “food addicts” especially in the context of how medical professionals define addiction. 1.5.2 Evidence for Neurotensin in Obesity As described previously, Nts suppresses food intake when administered peripherally or centrally, including directly to the VTA, and this is likely mediated by NtsR1. Nts also increases locomotor activity through NtsR1 when delivered to the VTA, but not centrally or peripherally92,191,299. Chronic Nts or NtsR1 agonists may therefore have therapeutic potential to induce weight loss if delivered to the VTA, but this has not been examined. The physiological role of Nts in energy balance also remains unclear. To date, a handful of studies have shown reduced levels of Nts protein or mRNA in the hypothalamus of genetically obese mice and rats300-303. Given the anorectic effects of pharmacologic Nts, reduced levels in obesity might lead to increased food intake, but there is insufficient data to evaluate whether decreased hypothalamic Nts content is a cause or effect of long-term hyperphagia. Peripheral injection of Nts analogs or NtsR1 agonists are sufficient to suppress food intake and either induce weight loss or prevent weight gain in lean rodents89,90. Interestingly, these compounds also restrain feeding and reduce body weight in genetically obese rodents, indicating that Nts may have translational potential as an anti-obesity agent89,90. In both of these studies however, the obese animal models were genetically deficient in leptin signaling; given the overlap between leptin and Nts action, inducing Nts signaling might have simply rescued their disrupted leptin-induced Nts signaling to potentiate weight loss. The anorectic potential of Nts in diet-induced obesity, which is the common cause of human obesity, has yet to be examined. Peripheral Nts or intraVTA Nts suppresses operant responding for sucrose257,304, suggesting that Nts may indeed be capable of restraining intake of palatable, obesogenic foods. Furthermore, mice genetically lacking NtsR1display increased sucrose preference and susceptibility to weight gain on palatable diet but not chow, implicating the requirement of NtsR1 for restraint of hedonic 28 intake238. Collectively these data suggest potential for Nts signaling as a novel target system for weight control, but much additional work is necessary to determine the precise circuits by which Nts acts to modify feeding and energy expenditure. In 2016, Nts received considerable attention in the context of obesity when Evers et al. published in Nature that Nts knockout mice were protected from obesity on high fat diet. This report focused on peripheral Nts, suggesting that Nts produced within the gastrointestinal tract is essential to facilitate intestinal fat absorption. As a result, loss of Nts-mediated fat absorption led to protection from diet-induced obesity. The authors also showed that circulating Nts levels were increased in the blood of obese individuals, and that elevated Nts in lean individuals predicted future weight gain305. While this study implicates an important role for peripheral Nts in energy accumulation, it also underscores the importance for site-specific investigation of Nts action to discern its contributions to physiology. The potentially differing roles of Nts in the brain (to suppress feeding) compared to the gut (to promote fat absorption) must also be taken into account in the development of future pharmaceuticals that target the Nts system. Given the very short half-life of circulating Nts, and the unlikelihood of the neuropeptide to traverse the blood brain barrier, the central effects of Nts are unlikely to be mediated by peripherallygenerated Nts. As such, Nts signaling in the gut is outside the scope of this dissertation, which instead focuses on elucidating the mechanisms by which central Nts engages the DA system to modify behavior and body weight. 1.6 Goals of the Dissertation Pharmacologic Nts modifies VTA DA signaling, increases physical activity, and restrains food intake, but the endogenous circuits by which Nts mediates these effects to regulate energy balance remain unclear. Previously, understanding of the physiological roles of Nts in engaging DA signaling was limited by the lack of reagents to visualize and manipulate specific populations 29 of Nts-expressing and Nts-regulated cells. To overcome this obstacle, we used existing mice that express Cre-recombinase in Nts neurons, and we developed mice that express Cre in NtsR1 or NtsR2 cells; these models allow us to reveal and interrogate specific Nts circuits using Cre-inducible molecular tools. Better understanding of the precise mechanisms by which Nts acts to coordinate food intake and energy expenditure will lead to novel therapeutic targets for the prevention and treatment of obesity. As such, the major goals of this dissertation are: 1. Define the Endogenous Nts Inputs to the VTA (Chapter 2) Hypothesis: The VTA receives afferent input from select anatomically-defined Nts populations that regulate reward rather than aversion. Method: We utilized retrograde tracing in mice expressing GFP selectively in Nts neurons to identify all Nts neurons that project to the VTA, and hence revealed which neural populations are anatomically positioned to mediate the weight-reducing effects of Nts within the VTA. 2. Decipher the Neurotransmitter Content, Ontogeny, and Projections of VTA NtsR1 and NtsR2 Neurons (Chapter 3) Hypothesis: NtsR1 is the dominant neuronal isoform by which Nts regulates DA signaling. Method: We utilized a dual recombinase strategy to reveal the VTA cells that express NtsR1 and NtsR2 during development and in the adult brain, and their projection sites. This study reveals that NtsR1 and NtsR2 are expressed by different VTA cell types, and implicate NtsR1 as the predominant receptor isoform on VTA DA neurons. 3. Understand How Loss of Nts Input to the VTA Modifies Energy Balance (Chapter 4) Hypothesis: NtsR1 signaling in the VTA is critical for coordination of behaviors that impact body weight. 30 Method: Genetic ablation of NtsR1 neurons in the VTA revealed the requirement for Nts signaling via the DA-ergic VTA NtsR1 neurons, and hence for control of DA-mediated behaviors that coordinate body weight. 4. Determine How LHA Nts Neurons Regulate Feeding and Body Weight (Chapter 5) Hypothesis: LHA Nts neurons regulate energy balance by engaging the mesolimbic DA system via NtsR1. Method: DREADD technology was used to selectively activate LHA Nts neurons in mice that have intact and disrupted NtsR1, permitting examination of how LHA NtsNtsR1 signaling contributes to feeding behavior and energy expenditure in lean and obese mice. Through these goals (Fig. 1), we establish a central circuit whereby LHA Nts neurons act via VTA NtsR1 neurons to coordinate feeding and volitional activity that modify body weight. These data establish a novel, physiologic mechanism to explain the anorectic and locomotor effects induced by pharmacologic Nts, and the translational implications of these findings for treating overweight and obesity will be discussed in Chapter 6. 31 Figure 1. Schematic of hypothesized neural circuit by which Nts acts to modify mesolimbic DA signaling and body weight. A) Hypothesis for dissertation: Nts regulates mesolimbic DA signaling via Nts receptors, and input from the LHA is critical for anorectic effects of Nts. B) Key questions addressed in each chapter of the dissertation to investigate how Nts regulates DA signaling energy balance. 32 REFERENCES 33 REFERENCES 1 Ng, M. et al. Global, regional, and national prevalence of overweight and obesity in children and adults during 1980-2013: a systematic analysis for the Global Burden of Disease Study 2013. Lancet 384, 766-781, doi:10.1016/s0140-6736(14)60460-8 (2014). 2 World Health Organization. Obesity and Overweight Fact Sheet, (2016). 3 Flegal, K. M., Carroll, M. D., Kit, B. K. & Ogden, C. L. Prevalence of obesity and trends in the distribution of body mass index among US adults, 1999-2010. JAMA : the journal of the American Medical Association 307, 491-497, doi:10.1001/jama.2012.39 (2012). 4 Fryar CD, C. M., Ogden CL. Prevalence of overweight and obesity among children and adolescents: United States, 1963-1965 through 2011-2012. (National Center for Health Statistics, Atlanta, GA, 2014). 5 Ogden, C. L. et al. Trends in Obesity Prevalence Among Children and Adolescents in the United States, 1988-1994 Through 2013-2014. JAMA : the journal of the American Medical Association 315, 2292-2299, doi:10.1001/jama.2016.6361 (2016). 6 Matt T, E. J., Bedell J, Selenic D, Young C, Deitcher D. . Obesity in the South Bronx: A look across generations. (New York City Department of Health and Metnal Hygeine, New York, NY, 2007). 7 Whitaker, R. C., Wright, J. A., Pepe, M. S., Seidel, K. D. & Dietz, W. H. Predicting obesity in young adulthood from childhood and parental obesity. The New England journal of medicine 337, 869-873, doi:10.1056/nejm199709253371301 (1997). 8 Lim, S. S. et al. A comparative risk assessment of burden of disease and injury attributable to 67 risk factors and risk factor clusters in 21 regions, 1990-2010: a systematic analysis for the Global Burden of Disease Study 2010. Lancet 380, 22242260, doi:10.1016/s0140-6736(12)61766-8 (2012). 9 Malnick, S. D. & Knobler, H. The medical complications of obesity. QJM : monthly journal of the Association of Physicians 99, 565-579, doi:10.1093/qjmed/hcl085 (2006). 10 Kushner, R. F. & Foster, G. D. Obesity and quality of life. Nutrition (Burbank, Los Angeles County, Calif.) 16, 947-952 (2000). 11 James, W. P. WHO recognition of the global obesity epidemic. International journal of obesity (2005) 32 Suppl 7, S120-126, doi:10.1038/ijo.2008.247 (2008). 12 Marketdata Enterprises Inc. The U.S. Weight Loss Market: 2015 Status Report and Forcast. (2015). 34 13 Mann, T. et al. Medicare's search for effective obesity treatments: diets are not the answer. The American psychologist 62, 220-233, doi:10.1037/0003-066x.62.3.220 (2007). 14 French, S. A. et al. Weight variability and incident disease in older women: the Iowa Women's Health Study. International journal of obesity and related metabolic disorders : journal of the International Association for the Study of Obesity 21, 217-223 (1997). 15 Bray, G. A. & Ryan, D. H. Update on obesity pharmacotherapy. Annals of the New York Academy of Sciences 1311, 1-13, doi:10.1111/nyas.12328 (2014). 16 Khera, R. et al. Association of Pharmacological Treatments for Obesity With Weight Loss and Adverse Events: A Systematic Review and Meta-analysis. JAMA : the journal of the American Medical Association 315, 2424-2434, doi:10.1001/jama.2016.7602 (2016). 17 Yanovski, S. Z. & Yanovski, J. A. Long-term drug treatment for obesity: a systematic and clinical review. JAMA : the journal of the American Medical Association 311, 74-86, doi:10.1001/jama.2013.281361 (2014). 18 Cheng, J., Gao, J., Shuai, X., Wang, G. & Tao, K. The comprehensive summary of surgical versus non-surgical treatment for obesity: a systematic review and metaanalysis of randomized controlled trials. Oncotarget 7, 39216-39230, doi:10.18632/oncotarget.9581 (2016). 19 Colquitt, J. L., Pickett, K., Loveman, E. & Frampton, G. K. Surgery for weight loss in adults. The Cochrane database of systematic reviews, CD003641, doi:10.1002/14651858.CD003641.pub4 (2014). 20 Janik, M. R., Rogula, T., Bielecka, I., Kwiatkowski, A. & Pasnik, K. Quality of Life and Bariatric Surgery: Cross-Sectional Study and Analysis of Factors Influencing Outcome. Obesity surgery 26, 2849-2855, doi:10.1007/s11695-016-2220-2 (2016). 21 Pories, W. J. Bariatric surgery: risks and rewards. The Journal of clinical endocrinology and metabolism 93, S89-96, doi:10.1210/jc.2008-1641 (2008). 22 National Institute of Diabetes and Digetsive and Kidney Diseases. Definition and Facts for Bariatric Surgery, (2016). 23 Ford, E. S. & Dietz, W. H. Trends in energy intake among adults in the United States: findings from NHANES. The American journal of clinical nutrition 97, 848-853, doi:10.3945/ajcn.112.052662 (2013). 24 Church, T. S. et al. Trends over 5 decades in U.S. occupation-related physical activity and their associations with obesity. PloS one 6, e19657, doi:10.1371/journal.pone.0019657 (2011). 25 Ward BW, C. T., Nugent CN, Schiller JS. Early release of selected estimates based on data from the 2015 National Health Interview Survey., (U.S. Department of Health and Human Services, 2016). 35 26 Cordain, L. et al. Origins and evolution of the Western diet: health implications for the 21st century. The American journal of clinical nutrition 81, 341-354 (2005). 27 Cota, D., Proulx, K. & Seeley, R. J. The role of CNS fuel sensing in energy and glucose regulation. Gastroenterology 132, 2158-2168, doi:10.1053/j.gastro.2007.03.049 (2007). 28 Muller, M. J., Bosy-Westphal, A. & Heymsfield, S. B. Is there evidence for a set point that regulates human body weight? F1000 medicine reports 2, 59, doi:10.3410/m2-59 (2010). 29 Poehlman, E. T. A review: exercise and its influence on resting energy metabolism in man. Medicine and science in sports and exercise 21, 515-525 (1989). 30 Jebb, S. A. et al. Changes in macronutrient balance during over- and underfeeding assessed by 12-d continuous whole-body calorimetry. The American journal of clinical nutrition 64, 259-266 (1996). 31 Diaz, E. O., Prentice, A. M., Goldberg, G. R., Murgatroyd, P. R. & Coward, W. A. Metabolic response to experimental overfeeding in lean and overweight healthy volunteers. The American journal of clinical nutrition 56, 641-655 (1992). 32 Taylor, H. L. & Keys, A. Adaptation to caloric restriction. Science (New York, N.Y.) 112, 215-218 (1950). 33 Leibel, R. L., Rosenbaum, M. & Hirsch, J. Changes in energy expenditure resulting from altered body weight. The New England journal of medicine 332, 621-628, doi:10.1056/nejm199503093321001 (1995). 34 Redman, L. M. et al. Metabolic and behavioral compensations in response to caloric restriction: implications for the maintenance of weight loss. PloS one 4, e4377, doi:10.1371/journal.pone.0004377 (2009). 35 Larder, R. & O'Rahilly, S. Shedding pounds after going under the knife: guts over glorywhy diets fail. Nature medicine 18, 666-667, doi:10.1038/nm.2747 (2012). 36 Sumithran, P. et al. Long-term persistence of hormonal adaptations to weight loss. The New England journal of medicine 365, 1597-1604, doi:10.1056/NEJMoa1105816 (2011). 37 Sumithran, P. & Proietto, J. The defence of body weight: a physiological basis for weight regain after weight loss. Clinical science (London, England : 1979) 124, 231-241, doi:10.1042/cs20120223 (2013). 38 Fothergill, E. et al. Persistent metabolic adaptation 6 years after "The Biggest Loser" competition. Obesity (Silver Spring, Md.) 24, 1612-1619, doi:10.1002/oby.21538 (2016). 39 Locke, A. E. et al. Genetic studies of body mass index yield new insights for obesity biology. Nature 518, 197-206, doi:10.1038/nature14177 (2015). 40 Friedman, J. M. & Halaas, J. L. Leptin and the regulation of body weight in mammals. Nature 395, 763-770, doi:10.1038/27376 (1998). 36 41 Muller, T. D. et al. Ghrelin. Molecular metabolism 4, 437-460, doi:10.1016/j.molmet.2015.03.005 (2015). 42 Richard, D. Cognitive and autonomic determinants of energy homeostasis in obesity. Nature reviews. Endocrinology 11, 489-501, doi:10.1038/nrendo.2015.103 (2015). 43 Krashes, M. J. et al. Rapid, reversible activation of AgRP neurons drives feeding behavior in mice. The Journal of clinical investigation 121, 1424-1428, doi:10.1172/jci46229 (2011). 44 Krashes, M. J., Shah, B. P., Koda, S. & Lowell, B. B. Rapid versus delayed stimulation of feeding by the endogenously released AgRP neuron mediators GABA, NPY, and AgRP. Cell metabolism 18, 588-595, doi:10.1016/j.cmet.2013.09.009 (2013). 45 Webber, E. S., Bonci, A. & Krashes, M. J. The elegance of energy balance: Insight from circuit-level manipulations. Synapse (New York, N.Y.) 69, 461-474, doi:10.1002/syn.21837 (2015). 46 Betley, J. N. et al. Neurons for hunger and thirst transmit a negative-valence teaching signal. Nature 521, 180-185, doi:10.1038/nature14416 (2015). 47 Chen, Y., Lin, Y. C., Kuo, T. W. & Knight, Z. A. Sensory detection of food rapidly modulates arcuate feeding circuits. Cell 160, 829-841, doi:10.1016/j.cell.2015.01.033 (2015). 48 Morrison, C. D., Morton, G. J., Niswender, K. D., Gelling, R. W. & Schwartz, M. W. Leptin inhibits hypothalamic Npy and Agrp gene expression via a mechanism that requires phosphatidylinositol 3-OH-kinase signaling. American journal of physiology. Endocrinology and metabolism 289, E1051-1057, doi:10.1152/ajpendo.00094.2005 (2005). 49 Zhan, C. et al. Acute and long-term suppression of feeding behavior by POMC neurons in the brainstem and hypothalamus, respectively. The Journal of neuroscience : the official journal of the Society for Neuroscience 33, 3624-3632, doi:10.1523/jneurosci.2742-12.2013 (2013). 50 Cowley, M. A. et al. Leptin activates anorexigenic POMC neurons through a neural network in the arcuate nucleus. Nature 411, 480-484, doi:10.1038/35078085 (2001). 51 Elias, C. F. et al. Leptin differentially regulates NPY and POMC neurons projecting to the lateral hypothalamic area. Neuron 23, 775-786 (1999). 52 You, P. et al. Effects of Melanocortin 3 and 4 Receptor Deficiency on Energy Homeostasis in Rats. Scientific reports 6, 34938, doi:10.1038/srep34938 (2016). 53 Huszar, D. et al. Targeted disruption of the melanocortin-4 receptor results in obesity in mice. Cell 88, 131-141 (1997). 54 Vaisse, C., Clement, K., Guy-Grand, B. & Froguel, P. A frameshift mutation in human MC4R is associated with a dominant form of obesity. Nature genetics 20, 113-114, doi:10.1038/2407 (1998). 37 55 Farooqi, I. S. et al. Clinical spectrum of obesity and mutations in the melanocortin 4 receptor gene. The New England journal of medicine 348, 1085-1095, doi:10.1056/NEJMoa022050 (2003). 56 Liu, H. et al. Transgenic mice expressing green fluorescent protein under the control of the melanocortin-4 receptor promoter. The Journal of neuroscience : the official journal of the Society for Neuroscience 23, 7143-7154 (2003). 57 Delgado, J. M. & Anand, B. K. Increase of food intake induced by electrical stimulation of the lateral hypothalamus. The American journal of physiology 172, 162-168 (1953). 58 Anand, B. K. & Brobeck, J. R. Localization of a "feeding center" in the hypothalamus of the rat. Proceedings of the Society for Experimental Biology and Medicine. Society for Experimental Biology and Medicine (New York, N.Y.) 77, 323-324 (1951). 59 Morrison, S. D., Barrnett, R. J. & Mayer, J. Localization of lesions in the lateral hypothalamus of rats with induced adipsia and aphagia. The American journal of physiology 193, 230-234 (1958). 60 Grossman, S. P., Dacey, D., Halaris, A. E., Collier, T. & Routtenberg, A. Aphagia and adipsia after preferential destruction of nerve cell bodies in hypothalamus. Science (New York, N.Y.) 202, 537-539 (1978). 61 Salamone, J. D. & Correa, M. The mysterious motivational functions of mesolimbic dopamine. Neuron 76, 470-485, doi:10.1016/j.neuron.2012.10.021 (2012). 62 Hernandez, L. & Hoebel, B. G. Feeding and hypothalamic stimulation increase dopamine turnover in the accumbens. Physiology & behavior 44, 599-606 (1988). 63 Olds, J. & Milner, P. Positive reinforcement produced by electrical stimulation of septal area and other regions of rat brain. Journal of comparative and physiological psychology 47, 419-427 (1954). 64 Saper, C. B., Swanson, L. W. & Cowan, W. M. An autoradiographic study of the efferent connections of the lateral hypothalamic area in the rat. The Journal of comparative neurology 183, 689-706, doi:10.1002/cne.901830402 (1979). 65 Leinninger, G. M. et al. Leptin acts via leptin receptor-expressing lateral hypothalamic neurons to modulate the mesolimbic dopamine system and suppress feeding. Cell metabolism 10, 89-98, doi:10.1016/j.cmet.2009.06.011 (2009). 66 Brown, J. A. et al. Loss of Action via Neurotensin-Leptin Receptor Neurons Disrupts Leptin and Ghrelin-Mediated Control of Energy Balance. Endocrinology, doi:10.1210/en.2017-00122 (2017). 67 Brown, J. A., Woodworth, H. L. & Leinninger, G. M. To ingest or rest? Specialized roles of lateral hypothalamic area neurons in coordinating energy balance. Frontiers in systems neuroscience 9, 9, doi:10.3389/fnsys.2015.00009 (2015). 68 Stuber, G. D. & Wise, R. A. Lateral hypothalamic circuits for feeding and reward. Nature neuroscience 19, 198-205, doi:10.1038/nn.4220 (2016). 38 69 de Lecea, L. et al. The hypocretins: hypothalamus-specific peptides with neuroexcitatory activity. Proceedings of the National Academy of Sciences of the United States of America 95, 322-327 (1998). 70 Yamanaka, A. et al. Hypothalamic orexin neurons regulate arousal according to energy balance in mice. Neuron 38, 701-713 (2003). 71 Inutsuka, A. et al. Concurrent and robust regulation of feeding behaviors and metabolism by orexin neurons. Neuropharmacology 85, 451-460, doi:10.1016/j.neuropharm.2014.06.015 (2014). 72 Cai, X. J. et al. Hypoglycemia activates orexin neurons and selectively increases hypothalamic orexin-B levels: responses inhibited by feeding and possibly mediated by the nucleus of the solitary tract. Diabetes 50, 105-112 (2001). 73 Moriguchi, T., Sakurai, T., Nambu, T., Yanagisawa, M. & Goto, K. Neurons containing orexin in the lateral hypothalamic area of the adult rat brain are activated by insulininduced acute hypoglycemia. Neuroscience letters 264, 101-104 (1999). 74 Marcus, J. N. et al. Differential expression of orexin receptors 1 and 2 in the rat brain. The Journal of comparative neurology 435, 6-25 (2001). 75 Korotkova, T. M., Sergeeva, O. A., Eriksson, K. S., Haas, H. L. & Brown, R. E. Excitation of ventral tegmental area dopaminergic and nondopaminergic neurons by orexins/hypocretins. The Journal of neuroscience : the official journal of the Society for Neuroscience 23, 7-11 (2003). 76 Rosin, D. L., Weston, M. C., Sevigny, C. P., Stornetta, R. L. & Guyenet, P. G. Hypothalamic orexin (hypocretin) neurons express vesicular glutamate transporters VGLUT1 or VGLUT2. The Journal of comparative neurology 465, 593-603, doi:10.1002/cne.10860 (2003). 77 Ludwig, D. S. et al. Melanin-concentrating hormone overexpression in transgenic mice leads to obesity and insulin resistance. The Journal of clinical investigation 107, 379386, doi:10.1172/jci10660 (2001). 78 Qu, D. et al. A role for melanin-concentrating hormone in the central regulation of feeding behaviour. Nature 380, 243-247, doi:10.1038/380243a0 (1996). 79 Jego, S. et al. Optogenetic identification of a rapid eye movement sleep modulatory circuit in the hypothalamus. Nature neuroscience 16, 1637-1643, doi:10.1038/nn.3522 (2013). 80 Harthoorn, L. F., Sane, A., Nethe, M. & Van Heerikhuize, J. J. Multi-transcriptional profiling of melanin-concentrating hormone and orexin-containing neurons. Cellular and molecular neurobiology 25, 1209-1223, doi:10.1007/s10571-005-8184-8 (2005). 81 Elias, C. F. et al. Characterization of CART neurons in the rat and human hypothalamus. The Journal of comparative neurology 432, 1-19 (2001). 39 82 Chee, M. J., Pissios, P. & Maratos-Flier, E. Neurochemical characterization of neurons expressing melanin-concentrating hormone receptor 1 in the mouse hypothalamus. The Journal of comparative neurology 521, 2208-2234, doi:10.1002/cne.23273 (2013). 83 Pissios, P. et al. Dysregulation of the mesolimbic dopamine system and reward in MCH/- mice. Biological psychiatry 64, 184-191, doi:10.1016/j.biopsych.2007.12.011 (2008). 84 Sears, R. M. et al. Regulation of nucleus accumbens activity by the hypothalamic neuropeptide melanin-concentrating hormone. The Journal of neuroscience : the official journal of the Society for Neuroscience 30, 8263-8273, doi:10.1523/jneurosci.585809.2010 (2010). 85 Leinninger, G. M. et al. Leptin action via neurotensin neurons controls orexin, the mesolimbic dopamine system and energy balance. Cell metabolism 14, 313-323, doi:10.1016/j.cmet.2011.06.016 (2011). 86 Cador, M., Kelley, A. E., Le Moal, M. & Stinus, L. Ventral tegmental area infusion of substance P, neurotensin and enkephalin: differential effects on feeding behavior. Neuroscience 18, 659-669 (1986). 87 Hawkins, M. F. Central nervous system neurotensin and feeding. Physiology & behavior 36, 1-8 (1986). 88 Hawkins, M. F. Aphagia in the rat following microinjection of neurotensin into the ventral tegmental area. Life sciences 38, 2383-2388 (1986). 89 Feifel, D., Goldenberg, J., Melendez, G. & Shilling, P. D. The acute and subchronic effects of a brain-penetrating, neurotensin-1 receptor agonist on feeding, body weight and temperature. Neuropharmacology 58, 195-198, doi:10.1016/j.neuropharm.2009.07.001 (2010). 90 Boules, M. et al. A novel neurotensin peptide analog given extracranially decreases food intake and weight in rodents. Brain research 865, 35-44 (2000). 91 Goforth, P. B., Leinninger, G. M., Patterson, C. M., Satin, L. S. & Myers, M. G., Jr. Leptin acts via lateral hypothalamic area neurotensin neurons to inhibit orexin neurons by multiple GABA-independent mechanisms. The Journal of neuroscience : the official journal of the Society for Neuroscience 34, 11405-11415, doi:10.1523/jneurosci.516713.2014 (2014). 92 Kalivas, P. W., Burgess, S. K., Nemeroff, C. B. & Prange, A. J., Jr. Behavioral and neurochemical effects of neurotensin microinjection into the ventral tegmental area of the rat. Neuroscience 8, 495-505 (1983). 93 Patterson, C. M. et al. Ventral tegmental area neurotensin signaling links the lateral hypothalamus to locomotor activity and striatal dopamine efflux in male mice. Endocrinology 156, 1692-1700, doi:10.1210/en.2014-1986 (2015). 94 Zhou, Q. Y. & Palmiter, R. D. Dopamine-deficient mice are severely hypoactive, adipsic, and aphagic. Cell 83, 1197-1209 (1995). 40 95 Elsworth, J. D. & Roth, R. H. Dopamine synthesis, uptake, metabolism, and receptors: relevance to gene therapy of Parkinson's disease. Experimental neurology 144, 4-9, doi:10.1006/exnr.1996.6379 (1997). 96 Kebabian, J. W. & Greengard, P. Dopamine-sensitive adenyl cyclase: possible role in synaptic transmission. Science (New York, N.Y.) 174, 1346-1349 (1971). 97 Enjalbert, A. & Bockaert, J. Pharmacological characterization of the D2 dopamine receptor negatively coupled with adenylate cyclase in rat anterior pituitary. Molecular pharmacology 23, 576-584 (1983). 98 Missale, C., Nash, S. R., Robinson, S. W., Jaber, M. & Caron, M. G. Dopamine receptors: from structure to function. Physiological reviews 78, 189-225 (1998). 99 Greengard, P. The neurobiology of dopamine signaling. Bioscience reports 21, 247-269 (2001). 100 Yao, W. D., Spealman, R. D. & Zhang, J. Dopaminergic signaling in dendritic spines. Biochemical pharmacology 75, 2055-2069, doi:10.1016/j.bcp.2008.01.018 (2008). 101 Onali, P., Olianas, M. C. & Bunse, B. Evidence that adenosine A2 and dopamine autoreceptors antagonistically regulate tyrosine hydroxylase activity in rat striatal synaptosomes. Brain research 456, 302-309 (1988). 102 Pothos, E. N., Przedborski, S., Davila, V., Schmitz, Y. & Sulzer, D. D2-Like dopamine autoreceptor activation reduces quantal size in PC12 cells. The Journal of neuroscience : the official journal of the Society for Neuroscience 18, 5575-5585 (1998). 103 Cass, W. A. & Zahniser, N. R. Potassium channel blockers inhibit D2 dopamine, but not A1 adenosine, receptor-mediated inhibition of striatal dopamine release. Journal of neurochemistry 57, 147-152 (1991). 104 Baik, J. H. Dopamine signaling in reward-related behaviors. Frontiers in neural circuits 7, 152, doi:10.3389/fncir.2013.00152 (2013). 105 Beaulieu, J. M. & Gainetdinov, R. R. The physiology, signaling, and pharmacology of dopamine receptors. Pharmacological reviews 63, 182-217, doi:10.1124/pr.110.002642 (2011). 106 Roeper, J. Dissecting the diversity of midbrain dopamine neurons. Trends in neurosciences 36, 336-342, doi:10.1016/j.tins.2013.03.003 (2013). 107 Salamone, J. D., Correa, M., Mingote, S. M. & Weber, S. M. Beyond the reward hypothesis: alternative functions of nucleus accumbens dopamine. Current opinion in pharmacology 5, 34-41, doi:10.1016/j.coph.2004.09.004 (2005). 108 Yin, H. H., Ostlund, S. B. & Balleine, B. W. Reward-guided learning beyond dopamine in the nucleus accumbens: the integrative functions of cortico-basal ganglia networks. The European journal of neuroscience 28, 1437-1448, doi:10.1111/j.14609568.2008.06422.x (2008). 41 109 Berridge, K. C. & Kringelbach, M. L. Pleasure systems in the brain. Neuron 86, 646-664, doi:10.1016/j.neuron.2015.02.018 (2015). 110 Berridge, K. C. & Robinson, T. E. What is the role of dopamine in reward: hedonic impact, reward learning, or incentive salience? Brain research. Brain research reviews 28, 309-369 (1998). 111 Berridge, K. C., Venier, I. L. & Robinson, T. E. Taste reactivity analysis of 6hydroxydopamine-induced aphagia: implications for arousal and anhedonia hypotheses of dopamine function. Behavioral neuroscience 103, 36-45 (1989). 112 Tobler, P. N., Fiorillo, C. D. & Schultz, W. Adaptive coding of reward value by dopamine neurons. Science (New York, N.Y.) 307, 1642-1645, doi:10.1126/science.1105370 (2005). 113 Schultz, W. Dopamine reward prediction-error signalling: a two-component response. Nature reviews. Neuroscience 17, 183-195, doi:10.1038/nrn.2015.26 (2016). 114 Morales, M. & Margolis, E. B. Ventral tegmental area: cellular heterogeneity, connectivity and behaviour. Nature reviews. Neuroscience 18, 73-85, doi:10.1038/nrn.2016.165 (2017). 115 Swanson, L. W. The projections of the ventral tegmental area and adjacent regions: a combined fluorescent retrograde tracer and immunofluorescence study in the rat. Brain research bulletin 9, 321-353 (1982). 116 Lammel, S. et al. Unique properties of mesoprefrontal neurons within a dual mesocorticolimbic dopamine system. Neuron 57, 760-773, doi:10.1016/j.neuron.2008.01.022 (2008). 117 Lammel, S., Ion, D. I., Roeper, J. & Malenka, R. C. Projection-specific modulation of dopamine neuron synapses by aversive and rewarding stimuli. Neuron 70, 855-862, doi:10.1016/j.neuron.2011.03.025 (2011). 118 Lammel, S. et al. Input-specific control of reward and aversion in the ventral tegmental area. Nature 491, 212-217, doi:10.1038/nature11527 (2012). 119 Ungerstedt, U. & Arbuthnott, G. W. Quantitative recording of rotational behavior in rats after 6-hydroxy-dopamine lesions of the nigrostriatal dopamine system. Brain research 24, 485-493 (1970). 120 Ungerstedt, U. Adipsia and aphagia after 6-hydroxydopamine induced degeneration of the nigro-striatal dopamine system. Acta physiologica Scandinavica. Supplementum 367, 95-122 (1971). 121 Salamone, J. D., Mahan, K. & Rogers, S. Ventrolateral striatal dopamine depletions impair feeding and food handling in rats. Pharmacology, biochemistry, and behavior 44, 605-610 (1993). 42 122 Salamone, J. D., Zigmond, M. J. & Stricker, E. M. Characterization of the impaired feeding behavior in rats given haloperidol or dopamine-depleting brain lesions. Neuroscience 39, 17-24 (1990). 123 Baldo, B. A., Sadeghian, K., Basso, A. M. & Kelley, A. E. Effects of selective dopamine D1 or D2 receptor blockade within nucleus accumbens subregions on ingestive behavior and associated motor activity. Behavioural brain research 137, 165-177 (2002). 124 Jicha, G. A. & Salamone, J. D. Vacuous jaw movements and feeding deficits in rats with ventrolateral striatal dopamine depletion: possible relation to parkinsonian symptoms. The Journal of neuroscience : the official journal of the Society for Neuroscience 11, 3822-3829 (1991). 125 Aberman, J. E. & Salamone, J. D. Nucleus accumbens dopamine depletions make rats more sensitive to high ratio requirements but do not impair primary food reinforcement. Neuroscience 92, 545-552 (1999). 126 Salamone, J. D. et al. Haloperidol and nucleus accumbens dopamine depletion suppress lever pressing for food but increase free food consumption in a novel food choice procedure. Psychopharmacology 104, 515-521 (1991). 127 Ikemoto, S. & Panksepp, J. Dissociations between appetitive and consummatory responses by pharmacological manipulations of reward-relevant brain regions. Behavioral neuroscience 110, 331-345 (1996). 128 Nicola, S. M. The flexible approach hypothesis: unification of effort and cue-responding hypotheses for the role of nucleus accumbens dopamine in the activation of rewardseeking behavior. The Journal of neuroscience : the official journal of the Society for Neuroscience 30, 16585-16600, doi:10.1523/jneurosci.3958-10.2010 (2010). 129 Cousins, M. S., Sokolowski, J. D. & Salamone, J. D. Different effects of nucleus accumbens and ventrolateral striatal dopamine depletions on instrumental response selection in the rat. Pharmacology, biochemistry, and behavior 46, 943-951 (1993). 130 Nowend, K. L., Arizzi, M., Carlson, B. B. & Salamone, J. D. D1 or D2 antagonism in nucleus accumbens core or dorsomedial shell suppresses lever pressing for food but leads to compensatory increases in chow consumption. Pharmacology, biochemistry, and behavior 69, 373-382 (2001). 131 Sink, K. S., Vemuri, V. K., Olszewska, T., Makriyannis, A. & Salamone, J. D. Cannabinoid CB1 antagonists and dopamine antagonists produce different effects on a task involving response allocation and effort-related choice in food-seeking behavior. Psychopharmacology 196, 565-574, doi:10.1007/s00213-007-0988-4 (2008). 132 Denk, F. et al. Differential involvement of serotonin and dopamine systems in costbenefit decisions about delay or effort. Psychopharmacology 179, 587-596, doi:10.1007/s00213-004-2059-4 (2005). 133 Mai, B., Sommer, S. & Hauber, W. Motivational states influence effort-based decision making in rats: the role of dopamine in the nucleus accumbens. Cognitive, affective & behavioral neuroscience 12, 74-84, doi:10.3758/s13415-011-0068-4 (2012). 43 134 Szczypka, M. S. et al. Feeding behavior in dopamine-deficient mice. Proceedings of the National Academy of Sciences of the United States of America 96, 12138-12143 (1999). 135 Hnasko, T. S. et al. Cre recombinase-mediated restoration of nigrostriatal dopamine in dopamine-deficient mice reverses hypophagia and bradykinesia. Proceedings of the National Academy of Sciences of the United States of America 103, 8858-8863, doi:10.1073/pnas.0603081103 (2006). 136 Steiner, J. E. The gustofacial response: observation on normal and anencephalic newborn infants. Symposium on Oral Sensation and Perception, 254-278 (1973). 137 Pecina, S., Cagniard, B., Berridge, K. C., Aldridge, J. W. & Zhuang, X. Hyperdopaminergic mutant mice have higher "wanting" but not "liking" for sweet rewards. The Journal of neuroscience : the official journal of the Society for Neuroscience 23, 9395-9402 (2003). 138 Wyvell, C. L. & Berridge, K. C. Intra-accumbens amphetamine increases the conditioned incentive salience of sucrose reward: enhancement of reward "wanting" without enhanced "liking" or response reinforcement. The Journal of neuroscience : the official journal of the Society for Neuroscience 20, 8122-8130 (2000). 139 Bassareo, V., De Luca, M. A. & Di Chiara, G. Differential Expression of Motivational Stimulus Properties by Dopamine in Nucleus Accumbens Shell versus Core and Prefrontal Cortex. The Journal of neuroscience : the official journal of the Society for Neuroscience 22, 4709-4719, doi:20026445 (2002). 140 Ahn, S. & Phillips, A. G. Dopaminergic correlates of sensory-specific satiety in the medial prefrontal cortex and nucleus accumbens of the rat. The Journal of neuroscience : the official journal of the Society for Neuroscience 19, RC29 (1999). 141 Beeler, J. A. et al. Taste uncoupled from nutrition fails to sustain the reinforcing properties of food. The European journal of neuroscience 36, 2533-2546, doi:10.1111/j.1460-9568.2012.08167.x (2012). 142 Hajnal, A. & Norgren, R. Sucrose sham feeding decreases accumbens norepinephrine in the rat. Physiology & behavior 82, 43-47, doi:10.1016/j.physbeh.2004.04.024 (2004). 143 Wheeler, R. A. et al. Cocaine cues drive opposing context-dependent shifts in reward processing and emotional state. Biological psychiatry 69, 1067-1074, doi:10.1016/j.biopsych.2011.02.014 (2011). 144 de Araujo, I. E. et al. Food reward in the absence of taste receptor signaling. Neuron 57, 930-941, doi:10.1016/j.neuron.2008.01.032 (2008). 145 Ferreira, J. G., Tellez, L. A., Ren, X., Yeckel, C. W. & de Araujo, I. E. Regulation of fat intake in the absence of flavour signalling. The Journal of physiology 590, 953-972, doi:10.1113/jphysiol.2011.218289 (2012). 146 Branch, S. Y. et al. Food restriction increases glutamate receptor-mediated burst firing of dopamine neurons. The Journal of neuroscience : the official journal of the Society for Neuroscience 33, 13861-13872, doi:10.1523/jneurosci.5099-12.2013 (2013). 44 147 Wilson, C., Nomikos, G. G., Collu, M. & Fibiger, H. C. Dopaminergic correlates of motivated behavior: importance of drive. The Journal of neuroscience : the official journal of the Society for Neuroscience 15, 5169-5178 (1995). 148 Zhen, J., Reith, M. E. & Carr, K. D. Chronic food restriction and dopamine transporter function in rat striatum. Brain research 1082, 98-101, doi:10.1016/j.brainres.2006.01.094 (2006). 149 Sevak, R. J. et al. Feeding conditions differentially affect the neurochemical and behavioral effects of dopaminergic drugs in male rats. European journal of pharmacology 592, 109-115, doi:10.1016/j.ejphar.2008.07.002 (2008). 150 Sharma, S., Hryhorczuk, C. & Fulton, S. Progressive-ratio responding for palatable highfat and high-sugar food in mice. Journal of visualized experiments : JoVE, e3754, doi:10.3791/3754 (2012). 151 Scheggi, S., Secci, M. E., Marchese, G., De Montis, M. G. & Gambarana, C. Influence of palatability on motivation to operate for caloric and non-caloric food in non food-deprived and food-deprived rats. Neuroscience 236, 320-331, doi:10.1016/j.neuroscience.2013.01.027 (2013). 152 Hommel, J. D. et al. Leptin receptor signaling in midbrain dopamine neurons regulates feeding. Neuron 51, 801-810, doi:10.1016/j.neuron.2006.08.023 (2006). 153 Fulton, S. et al. Leptin regulation of the mesoaccumbens dopamine pathway. Neuron 51, 811-822, doi:10.1016/j.neuron.2006.09.006 (2006). 154 Leshan, R. L. et al. Ventral tegmental area leptin receptor neurons specifically project to and regulate cocaine- and amphetamine-regulated transcript neurons of the extended central amygdala. The Journal of neuroscience : the official journal of the Society for Neuroscience 30, 5713-5723, doi:10.1523/jneurosci.1001-10.2010 (2010). 155 Ahima, R. S. et al. Role of leptin in the neuroendocrine response to fasting. Nature 382, 250-252, doi:10.1038/382250a0 (1996). 156 Figlewicz, D. P., Higgins, M. S., Ng-Evans, S. B. & Havel, P. J. Leptin reverses sucroseconditioned place preference in food-restricted rats. Physiology & behavior 73, 229-234 (2001). 157 Figlewicz, D. P. et al. Intraventricular insulin and leptin reverse place preference conditioned with high-fat diet in rats. Behavioral neuroscience 118, 479-487, doi:10.1037/0735-7044.118.3.479 (2004). 158 Fernandes, M. F. et al. Leptin Suppresses the Rewarding Effects of Running via STAT3 Signaling in Dopamine Neurons. Cell metabolism 22, 741-749, doi:10.1016/j.cmet.2015.08.003 (2015). 159 Liu, J., Perez, S. M., Zhang, W., Lodge, D. J. & Lu, X. Y. Selective deletion of the leptin receptor in dopamine neurons produces anxiogenic-like behavior and increases dopaminergic activity in amygdala. Molecular psychiatry 16, 1024-1038, doi:10.1038/mp.2011.36 (2011). 45 160 Opland, D. M., Leinninger, G. M. & Myers, M. G., Jr. Modulation of the mesolimbic dopamine system by leptin. Brain research 1350, 65-70, doi:10.1016/j.brainres.2010.04.028 (2010). 161 Abizaid, A. et al. Ghrelin modulates the activity and synaptic input organization of midbrain dopamine neurons while promoting appetite. The Journal of clinical investigation 116, 3229-3239, doi:10.1172/jci29867 (2006). 162 Egecioglu, E. et al. Ghrelin increases intake of rewarding food in rodents. Addiction biology 15, 304-311, doi:10.1111/j.1369-1600.2010.00216.x (2010). 163 Jerlhag, E. et al. Ghrelin administration into tegmental areas stimulates locomotor activity and increases extracellular concentration of dopamine in the nucleus accumbens. Addiction biology 12, 6-16, doi:10.1111/j.1369-1600.2006.00041.x (2007). 164 Skibicka, K. P., Hansson, C., Alvarez-Crespo, M., Friberg, P. A. & Dickson, S. L. Ghrelin directly targets the ventral tegmental area to increase food motivation. Neuroscience 180, 129-137, doi:10.1016/j.neuroscience.2011.02.016 (2011). 165 Skibicka, K. P., Hansson, C., Egecioglu, E. & Dickson, S. L. Role of ghrelin in food reward: impact of ghrelin on sucrose self-administration and mesolimbic dopamine and acetylcholine receptor gene expression. Addiction biology 17, 95-107, doi:10.1111/j.1369-1600.2010.00294.x (2012). 166 Skibicka, K. P. et al. Divergent circuitry underlying food reward and intake effects of ghrelin: dopaminergic VTA-accumbens projection mediates ghrelin's effect on food reward but not food intake. Neuropharmacology 73, 274-283, doi:10.1016/j.neuropharm.2013.06.004 (2013). 167 Kelly, P. H., Seviour, P. W. & Iversen, S. D. Amphetamine and apomorphine responses in the rat following 6-OHDA lesions of the nucleus accumbens septi and corpus striatum. Brain Res 94, 507-522, doi:0006-8993(75)90233-4 [pii] (1975). 168 Beninger, R. J. The role of dopamine in locomotor activity and learning. Brain research 287, 173-196 (1983). 169 Cagniard, B., Balsam, P. D., Brunner, D. & Zhuang, X. Mice with chronically elevated dopamine exhibit enhanced motivation, but not learning, for a food reward. Neuropsychopharmacology : official publication of the American College of Neuropsychopharmacology 31, 1362-1370, doi:10.1038/sj.npp.1300966 (2006). 170 Giros, B., Jaber, M., Jones, S. R., Wightman, R. M. & Caron, M. G. Hyperlocomotion and indifference to cocaine and amphetamine in mice lacking the dopamine transporter. Nature 379, 606-612, doi:10.1038/379606a0 (1996). 171 Beeler, J. A., Frazier, C. R. & Zhuang, X. Putting desire on a budget: dopamine and energy expenditure, reconciling reward and resources. Frontiers in integrative neuroscience 6, 49, doi:10.3389/fnint.2012.00049 (2012). 46 172 Belke, T. W. Exclusive preference develops less readily on concurrent ratio schedules with wheel-running than with sucrose reinforcement. Journal of the experimental analysis of behavior 94, 135-158, doi:10.1901/jeab.2010.94-135 (2010). 173 Lett, B. T., Grant, V. L., Byrne, M. J. & Koh, M. T. Pairings of a distinctive chamber with the aftereffect of wheel running produce conditioned place preference. Appetite 34, 8794, doi:10.1006/appe.1999.0274 (2000). 174 Greenwood, B. N. et al. Long-term voluntary wheel running is rewarding and produces plasticity in the mesolimbic reward pathway. Behavioural brain research 217, 354-362, doi:10.1016/j.bbr.2010.11.005 (2011). 175 Yang, H. S., Shimomura, K., Vitaterna, M. H. & Turek, F. W. High-resolution mapping of a novel genetic locus regulating voluntary physical activity in mice. Genes, brain, and behavior 11, 113-124, doi:10.1111/j.1601-183X.2011.00737.x (2012). 176 Mathes, W. F. et al. Dopaminergic dysregulation in mice selectively bred for excessive exercise or obesity. Behavioural brain research 210, 155-163, doi:10.1016/j.bbr.2010.02.016 (2010). 177 Belke, T. W. & Pierce, W. D. Wheel-running reinforcement in free-feeding and fooddeprived rats. Behavioural processes 124, 1-9, doi:10.1016/j.beproc.2015.11.018 (2016). 178 Cone, J. J., McCutcheon, J. E. & Roitman, M. F. Ghrelin acts as an interface between physiological state and phasic dopamine signaling. The Journal of neuroscience : the official journal of the Society for Neuroscience 34, 4905-4913, doi:10.1523/jneurosci.4404-13.2014 (2014). 179 Stuber, G. D., Evans, S. B., Higgins, M. S., Pu, Y. & Figlewicz, D. P. Food restriction modulates amphetamine-conditioned place preference and nucleus accumbens dopamine release in the rat. Synapse (New York, N.Y.) 46, 83-90, doi:10.1002/syn.10120 (2002). 180 Marinelli, M., Le Moal, M. & Piazza, P. V. Acute pharmacological blockade of corticosterone secretion reverses food restriction-induced sensitization of the locomotor response to cocaine. Brain research 724, 251-255 (1996). 181 Carr, K. D. & Wolinsky, T. D. Chronic food restriction and weight loss produce opioid facilitation of perifornical hypothalamic self-stimulation. Brain research 607, 141-148 (1993). 182 Geisler, S. & Zahm, D. S. Neurotensin afferents of the ventral tegmental area in the rat: [1] re-examination of their origins and [2] responses to acute psychostimulant and antipsychotic drug administration. The European journal of neuroscience 24, 116-134, doi:10.1111/j.1460-9568.2006.04928.x (2006). 183 Quarta, D. & Smolders, I. Rewarding, reinforcing and incentive salient events involve orexigenic hypothalamic neuropeptides regulating mesolimbic dopaminergic neurotransmission. European journal of pharmaceutical sciences : official journal of the 47 European Federation for Pharmaceutical Sciences 57, 2-10, doi:10.1016/j.ejps.2014.01.008 (2014). 184 Borgland, S. L., Ungless, M. A. & Bonci, A. Convergent actions of orexin/hypocretin and CRF on dopamine neurons: Emerging players in addiction. Brain research 1314, 139144, doi:10.1016/j.brainres.2009.10.068 (2010). 185 van den Pol, A. N. Neuropeptide transmission in brain circuits. Neuron 76, 98-115, doi:10.1016/j.neuron.2012.09.014 (2012). 186 Carraway, R. & Leeman, S. E. The isolation of a new hypotensive peptide, neurotensin, from bovine hypothalami. The Journal of biological chemistry 248, 6854-6861 (1973). 187 Carraway, R. & Leeman, S. E. Characterization of radioimmunoassayable neurotensin in the rat. Its differential distribution in the central nervous system, small intestine, and stomach. The Journal of biological chemistry 251, 7045-7052 (1976). 188 Geisler, S., Berod, A., Zahm, D. S. & Rostene, W. Brain neurotensin, psychostimulants, and stress--emphasis on neuroanatomical substrates. Peptides 27, 2364-2384, doi:10.1016/j.peptides.2006.03.037 (2006). 189 Kitabgi, P., De Nadai, F., Rovere, C. & Bidard, J. N. Biosynthesis, maturation, release, and degradation of neurotensin and neuromedin N. Annals of the New York Academy of Sciences 668, 30-42 (1992). 190 Bayer, V. E., Towle, A. C. & Pickel, V. M. Vesicular and cytoplasmic localization of neurotensin-like immunoreactivity (NTLI) in neurons postsynaptic to terminals containing NTLI and/or tyrosine hydroxylase in the rat central nucleus of the amygdala. Journal of neuroscience research 30, 398-413, doi:10.1002/jnr.490300216 (1991). 191 Elliott, P. J., Chan, J., Parker, Y. M. & Nemeroff, C. B. Behavioral effects of neurotensin in the open field: structure-activity studies. Brain research 381, 259-265 (1986). 192 Bissette, G., Luttinger, D., Mason, G. A., Hernandez, D. E. & Loosen, P. T. Neurotensin and thermoregulation. Annals of the New York Academy of Sciences 400, 268-282 (1982). 193 Leeman, S. E. & Carraway, R. E. Neurotensin: discovery, isolation, characterization, synthesis and possible physiological roles. Annals of the New York Academy of Sciences 400, 1-16 (1982). 194 Lein, E. S. et al. Genome-wide atlas of gene expression in the adult mouse brain. Nature 445, 168-176, doi:10.1038/nature05453 (2007). 195 Tanaka, K., Masu, M. & Nakanishi, S. Structure and functional expression of the cloned rat neurotensin receptor. Neuron 4, 847-854 (1990). 196 Vita, N. et al. Cloning and expression of a complementary DNA encoding a high affinity human neurotensin receptor. FEBS letters 317, 139-142 (1993). 48 197 Mazella, J. et al. Structure, functional expression, and cerebral localization of the levocabastine-sensitive neurotensin/neuromedin N receptor from mouse brain. The Journal of neuroscience : the official journal of the Society for Neuroscience 16, 56135620 (1996). 198 Chalon, P. et al. Molecular cloning of a levocabastine-sensitive neurotensin binding site. FEBS letters 386, 91-94 (1996). 199 Hermans, E., Jeanjean, A. P., Laduron, P. M., Octave, J. N. & Maloteaux, J. M. Postnatal ontogeny of the rat brain neurotensin receptor mRNA. Neuroscience letters 157, 45-48 (1993). 200 Amar, S., Kitabgi, P. & Vincent, J. P. Activation of phosphatidylinositol turnover by neurotensin receptors in the human colonic adenocarcinoma cell line HT29. FEBS letters 201, 31-36 (1986). 201 Vita, N. et al. Neurotensin is an antagonist of the human neurotensin NT2 receptor expressed in Chinese hamster ovary cells. European journal of pharmacology 360, 265272 (1998). 202 Mazella, J. et al. The 100-kDa neurotensin receptor is gp95/sortilin, a non-G-proteincoupled receptor. The Journal of biological chemistry 273, 26273-26276 (1998). 203 Kleczkowska, P. & Lipkowski, A. W. Neurotensin and neurotensin receptors: characteristic, structure-activity relationship and pain modulation--a review. European journal of pharmacology 716, 54-60, doi:10.1016/j.ejphar.2013.03.004 (2013). 204 Lepee-Lorgeoux, I., Betancur, C., Rostene, W. & Pelaprat, D. Differential ontogenetic patterns of levocabastine-sensitive neurotensin NT2 receptors and of NT1 receptors in the rat brain revealed by in situ hybridization. Brain research. Developmental brain research 113, 115-131 (1999). 205 Palacios, J. M., Pazos, A., Dietl, M. M., Schlumpf, M. & Lichtensteiger, W. The ontogeny of brain neurotensin receptors studied by autoradiography. Neuroscience 25, 307-317 (1988). 206 Sarret, P., Beaudet, A., Vincent, J. P. & Mazella, J. Regional and cellular distribution of low affinity neurotensin receptor mRNA in adult and developing mouse brain. The Journal of comparative neurology 394, 344-356 (1998). 207 Walker, N. et al. Tissue distribution and cellular localization of the levocabastinesensitive neurotensin receptor mRNA in adult rat brain. Brain research. Molecular brain research 57, 193-200 (1998). 208 Yamauchi, R. et al. Neurotensin type 2 receptor is involved in fear memory in mice. Journal of neurochemistry 102, 1669-1676, doi:10.1111/j.1471-4159.2007.04805.x (2007). 209 Nouel, D., Sarret, P., Vincent, J. P., Mazella, J. & Beaudet, A. Pharmacological, molecular and functional characterization of glial neurotensin receptors. Neuroscience 94, 1189-1197 (1999). 49 210 Nouel, D. et al. Differential binding profile and internalization process of neurotensin via neuronal and glial receptors. The Journal of neuroscience : the official journal of the Society for Neuroscience 17, 1795-1803 (1997). 211 Sarret, P., Perron, A., Stroh, T. & Beaudet, A. Immunohistochemical distribution of NTS2 neurotensin receptors in the rat central nervous system. The Journal of comparative neurology 461, 520-538, doi:10.1002/cne.10718 (2003). 212 Remaury, A. et al. Targeted inactivation of the neurotensin type 1 receptor reveals its role in body temperature control and feeding behavior but not in analgesia. Brain research 953, 63-72 (2002). 213 Pettibone, D. J. et al. The effects of deleting the mouse neurotensin receptor NTR1 on central and peripheral responses to neurotensin. The Journal of pharmacology and experimental therapeutics 300, 305-313 (2002). 214 Leonetti, M. et al. Specific involvement of neurotensin type 1 receptor in the neurotensinmediated in vivo dopamine efflux using knock-out mice. Journal of neurochemistry 89, 16, doi:10.1046/j.1471-4159.2003.02231.x (2004). 215 Kim, E. R. & Mizuno, T. M. Role of neurotensin receptor 1 in the regulation of food intake by neuromedins and neuromedin-related peptides. Neuroscience letters 468, 64-67, doi:10.1016/j.neulet.2009.10.064 (2010). 216 Rouibi, K., Bose, P., Rompre, P. P. & Warren, R. A. Ventral Midbrain NTS1 Receptors Mediate Conditioned Reward Induced by the Neurotensin Analog, D-Tyr[11]neurotensin. Frontiers in neuroscience 9, 470, doi:10.3389/fnins.2015.00470 (2015). 217 Kempadoo, K. A. et al. Hypothalamic neurotensin projections promote reward by enhancing glutamate transmission in the VTA. The Journal of neuroscience : the official journal of the Society for Neuroscience 33, 7618-7626, doi:10.1523/jneurosci.258812.2013 (2013). 218 Maeno, H. et al. Comparison of mice deficient in the high- or low-affinity neurotensin receptors, Ntsr1 or Ntsr2, reveals a novel function for Ntsr2 in thermal nociception. Brain research 998, 122-129 (2004). 219 St-Gelais, F., Legault, M., Bourque, M. J., Rompre, P. P. & Trudeau, L. E. Role of calcium in neurotensin-evoked enhancement in firing in mesencephalic dopamine neurons. The Journal of neuroscience : the official journal of the Society for Neuroscience 24, 2566-2574, doi:10.1523/jneurosci.5376-03.2004 (2004). 220 Legault, M., Congar, P., Michel, F. J. & Trudeau, L. E. Presynaptic action of neurotensin on cultured ventral tegmental area dopaminergic neurones. Neuroscience 111, 177-187 (2002). 221 Werkman, T. R., Kruse, C. G., Nievelstein, H., Long, S. K. & Wadman, W. J. Neurotensin attenuates the quinpirole-induced inhibition of the firing rate of dopamine neurons in the rat substantia nigra pars compacta and the ventral tegmental area. Neuroscience 95, 417-423 (2000). 50 222 Stuhrman, K. & Roseberry, A. G. Neurotensin inhibits both dopamine- and GABAmediated inhibition of ventral tegmental area dopamine neurons. Journal of neurophysiology 114, 1734-1745, doi:10.1152/jn.00279.2015 (2015). 223 Thibault, D., Albert, P. R., Pineyro, G. & Trudeau, L. E. Neurotensin triggers dopamine D2 receptor desensitization through a protein kinase C and beta-arrestin1-dependent mechanism. The Journal of biological chemistry 286, 9174-9184, doi:10.1074/jbc.M110.166454 (2011). 224 Bose, P., Rompre, P. P. & Warren, R. A. Neurotensin enhances glutamatergic EPSCs in VTA neurons by acting on different neurotensin receptors. Peptides 73, 43-50, doi:10.1016/j.peptides.2015.08.008 (2015). 225 Sotty, F. et al. Differential effects of neurotensin on dopamine release in the caudal and rostral nucleus accumbens: a combined in vivo electrochemical and electrophysiological study. Neuroscience 85, 1173-1182 (1998). 226 Kalivas, P. W., Nemeroff, C. B. & Prange, A. J., Jr. Increase in spontaneous motor activity following infusion of neurotensin into the ventral tegmental area. Brain research 229, 525-529 (1981). 227 Kalivas, P. W. & Duffy, P. Effect of acute and daily neurotensin and enkephalin treatments on extracellular dopamine in the nucleus accumbens. The Journal of neuroscience : the official journal of the Society for Neuroscience 10, 2940-2949 (1990). 228 Steinberg, R. et al. SR 48692, a non-peptide neurotensin receptor antagonist differentially affects neurotensin-induced behaviour and changes in dopaminergic transmission. Neuroscience 59, 921-929 (1994). 229 Sotty, F. et al. Comparative effects of neurotensin, neurotensin(8-13) and [DTyr(11)]neurotensin applied into the ventral tegmental area on extracellular dopamine in the rat prefrontal cortex and nucleus accumbens. Neuroscience 98, 485-492 (2000). 230 Laitinen, K., Crawley, J. N., Mefford, I. N. & De Witte, P. Neurotensin and cholecystokinin microinjected into the ventral tegmental area modulate microdialysate concentrations of dopamine and metabolites in the posterior nucleus accumbens. Brain research 523, 342-346 (1990). 231 Steinberg, R. et al. Neurochemical and behavioural effects of neurotensin vs [DTyr11]neurotensin on mesolimbic dopaminergic function. Neuropeptides 28, 43-50 (1995). 232 Rivest, R., Jolicoeur, F. B. & Marsden, C. A. Use of amfonelic acid to discriminate between classical and atypical neuroleptics and neurotensin: an in vivo voltammetric study. Brain research 544, 86-93 (1991). 233 Glimcher, P. W., Giovino, A. A. & Hoebel, B. G. Neurotensin self-injection in the ventral tegmental area. Brain research 403, 147-150 (1987). 51 234 Rompre, P. P. & Gratton, A. Mesencephalic microinjections of neurotensin-(1-13) and its C-terminal fragment, neurotensin-(8-13), potentiate brain stimulation reward. Brain research 616, 154-162 (1993). 235 Glimcher, P. W., Margolin, D. H., Giovino, A. A. & Hoebel, B. G. Neurotensin: a new 'reward peptide'. Brain research 291, 119-124 (1984). 236 Elliott, P. J. & Nemeroff, C. B. Repeated neurotensin administration in the ventral tegmental area: effects on baseline and D-amphetamine-induced locomotor activity. Neuroscience letters 68, 239-244 (1986). 237 Kalivas, P. W. & Taylor, S. Behavioral and neurochemical effect of daily injection with neurotensin into the ventral tegmental area. Brain research 358, 70-76 (1985). 238 Opland, D. et al. Loss of neurotensin receptor-1 disrupts the control of the mesolimbic dopamine system by leptin and promotes hedonic feeding and obesity. Molecular metabolism 2, 423-434, doi:10.1016/j.molmet.2013.07.008 (2013). 239 Elde, R., Schalling, M., Ceccatelli, S., Nakanishi, S. & Hokfelt, T. Localization of neuropeptide receptor mRNA in rat brain: initial observations using probes for neurotensin and substance P receptors. Neuroscience letters 120, 134-138 (1990). 240 Nicot, A., Rostene, W. & Berod, A. Neurotensin receptor expression in the rat forebrain and midbrain: a combined analysis by in situ hybridization and receptor autoradiography. The Journal of comparative neurology 341, 407-419, doi:10.1002/cne.903410310 (1994). 241 Hokfelt, T., Everitt, B. J., Theodorsson-Norheim, E. & Goldstein, M. Occurrence of neurotensinlike immunoreactivity in subpopulations of hypothalamic, mesencephalic, and medullary catecholamine neurons. The Journal of comparative neurology 222, 543559, doi:10.1002/cne.902220407 (1984). 242 Szigethy, E. & Beaudet, A. Correspondence between high affinity 125I-neurotensin binding sites and dopaminergic neurons in the rat substantia nigra and ventral tegmental area: a combined radioautographic and immunohistochemical light microscopic study. The Journal of comparative neurology 279, 128-137, doi:10.1002/cne.902790111 (1989). 243 Woulfe, J. & Beaudet, A. Immunocytochemical evidence for direct connections between neurotensin-containing axons and dopaminergic neurons in the rat ventral midbrain tegmentum. Brain research 479, 402-406 (1989). 244 Beaudet, A. & Woulfe, J. Morphological substrate for neurotensin-dopamine interactions in the rat midbrain tegmentum. Annals of the New York Academy of Sciences 668, 173185 (1992). 245 Seroogy, K. et al. A subpopulation of dopaminergic neurons in rat ventral mesencephalon contains both neurotensin and cholecystokinin. Brain research 455, 8898 (1988). 52 246 Studler, J. M. et al. Extensive co-localization of neurotensin with dopamine in rat mesocortico-frontal dopaminergic neurons. Neuropeptides 11, 95-100 (1988). 247 Smits, S. M., Terwisscha van Scheltinga, A. F., van der Linden, A. J., Burbach, J. P. & Smidt, M. P. Species differences in brain pre-pro-neurotensin/neuromedin N mRNA distribution: the expression pattern in mice resembles more closely that of primates than rats. Brain research. Molecular brain research 125, 22-28, doi:10.1016/j.molbrainres.2004.03.001 (2004). 248 Berger, B., Gaspar, P. & Verney, C. Colocalization of neurotensin in the mesocortical dopaminergic system. Restricted regional and laminar distribution in rat, lack of colocalization in human. Annals of the New York Academy of Sciences 668, 307-310 (1992). 249 Zahm, D. S., Grosu, S., Williams, E. A., Qin, S. & Berod, A. Neurons of origin of the neurotensinergic plexus enmeshing the ventral tegmental area in rat: retrograde labeling and in situ hybridization combined. Neuroscience 104, 841-851 (2001). 250 Geisler, S. & Zahm, D. S. On the retention of neurotensin in the ventral tegmental area (VTA) despite destruction of the main neurotensinergic afferents of the VTA--implications for the organization of forebrain projections to the VTA. Brain research 1087, 87-104, doi:10.1016/j.brainres.2006.02.108 (2006). 251 Luttinger, D. et al. The effect of neurotensin on food consumption in the rat. European journal of pharmacology 81, 499-503 (1982). 252 Cooke, J. H. et al. Peripheral and central administration of xenin and neurotensin suppress food intake in rodents. Obesity (Silver Spring, Md.) 17, 1135-1143, doi:10.1038/oby.2008.652 (2009). 253 Kim, E. R., Leckstrom, A. & Mizuno, T. M. Impaired anorectic effect of leptin in neurotensin receptor 1-deficient mice. Behavioural brain research 194, 66-71, doi:10.1016/j.bbr.2008.06.024 (2008). 254 Levine, A. S., Kneip, J., Grace, M. & Morley, J. E. Effect of centrally administered neurotensin on multiple feeding paradigms. Pharmacology, biochemistry, and behavior 18, 19-23 (1983). 255 Bailey, C. J. & Flatt, P. R. Anorectic effect of fenfluramine, cholecystokinin and neurotensin in genetically obese (ob/ob) mice. Comparative biochemistry and physiology. A, Comparative physiology 84, 451-454 (1986). 256 Sandoval, S. L. & Kulkosky, P. J. Effects of peripheral neurotensin on behavior of the rat. Pharmacology, biochemistry, and behavior 41, 385-390 (1992). 257 Kelley, A. E., Cador, M., Stinus, L. & Le Moal, M. Neurotensin, substance P, neurokininalpha, and enkephalin: injection into ventral tegmental area in the rat produces differential effects on operant responding. Psychopharmacology 97, 243-252 (1989). 258 Stanley, B. G., Hoebel, B. G. & Leibowitz, S. F. Neurotensin: effects of hypothalamic and intravenous injections on eating and drinking in rats. Peptides 4, 493-500 (1983). 53 259 Richy, S., Burlet, A., Max, J., Burlet, C. & Beck, B. Effect of chronic intraperitoneal injections of leptin on hypothalamic neurotensin content and food intake. Brain Res 862, 276-279 (2000). 260 Sahu, A. Evidence suggesting that galanin (GAL), melanin-concentrating hormone (MCH), neurotensin (NT), proopiomelanocortin (POMC) and neuropeptide Y (NPY) are targets of leptin signaling in the hypothalamus. Endocrinology 139, 795-798, doi:10.1210/endo.139.2.5909 (1998). 261 Sahu, A., Carraway, R. E. & Wang, Y. P. Evidence that neurotensin mediates the central effect of leptin on food intake in rat. Brain research 888, 343-347 (2001). 262 Blum, K., Thanos, P. K. & Gold, M. S. Dopamine and glucose, obesity, and reward deficiency syndrome. Frontiers in psychology 5, 919, doi:10.3389/fpsyg.2014.00919 (2014). 263 Dawe, S. & Loxton, N. J. The role of impulsivity in the development of substance use and eating disorders. Neuroscience and biobehavioral reviews 28, 343-351, doi:10.1016/j.neubiorev.2004.03.007 (2004). 264 Stice, E., Yokum, S., Burger, K. S., Epstein, L. H. & Small, D. M. Youth at risk for obesity show greater activation of striatal and somatosensory regions to food. The Journal of neuroscience : the official journal of the Society for Neuroscience 31, 4360-4366, doi:10.1523/jneurosci.6604-10.2011 (2011). 265 Wang, G. J. et al. Brain dopamine and obesity. Lancet 357, 354-357 (2001). 266 Volkow, N. D. et al. Low dopamine striatal D2 receptors are associated with prefrontal metabolism in obese subjects: possible contributing factors. NeuroImage 42, 1537-1543, doi:10.1016/j.neuroimage.2008.06.002 (2008). 267 de Weijer, B. A. et al. Lower striatal dopamine D2/3 receptor availability in obese compared with non-obese subjects. EJNMMI research 1, 37, doi:10.1186/2191-219x-137 (2011). 268 Kessler, R. M., Zald, D. H., Ansari, M. S., Li, R. & Cowan, R. L. Changes in dopamine release and dopamine D2/3 receptor levels with the development of mild obesity. Synapse (New York, N.Y.) 68, 317-320, doi:10.1002/syn.21738 (2014). 269 Barnard, N. D. et al. D2 dopamine receptor Taq1A polymorphism, body weight, and dietary intake in type 2 diabetes. Nutrition (Burbank, Los Angeles County, Calif.) 25, 5865, doi:10.1016/j.nut.2008.07.012 (2009). 270 Carpenter, C. L., Wong, A. M., Li, Z., Noble, E. P. & Heber, D. Association of dopamine D2 receptor and leptin receptor genes with clinically severe obesity. Obesity (Silver Spring, Md.) 21, E467-473, doi:10.1002/oby.20202 (2013). 271 Stice, E., Spoor, S., Bohon, C. & Small, D. M. Relation between obesity and blunted striatal response to food is moderated by TaqIA A1 allele. Science (New York, N.Y.) 322, 449-452, doi:10.1126/science.1161550 (2008). 54 272 Thompson, J. et al. D2 dopamine receptor gene (DRD2) Taq1 A polymorphism: reduced dopamine D2 receptor binding in the human striatum associated with the A1 allele. Pharmacogenetics 7, 479-484 (1997). 273 Stice, E., Yokum, S., Bohon, C., Marti, N. & Smolen, A. Reward circuitry responsivity to food predicts future increases in body mass: moderating effects of DRD2 and DRD4. NeuroImage 50, 1618-1625, doi:10.1016/j.neuroimage.2010.01.081 (2010). 274 Carlin, J., Hill-Smith, T. E., Lucki, I. & Reyes, T. M. Reversal of dopamine system dysfunction in response to high-fat diet. Obesity (Silver Spring, Md.) 21, 2513-2521, doi:10.1002/oby.20374 (2013). 275 Vucetic, Z., Carlin, J. L., Totoki, K. & Reyes, T. M. Epigenetic dysregulation of the dopamine system in diet-induced obesity. Journal of neurochemistry 120, 891-898, doi:10.1111/j.1471-4159.2012.07649.x (2012). 276 Davis, J. F. et al. Exposure to elevated levels of dietary fat attenuates psychostimulant reward and mesolimbic dopamine turnover in the rat. Behavioral neuroscience 122, 1257-1263, doi:10.1037/a0013111 (2008). 277 Geiger, B. M. et al. Deficits of mesolimbic dopamine neurotransmission in rat dietary obesity. Neuroscience 159, 1193-1199, doi:10.1016/j.neuroscience.2009.02.007 (2009). 278 Johnson, P. M. & Kenny, P. J. Dopamine D2 receptors in addiction-like reward dysfunction and compulsive eating in obese rats. Nature neuroscience 13, 635-641, doi:10.1038/nn.2519 (2010). 279 Caravaggio, F. et al. Ventral striatum binding of a dopamine D2/3 receptor agonist but not antagonist predicts normal body mass index. Biological psychiatry 77, 196-202, doi:10.1016/j.biopsych.2013.02.017 (2015). 280 Karlsson, H. K. et al. Obesity is associated with decreased mu-opioid but unaltered dopamine D2 receptor availability in the brain. The Journal of neuroscience : the official journal of the Society for Neuroscience 35, 3959-3965, doi:10.1523/jneurosci.474414.2015 (2015). 281 Tuominen, L. et al. Aberrant mesolimbic dopamine-opiate interaction in obesity. NeuroImage 122, 80-86, doi:10.1016/j.neuroimage.2015.08.001 (2015). 282 Eisenstein, S. A. et al. A comparison of D2 receptor specific binding in obese and normal-weight individuals using PET with (N-[(11)C]methyl)benperidol. Synapse (New York, N.Y.) 67, 748-756, doi:10.1002/syn.21680 (2013). 283 Kravitz, A. V., O'Neal, T. J. & Friend, D. M. Do Dopaminergic Impairments Underlie Physical Inactivity in People with Obesity? Frontiers in human neuroscience 10, 514, doi:10.3389/fnhum.2016.00514 (2016). 284 Friend, D. M. et al. Basal Ganglia Dysfunction Contributes to Physical Inactivity in Obesity. Cell metabolism 25, 312-321, doi:10.1016/j.cmet.2016.12.001 (2017). 55 285 Beeler, J. A., Faust, R. P., Turkson, S., Ye, H. & Zhuang, X. Low Dopamine D2 Receptor Increases Vulnerability to Obesity Via Reduced Physical Activity, Not Increased Appetitive Motivation. Biological psychiatry 79, 887-897, doi:10.1016/j.biopsych.2015.07.009 (2016). 286 Stice, E., Spoor, S., Bohon, C., Veldhuizen, M. G. & Small, D. M. Relation of reward from food intake and anticipated food intake to obesity: a functional magnetic resonance imaging study. Journal of abnormal psychology 117, 924-935, doi:10.1037/a0013600 (2008). 287 Drewnowski, A., Brunzell, J. D., Sande, K., Iverius, P. H. & Greenwood, M. R. Sweet tooth reconsidered: taste responsiveness in human obesity. Physiology & behavior 35, 617-622 (1985). 288 Rissanen, A. et al. Acquired preference especially for dietary fat and obesity: a study of weight-discordant monozygotic twin pairs. International journal of obesity and related metabolic disorders : journal of the International Association for the Study of Obesity 26, 973-977, doi:10.1038/sj.ijo.0802014 (2002). 289 Rothemund, Y. et al. Differential activation of the dorsal striatum by high-calorie visual food stimuli in obese individuals. NeuroImage 37, 410-421, doi:10.1016/j.neuroimage.2007.05.008 (2007). 290 Stoeckel, L. E. et al. Widespread reward-system activation in obese women in response to pictures of high-calorie foods. NeuroImage 41, 636-647, doi:10.1016/j.neuroimage.2008.02.031 (2008). 291 Berridge, K. C., Ho, C. Y., Richard, J. M. & DiFeliceantonio, A. G. The tempted brain eats: pleasure and desire circuits in obesity and eating disorders. Brain research 1350, 43-64, doi:10.1016/j.brainres.2010.04.003 (2010). 292 Volkow, N. D. et al. Decreased dopamine D2 receptor availability is associated with reduced frontal metabolism in cocaine abusers. Synapse (New York, N.Y.) 14, 169-177, doi:10.1002/syn.890140210 (1993). 293 Volkow, N. D. et al. Low level of brain dopamine D2 receptors in methamphetamine abusers: association with metabolism in the orbitofrontal cortex. The American journal of psychiatry 158, 2015-2021, doi:10.1176/appi.ajp.158.12.2015 (2001). 294 Wang, G. J. et al. Dopamine D2 receptor availability in opiate-dependent subjects before and after naloxone-precipitated withdrawal. Neuropsychopharmacology : official publication of the American College of Neuropsychopharmacology 16, 174-182, doi:10.1016/s0893-133x(96)00184-4 (1997). 295 Blum, K. et al. Increased prevalence of the Taq I A1 allele of the dopamine receptor gene (DRD2) in obesity with comorbid substance use disorder: a preliminary report. Pharmacogenetics 6, 297-305 (1996). 296 Stice, E., Figlewicz, D. P., Gosnell, B. A., Levine, A. S. & Pratt, W. E. The contribution of brain reward circuits to the obesity epidemic. Neuroscience and biobehavioral reviews 37, 2047-2058, doi:10.1016/j.neubiorev.2012.12.001 (2013). 56 297 Colantuoni, C. et al. Evidence that intermittent, excessive sugar intake causes endogenous opioid dependence. Obesity research 10, 478-488, doi:10.1038/oby.2002.66 (2002). 298 Avena, N. M., Rada, P. & Hoebel, B. G. Evidence for sugar addiction: behavioral and neurochemical effects of intermittent, excessive sugar intake. Neuroscience and biobehavioral reviews 32, 20-39, doi:10.1016/j.neubiorev.2007.04.019 (2008). 299 Feifel, D. et al. The reversal of amphetamine-induced locomotor activation by a selective neurotensin-1 receptor agonist does not exhibit tolerance. Psychopharmacology 200, 197-203, doi:10.1007/s00213-008-1197-5 (2008). 300 Beck, B., Nicolas, J. P. & Burlet, C. Neurotensin decreases with fasting in the ventromedian nucleus of obese Zucker rats. Metabolism: clinical and experimental 44, 972-975 (1995). 301 Beck, B., Burlet, A., Nicolas, J. P. & Burlet, C. Hyperphagia in obesity is associated with a central peptidergic dysregulation in rats. The Journal of nutrition 120, 806-811 (1990). 302 Wilding, J. P. et al. Increased neuropeptide-Y messenger ribonucleic acid (mRNA) and decreased neurotensin mRNA in the hypothalamus of the obese (ob/ob) mouse. Endocrinology 132, 1939-1944, doi:10.1210/endo.132.5.7682936 (1993). 303 Williams, G. et al. Reduced hypothalamic neurotensin concentrations in the genetically obese diabetic (ob/ob) mouse: possible relationship to obesity. Metabolism: clinical and experimental 40, 1112-1116 (1991). 304 Boules, M. et al. The neurotensin receptor agonist NT69L suppresses sucrosereinforced operant behavior in the rat. Brain research 1127, 90-98, doi:10.1016/j.brainres.2006.10.025 (2007). 305 Li, J. et al. An obligatory role for neurotensin in high-fat-diet-induced obesity. Nature 533, 411-415, doi:10.1038/nature17662 (2016). 306 Thomas, J. B. et al. Identification of 2-({[1-(4-Fluorophenyl)-5-(2-methoxyphenyl)-1Hpyrazol-3-yl]carbonyl}amino)tricyc lo[3.3.1.13,7]decane-2-carboxylic Acid (NTRC-844) as a Selective Antagonist for the Rat Neurotensin Receptor Type 2. ACS chemical neuroscience 7, 1225-1231, doi:10.1021/acschemneuro.6b00097 (2016). 57 Chapter 2. Determination of Neurotensin Projections to the Ventral Tegmental Area in Mice Authors: Hillary L. Woodworth, Juliette A. Brown, Hannah M. Batchelor, Raluca Bugescu, and Gina M. Leinninger This chapter is a modified version of a manuscript under review. 2.1 Abstract Pharmacologic treatment with the neuropeptide neurotensin (Nts) modifies motivated behaviors such as feeding, locomotor activity, and reproduction. Dopamine (DA) neurons of the ventral tegmental area (VTA) control these behaviors, and Nts directly modulates the activity of DA neurons via Nts receptor-1. However, the endogenous sources of Nts to the VTA remain incompletely understood, impeding determination of which Nts circuits orchestrate specific behaviors. To overcome this obstacle we injected the retrograde tracer fluorogold into the VTA of mice that express GFP in Nts neurons. Identification of GFP-Nts cells that accumulate fluorogold revealed the Nts afferents to the VTA in mice. Similar to rats, most Nts afferents to the VTA of mice arise from the medial and lateral preoptic areas (POA) and the lateral hypothalamic area (LHA), brain regions that are critical for coordination of feeding and reproduction. Additionally, the VTA receives dense input from Nts neurons in the nucleus accumbens shell (NAsh) of mice, and minor Nts projections from the amygdala and periaqueductal gray area. Collectively, our data reveal multiple populations of Nts neurons that provide direct afferents to the VTA and which may regulate specific aspects of motivated behavior. This work lays the foundation for understanding endogenous Nts actions in the VTA, 58 and how circuit-specific Nts modulation may be useful to correct motivational and affective deficits in neuropsychiatric disease. 2.2 Introduction Neurotensin (Nts) is a 13 amino-acid neuropeptide that was first extracted from the bovine hypothalamus by Carraway and Leeman. Nts has subsequently been identified throughout the brain1-4 and has been implicated in regulating a diverse repertoire of physiology and motivated behaviors including feeding, locomotor activity, social behavior, sleep, and response to addictive drugs5-10. Nts may direct certain behaviors via actions in the ventral tegmental area (VTA), based on findings that different types of VTA neurons orchestrate distinct goal-directed behaviors11-13. The VTA is predominantly comprised of dopamine (DA) neurons that project to and release DA in the nucleus accumbens (NA), prefrontal cortex (PFC), hippocampus, or amygdala to modify goal-directed behaviors (see14,15 for review). The VTA also contains GABA-ergic neurons including those within the newly defined “tail” of the VTA or rostromedial tegmental nucleus (RMTg) that locally inhibit VTA DA neurons, thus serving as negative regulators of DA-mediated behaviors12,16-18. Given that alterations in both VTA DA and Nts signaling have been implicated in the pathogenesis of drug addiction, depression, anxiety, schizophrenia, autism, and obesity6,8,19-28, there is likely some functional overlap of the DA and Nts signaling systems. It is therefore critical to define the precise neural mechanisms by which Nts engages the VTA to understand how it regulates diverse motivated behaviors, and how they become maladaptive in disease. Numerous lines of evidence support a role for Nts acting directly within the VTA to modify behavior. Nts signals via the G-protein coupled receptors, Nts receptor 1 and 2 (NtsR1 and NtsR2), both of which are expressed within the VTA of rodents and humans29-38, including on some VTA DA neurons31,38-41. Pharmacologic administration of Nts or Nts analogs into the 59 VTA activates DA neurons37,42-46 and increases DA release in the NA43,44,47,48. Mechanistically, Nts may promote DA signalling via direct stimulation of NtsRs on DA neurons, by enhancing excitatory input to VTA DA neurons and/or by decreasing D2 receptor-mediated autoinhibition42,46,49-51, all of which can lead to DA release that modifies behavior. Indeed, Nts within the VTA suppresses homeostatic and motivated feeding52,53 and increases locomotor activity47,48,52,54-58, identifying dual VTA-mediated behaviors by which Nts modifies energy balance. Pharmacologic Nts in the VTA also supports self-administration50,59,60, conditioned place preference (CPP) 61,62 and locomotor sensitization similar to addictive drugs47,56,63,64, suggesting that Nts may be rewarding. Furthermore, blockade of Nts receptor signalling disrupts or delays sensitization to amphetamine, cocaine, and morphine58,65-67. Intriguingly, many of these behavioral effects are specific to the VTA because Nts administration outside of the VTA elicits different effects. For example, while Nts injection in the VTA increases locomotor activity and DA release, intra-NA or central Nts decreases locomotor activity and does not alter DA release48,68-72. Similarly, intra-VTA Nts does not alter acute locomotor response to psychostimulants56, whereas ICV or intra-NA Nts reduces psychostimulant-induced hyperactivity73-76. One notable exception to this is Nts-mediated suppression of feeding behavior, as both central and direct administration of Nts in the VTA suppress feeding in both fasted and satiated animals52,77,78. Taken together, these data indicate that Nts can act directly in the VTA to regulate motivated locomotor and reward behaviors. Although it is clear that exogenous Nts impacts the mesolimbic DA system, the endogenous sources of Nts input to the VTA remain incompletely understood. Abundant Ntsimmunoreactive terminals are found within the VTA38,40,79,80, including in close apposition to 39 VTA DA neurons , indicating that some Nts is released to the VTA. Nts afferents to the VTA were examined in rats by injecting the retrograde tracer fluorogold (FG) into the VTA and using 60 in situ hybridization (ISH) to label Nts cell bodies, demonstrating that most Nts-ergic inputs to the VTA originate from the preoptic area (POA) and rostral lateral hypothalamic area (LHA)81. Yet, lesioning the POA and LHA in rats does not substantially reduce Nts terminals in the VTA, 82 suggesting there may be other important sources of endogenous input . Indeed, injection of the wheat germ agglutinin transynaptic tracer into the VTA of rats confirmed afferents from the POA and LHA, and identified putative afferents from the NA shell, dorsal raphe (DR), ventral endopiriform area, lateral septum (LS), pedunculopontine tegmental nucleus (PPTg), and 83 laterodorsal tegmental nucleus (LDTg) . Wheat germ agglutinin can move both retrogradely and anterogradely, however, so some of these sites may be targets of VTA neurons rather than providing afferents84,85. Further studies were limited by the inability to easily identify Ntsexpressing cells, but the recent development of NtsCre mice enables the facile detection and manipulation of Nts neurons using cre-lox technology. Using these mice we have identified a large population of Nts neurons in the LHA that project to the VTA, consistent with the prior afferent mapping done in rats, and manipulation of these LHA Nts neurons reveals their crucial contributions to motivated behavior and energy balance39,86-88. The LHA may be just one of several sites by which Nts orchestrates distinct behavioral responses in the VTA, thus it will be important to define all Nts afferents to the VTA and test their roles individually. Since mice and rats differ in Nts expression89, they may also differ in the distribution of Nts afferents to the VTA, thus it is crucial to characterize these in mice. We therefore defined the Nts neurons that project to the VTA in mice by injecting the retrograde tracer FG into the VTA of NtsCre;GFP reporter mice, permitting robust, simultaneous detection of Nts neurons and VTA afferents. 2.3 Results The VTA Contains a Small Population of Non-Dopaminergic Nts Neurons: Nts neurons have been previously difficult to identify because immunohistochemical reagents label fibers, 61 but are insufficient to label cell bodies unless animals are pre-treated with the axonal transport inhibitor, colchicine88. To overcome this, we bred mice expressing Cre-recombinase in Nts neurons88 to a Cre-inducible L10-eGFP reporter90, producing progeny that express GFP selectively in Nts neurons (we refer to these as NtsCre;GFP mice). To understand whether the VTA itself could serve as a local source of Nts input, we first examined the VTA of NtsCre;GFP mice for GFP expression. We observed a small population of Nts neurons within the boundaries of the VTA (372 ± 44 total neurons), which were predominantly localized in the caudolateral region (Fig. 2). Interestingly, 97% of Nts neurons within the VTA did not colocalize with TH, indicating that most VTA Nts neurons are not DA-ergic. Since VTA neurons that express GABA or glutamate can project to VTA DA neurons91,92, these data support the possibility that VTA Nts neurons could locally regulate neighboring VTA neurons. Given the relatively small number of VTA Nts neurons, but the substantial behavioral modifications induced by pharmacological Nts administration to the VTA, we hypothesized that other Nts populations may provide more substantial Nts input to the VTA. Characterization of Nts Neurons that Project to the VTA: To identify all Nts neurons that project to the VTA, we stereotaxically injected NtsCre;GFP mice with the retrograde tracer FG, which labels VTA afferents throughout the brain. Any cell body found to contain both FG (magenta) and GFP (green) represents a Nts neuron that projects to the VTA (gray) (Fig. 3A). We verified VTA targeting by mapping the locations of the midbrain injection tracts and by carefully assessing FG labeling within brain regions known to preferentially project to the VTA vs. the SN, based on the previous work of Watabe-Uchida et al93. For example, at the level of the hypothalamus, VTA afferents are found in the LHA and zona incerta (ZI), but not in the neighboring dorsal striatum (DS), while SN afferents reside preferentially in the DS. Similarly, the NA heavily innervates the VTA, while the DS preferentially innervates the SN93 (Fig. 3B). Out of 10 animals injected with FG, 5 had injection sites within the VTA and had many FG62 labeled neurons in the LHA, ZI, and NA, but not the DS (Fig. 3C-E); these 5 mice were subsequently analyzed to define Nts afferents to the VTA. Although the injection tracts of cases F19 and F29 appeared on the lateral border of the VTA, these mice were included for analysis because their FG expression was consistent with VTA rather than SN targeting (e.g. robust in the LHA, ZI and NA, with minimal FG+ neurons in the DS). While our analysis revealed substantial numbers of GFP-labeled Nts neurons and FG-labeled neurons throughout the brain, we will describe only the select brain regions containing colocalized neurons, henceforth referred to as Nts/FG+ neurons. Striatum and Septum: We observed many Nts/FG+ neurons in the NA shell (NAsh) between +1.75 and 0.85 bregma, but few Nts/FG+ in the NA core (Fig. 4A and Fig. 10). We investigated whether these Nts/FG+ neurons localized to a specific NAsh subregion that might suggest their function, given that site-specific cells along the rostrocaudal NAsh mediate behaviors with either positive or negative valence94. The Nts/FG+ neurons, however, were found within a continuous band across the rostrocaudal axis of the NAsh, with no clear localization to rostral or caudal regions. While Geisler and Zahm reported a minor NAsh Nts input to the VTA in rats83, our quantitative analysis in mice identifies the NAsh as one of the most significant contributors of Nts input to the VTA (Fig. 10). GFP-labeled neurons were also abundant in the olfactory tubercle (OFT), a structure considered to be ventral extension of the NA with similar roles in DA-mediated motivated behaviors95. However, the OFT does not provide significant VTA afferents in mice93,96 and accordingly, we detected few FG-labeled neurons in the OFT (Fig. 4B), suggesting that OFT Nts neurons project to and act at sites other than the VTA. Compared to the ventral striatum, the dorsal striatum contained relatively few Nts neurons, which clustered around the lateral and medial borders of the caudate/putamen (CPu) 63 (Fig. 4C, D). The CPu preferentially projects to the SN with minimal afferents to the VTA93,96; similarly, we observed some FG+ neurons in the ventral CPu between +1.94 and 0.62 bregma, but the vast majority did not colocalize with GFP (Fig. 4C). Nts neurons in lateral septum (LS) project to the VTA in rats83, but we observed ≤1 Nts/FG+ neuron per coronal section in the LS of mice. While Nts neurons were abundant in the LS, there were few FG+ neurons (Fig. 4D), consistent with previous work that the LS does not send substantial projections to the VTA in mice93,96. Cortex and Amygdala: The neocortex contained relatively few GFP-labeled Nts neurons with the exception of a prominent band of Nts neurons in the prefrontal cortex (PFC). The PFC is the predominant source of cortical input to the VTA, while the lateral orbital cortex (LO) also provides afferents specifically to VTA DA neurons93. We noted distinct populations of FGlabeled or Nts neurons in separate layers of the PFC and the LO, but no Nts/FG+ neurons (Fig. 5A, B). Thus, neither the PFC nor the associated neocortex are significant sources of Nts input to the VTA, indicating that they mediate control of the VTA via other, non-Nts signals. Examination of the amygdala revealed FG-labeled neurons in the central amygdala (CeA) and extended amygdala (EA) with relatively few in the basolateral amygdala (BLA), consistent with previous reports93,96,97. Similar to findings from rats, we detected a concentrated area of Nts neurons in the CeA of mice, which were more numerous in the lateral division in caudal sections (Fig. 5C). Some CeA Nts neurons project to the VTA, and were predominantly found between -1.30 and -0.80 bregma (Fig. 5D). A smaller population of Nts/FG+ neurons was observed in the EA. Together these data indicate that the amygdala provides modest Nts input to the VTA in mice (Fig. 10). 64 IPAC, Pallidum and Stria Terminalis: The interstitial nucleus of the posterior limb of the anterior commissure (IPAC) is closely related to both the ventral striatum and the amygdala98,99 and also provides afferents to the VTA93,96. We noted a few Nts/FG+ neurons in the lateral IPAC (Fig. 6A), which appeared to represent a caudal extension of the Nts/FG+ neurons described in the NAsh. By contrast, Nts expression was consistently scarce in the ventral pallidum (VP) and globus pallidus (GP) (Fig. 6A, B), coinciding with previous work showing low levels of Nts in these structures3,100. A few Nts/FG+ neurons were found in the VP, especially near the border with the LPO (Fig. 10). In rats, the stria terminalis (ST) provides Nts afferents to the VTA83, and similarly we identified a few Nts/FG+ neurons in the ST of mice. Together, the ST, VP, and IPAC provided relatively minor Nts inputs to the VTA (Fig. 10). Hypothalamus: Compared to other brain regions, the hypothalamus was the largest contributor of Nts neurons that project to the VTA, with most Nts/FG+ neurons located in the preoptic area (POA), LHA, or zona incerta (ZI). The LHA is a key site for integration of peripheral energy cues with motivated behaviors (see 5 for review) and contained a robust population of Nts neurons between -1.30 to -1.70 bregma. Many of these LHA Nts neurons were labeled with FG from the VTA (Fig. 7B), consistent with other reports that LHA Nts neurons directly project to the VTA39,87,101, and supporting a mechanism by which LHA Nts neurons can coordinate VTA-mediated behaviors such as feeding. Compared to the LHA, other hypothalamic sub-regions contained sparser populations of Nts neurons, including the dorsomedial hypothalamus (DMH), ventromedial hypothalamus (VMH), arcuate nucleus (ARC), and paraventricular hypothalamus (PVH) (Fig. 7B-D). Similar to rats83, we identified a few Nts/FG+ neurons in the ARC and PVH of mice, though the number of ARC and PVH Nts afferents to the VTA are minor compared to the robust number of Nts afferents from the LHA (Fig. 10). The ZI runs along the dorsal border of the LHA but appears to be functionally unique from the LHA and neighboring thalamic areas102. Interestingly, while there are minimal Nts 65 neurons in the ZI, many of them are Nts/FG+ (Fig. 7A, top panel), such that the ZI and LHA contain comparable numbers of Nts neurons projecting to the VTA (Fig. 10). Our examination also revealed a dense population of Nts neurons in the neighboring subthalamic nucleus (STN) (Fig. 7A, bottom panel), but we did not observe FG+ neurons in the STN, which is more likely to provide afferent input to the SN than the VTA93. Indeed, the absence of FG labeling in the STN further verified that FG injections were confined to the VTA in the mice analyzed for this study. Along with the LHA and ZI, the POA provided a large number of Nts afferents to the VTA, whose distribution varied along the rostro-caudal axis. In the caudal POA, the Nts population was most dense in the medial preoptic area (MPO) adjacent to the third ventricle. Many Nts/FG+ neurons were found in the MPO (Fig. 7E) and these VTA-projecting neurons formed a dense cluster centered between -0.22 and +0.02 bregma. Our finding of POA Nts afferents to the VTA in mice is consistent with the recent description of MPO Nts neurons that regulate social reward via the VTA103. Collectively, this data and our work characterizing LHA Nts neurons that project to the VTA39,87,88 provides support for the idea that region-specific Nts input to the VTA controls distinct motivated behaviors (e.g. POA Nts modifies social interaction while LHA Nts modifies energy balance). In the rostral POA the Nts population was sparse in the MPO but dense within the lateral preoptic area (LPO), where we observed many Nts/FG+ neurons (Fig. 7F) spanning between -0.10 and +0.40 bregma. One potential role for these LPO Nts neurons may be initiation of locomotor activity, as LPO-mediated locomotor activity is blunted by microinjection of Nts receptor antagonists in the VTA104. Midbrain and Hindbrain: Brainstem nuclei provide direct sensory afferents to the VTA, thus our identification of Nts/FG+ neurons throughout the midbrain and hindbrain suggests roles for Nts in conveying sensory information to the VTA. First, examination of the dorsal midbrain revealed Nts/FG+ neurons in the superior colliculus (SC) and anterior prectectal nucleus (APT) 66 (Fig. 8A, B), midbrain regions involved in visual and somatosensory processing, respectively105,106. Dense populations of GFP-labeled Nts neurons were noted in the parabrachial nucleus (PBN) and the DR, but few of these contained FG, and thus provided minimal projections to the VTA (Fig. 8F, G). The lateral PBN contained segregated groups of either GFP or FG-labeled neurons with the GFP-labeled neurons clustered more laterally, and most co-labeled neurons were found in the middle where the two populations overlap (Fig. 8G). Nts neurons in the DR were most abundant in caudal sections between -4.8 to -5.0 bregma, which also gave rise to the highest numbers of colocalized neurons per section. VTA afferents arising from the pedunculopontine nucleus (PPTg) and laterodorsal tegmentum (LDTg) are associated with reward behaviors11,107 and we detected small numbers of Nts/FG+ neurons in these regions (Fig. 8D,H). The periaqueductal grey (PAG) contained the highest number of brainstem Nts afferents to the VTA, which were evenly distributed throughout its rostrocaudal extent. Thus while only 4-6 Nts/FG+ neurons were present in each coronal section of the PAG, the cumulative neurons observed across the large rostrocaudal span of the PAG (approximately 2.5 mm) amounted to an appreciable number of Nts afferents from this structure. Habenula and RMTg: Both the lateral habenula (LHb) and RMTg provide afferent input to the VTA that inhibits DA neurons and mediates aversive behaviors11,108. Although most literature strongly suggests that Nts action in the VTA supports reward rather than aversion50,61,62, we examined the LHb and RMTg to assess whether Nts is co-expressed in aversive inputs to the VTA. Analysis of the LHb revealed many FG+ neurons, but sparse GFPNts neurons, and few LHb neurons contained both labels (Fig. 9A). When quantified, we found the LHb Nts neurons provided minimal afferents to the VTA, similar to the modest input from the DMH, VMH and AHA (Fig. 10). 67 The anatomic boundaries of the RMTg in mouse were recently mapped onto the Allen Brain Atlas109, and we used these to locate the RMTg in our experiments. In the caudal RMTg, FG+ neurons occasionally colocalized with GFP, but the majority were GFP-negative (Fig. 9B), suggesting that Nts neurons only occasionally provide aversive input to the VTA. Assessment of rostral RMTg Nts afferents was hindered by its close proximity to the FG injection site, and due to the mechanical tissue damage in this region we could not determine whether RMTg Nts neurons also contained FG. Examination of the rostral RMTg in uninjected NtsCre;GFP reporter mice, however, revealed an average of 3-4 TH-negative Nts neurons per coronal section (Fig. 9C, D). Thus although we cannot determine whether these RMTg Nts neurons project locally, their modest number suggests that Nts is unlikely to mediate aversion via input to the VTA. 2.4 Discussion Here, we characterized the endogenous sources of Nts that can directly regulate the VTA in mouse. Our findings reveal that the largest source of Nts input to the VTA comes from the hypothalamus, predominantly from the LPO, MPO, and LHA. The NAsh also provides significant Nts afferents to the VTA, while the amygdala, ST, VP, and sub-regions of the hindbrain provide more modest Nts input. To our knowledge, this is the first comprehensive definition of mouse Nts afferents to the VTA, and is a first step toward understanding how Nts coordinates a diverse repertoire of behaviors. Indeed, since Nts afferents from the MPO and LHA modify DA-mediated social interaction and energy balance, respectively, it is possible that each Nts circuit orchestrates a specific motivated behavior via the VTA. Going forward it will be important to determine the unique roles of each Nts projection, and hence precise mechanisms to influence feeding, locomotor activity, social behavior, sleep, and response to addictive drugs. While the VTA receives input from many brain structures93,96,110,111, the Nts afferents were confined to a limited number of regions. For example, the DR, VP, and PBN provide major 68 input to the VTA 96,112, and we also observed many FG-labeled neurons in these regions but very few were Nts-specific VTA afferents. By contrast, the highest number of Nts-labeled VTA afferents came from the POA (the LPO and MPO, specifically), yet the POA only provides ~2% of all VTA input in mouse96. Overall, our data support Geisler and Zahm’s hypothesis that Nts inputs to the VTA preferentially originate from regions associated with reward rather than aversion83. Indeed, in mice we find that the largest Nts projections to the VTA originate from the POA, LHA and NAsh, regions that are essential for mediating ingestive and social reward behavior113-115. By contrast, we detected few Nts afferents from aversion-inducing areas that project to the VTA, such as the LHb and RMTg11,116,117. In sum, these data identify specific neural circuits by which Nts may be released to the VTA to mediate positive reinforcement50,59,61,62. Comparison to Previous Work: Our work in mice, taken together with tracing studies in rats, confirms that the LHA and POA are the predominant endogenous sources of Nts to the VTA of rodents83. We did, however, note several discrepancies in the distribution of Nts afferents in mice compared to those in rats, indicative of species differences in Nts signaling. For example, rats have a continuous band of Nts afferents from the LPO and rostral LHA, while mice exhibit significant clusters of Nts afferents in the perifornical LHA (between -1.30 and -1.80 from bregma) and the LPO with sparse afferents between these regions. Additionally, although mice and rats have many Nts neurons in the LS and endopiriform areas118,119, mice lack the substantial Nts-afferents to the VTA observed in rats. This discrepancy may be explained by the general paucity of VTA projections from the LS and endopiriform areas in mice (0.1-0.3% of all VTA afferents96) compared to rats (~2-3% of all VTA afferents112). Thus, Nts from the LS and endopiriform regions of mice and rats likely act via distinct targets, and may differentially influence physiology. Additionally, while the DR provides significant Nts input to the VTA in rats, mice have only modest Nts projections from the DR and other mid- and hindbrain nuclei (e.g. 69 the LDTg, PBN, PPTg). We did, however, detect small populations of VTA-projecting Nts neurons in the APTN, PAG, and SC, which were not previously identified in rats. Additionally, our mouse data are consistent with reports of rat Nts afferents to the VTA from the NAsh, ST and amygdala, though the density of afferents differs between species. Overall, our data reveal species-conserved Nts circuits from the LHA and POA that can engage the VTA, but also point to other species-specific mechanisms for Nts action that may have functional importance. We also addressed whether there are Nts neurons within the VTA that provide a local source of Nts. Previously, immunohistochemistry was used to identify a large population of Nts cell bodies that co-express TH, and hence are DA-ergic79,120 but very few Nts-expressing cells are detected in the VTA via ISH100. This discrepancy may be explained by the subsequent recognition that immunoreagents do not reliably detect Nts in cell bodies without the use of colchicine treatment to inhibit axonal transport. Using NtsCre;GFP reporter mice, we observed only a small population of GFP-labeled Nts neurons within the VTA, similar to Nts ISH. Only 3% of these VTA Nts neurons co-expressed TH, thus the vast majority are not DA-ergic. VTA Nts neurons may instead contain either glutamate or GABA and project locally within the VTA to directly regulate DA neuron firing91. Intriguingly, the small number of TH-negative Nts neurons was primarily detected in the caudal VTA along the rostral lateral borders, where the majority of VTA GABA neurons reside121. These data suggest that VTA Nts neurons may be GABAergic, and could provide local inhibitory control over neighboring DA neurons. Overall, our finding of few local VTA Nts neurons compared to the robust Nts input from the POA, LHA and NAsh suggests that non-VT sources of Nts are the predominant mechanism to modulate the VTA and motivated behaviors. Functional Implications of Nts Afferents to the VTA from the POA: Our results reveal that the both the MPO and LPO subregions of the POA contribute substantial Nts input to the 70 VTA. The POA is sexually dimorphic122,123, and in female rodents MPO-mediated regulation of the mesolimbic DA system influences mating and care of offspring124-126. Nts signaling from the MPO may play a role in maternal behaviors, given that Nts expression in the MPO is altered post-partum127 and in maternal aggression128, and mouse MPO Nts neurons are activated during interaction with pups129. Nts in the POA may also mediate mating behavior. Indeed, Nts immunoreactivity in the VTA is positively correlated with sexually-motivated behaviors in European starlings9,130, while optogenetic stimulation of MPO Nts neurons projecting to the VTA mediates social attraction and mesolimbic DA release103. Taken together, these data suggest that the dense MPO Nts circuit to the VTA is a mechanism by which Nts can modify mating, parenting, and other social behaviors. The LPO provides similarly large numbers of Nts afferents to the VTA as the MPO, yet much less is known about the role of the LPO coordinating motivated behaviors. The LPO is implicated in fluid balance and thirst regulation131, thus it is possible that the abundant Nts neurons in the LPO may modify water intake. However, while central administration of Nts increases water intake132-134, application of Nts directly into the VTA does not alter drinking behavior52, suggesting that LPO Nts neurons may modify drinking via other non-VTA targets. The LPO has also been implicated in regulating locomotor activity. Dis-inhibition of the LPO and rostral LHA continuum in rats induces locomotor activity, which is prevented by blockade of NtsR1-signalling in the VTA104. These data support a role for endogenous LPO Nts release to the VTA in mediating locomotor activity, and may account for the induction of locomotor activity observed after pharmacologic administration of Nts into the VTA. Given the role of the LPO in drinking and locomotor behavior, it will be important to determine whether the large number of LPO Nts afferents to the VTA selectively modulate these, or other motivated behaviors. 71 Nts released from the POA may also modify sleeping behavior via the VTA. Modulation of VTA DA neurons directly regulates arousal, such that the activation of VTA DA neurons promotes wakefulness while inhibition drives sleep and sleep-oriented nesting behaviors135. The afferent inputs to the VTA controlling sleep are not clearly defined, but we demonstrate dense Nts inputs from the POA to the VTA, and the POA is an established regulator of sleepwake behaviors (see 136 for review). GABAergic neurons in the ventrolateral POA promote induction and maintenance of sleep137,138 and project to the VTA139, though it remains to be determined if these neurons co-express Nts. Nts has, however, been implicated in sleeping behavior. For example, central administration of Nts alters sleeping patterns7,140 and NtsR1 knockout mice have altered baseline sleeping behavior and disrupted response to sleep deprivation6. Thus, the large number of Nts inputs to the VTA from the LPO and MPO therefore could potentially play a role in mediating sleep, and may be useful targets to treat the sleep disturbances that commonly co-occur in many neuropsychiatric diseases. Functional Implications of Nts Afferents to the VTA from the LHA: The LHA provided a large number of Nts afferents to the VTA of mice, consistent with prior tracing studies from mice141 and rats83. The LHA is a hub for coordinating peripheral energy status with motivated behaviors such as feeding and locomotor activity that impact body weight (see 5 for review). Indeed, pharmacological Nts treatment reduces feeding and increases locomotor activity in a NtsR1-dependent manner52,53,77,78,142-144, and these effects are mediated, in part, by LHA Nts neurons that project to the VTA39,87,88. The LHA Nts neurons are molecularly distinct from other orexigenic neuropeptide-containing populations such as orexin (OX) and melanin-concentrating hormone (MCH) neurons86, and instead are selectively activated by peripheral cues that suppress feeding such as leptin, inflammation, and dehydration88,145,146. Furthermore, loss of leptin receptor signaling from LHA Nts neurons leads to hypoactivity and weight gain88. Given that LHA Nts neurons respond to cues that suppress feeding and project directly to the VTA, we 72 hypothesize that they coordinate peripheral energy status with feeding behavior and energy expenditure, a process that becomes disrupted in obesity and eating disorders. Since stimulation of LHA fibers projecting to the VTA and Nts administration into the VTA are both rewarding50,59,62,147, LHA Nts afferents may also regulate other motivated via the VTA. For example, some LHA neurons that project to the VTA are activated by acute amphetamine exposure148, expression of morphine CPP149, and cue-induced reinstatement of cocaine150. Going forward, it will therefore be important to define the specific contributions of LHA Nts inputs to the VTA to Nts-mediated behaviors, and whether they modify consumption of both natural and pharmacologic rewards. Functional Implications of Nts Afferents to the VTA from the NA: The NAsh provides substantial Nts projections to the VTA similar to the magnitude from the MPO and LPO. NAsh Nts neurons likely overlap with the GABAergic medium spiny neurons in this region that project to the VTA151,152, which are thought to comprise a feedback loop to regulate the activity of VTA DA neurons153,154. For example, NA medium spiny neuron projections to VTA GABA neurons cause dis-inhibition of neighboring DA neurons, which contributes to the reinforcing effects of cocaine155. NA afferents to the VTA are also preferentially activated during acute amphetamine exposure148 and cue-induced reinstatement of cocaine150. Together these data indicate that NA projections to the VTA may be involved in drug-seeking behavior, thus NAsh Nts projections are anatomically positioned to modify intake of pharmacological rewards. While many studies implicate roles for central or systemic Nts treatment in modifying drug-intake, it remains unclear what part of these effects are mediated specifically by Nts release into the VTA (see 8 for review). Injection of Nts into the VTA does not alter the acute response to amphetamine56, but may be required for the expression of amphetamine sensitization58. Thus, although administration of Nts to the VTA is rewarding, it is unclear whether NAsh Nts inputs specifically modulate the strong reinforcing properties of abusive drugs or natural rewards. 73 Conclusions: Collectively, our work defines the central circuits capable of providing endogenous Nts to the VTA in mice, which are summarized in Fig. 11. Identification of these Nts afferents to the VTA also suggests specific mechanisms that presumably underlie the ability of Nts to modify feeding, sleeping, and social behavior. These data will be useful to direct the application of optogenetic and/or chemogenetic strategies to selectively modulate the activity of site-specified Nts populations, and thereby to define their contributions to behavior and physiology. One limitation of our present work is that it identifies Nts neurons that project to the VTA in general, but does not reveal which specific VTA neurons are targets of Nts signaling (e.g. DAergic, GABAergic, and/or glutamatergic neurons). Since it is now appreciated that VTA neuronal subpopulations control specific motivated behaviors and their reinforcing or aversive valence, it will be crucial to establish how Nts neurons mechanistically engage VTA neurons to fully understand their role in behavior. Indeed, given that Nts inputs from the LHA and POA appear to differentially modify motivated behaviors, there may be distributed Nts circuits and Nts mechanisms to appropriately coordinate environmental stimuli with feeding, arousal and/or social interaction responses. Modulation of specific Nts afferents to the VTA may therefore be useful to normalize maladaptive behaviors associated with neuropsychiatric diseases. 2.5 Methods Animals: Mice were bred and housed in a 12h light/12h dark cycle and cared for by Campus Animal Resources (CAR) at Michigan State University. Animals had ad lib access to chow (Teklad 7913) and water. All animal protocols were approved by the Institutional Animal Care and Use Committee (IACUC) at Michigan State University. Ntscre mice 88 [Jackson stock ♯ 017525], were bred to wild-type C57/Bl6 mice for seven generations to obtain fully backcrossed animals. To visualize Nts neurons, heterozygous Ntscre 74 mice were crossed with homozygous Rosa26EGFP-L10a mice 90, and progeny heterozygous for both Ntscre and Rosa26EGFP-L10a alleles were studied (referred to as Ntscre;GFP). Genotyping was performed using standard PCR using the following primer sequences: Ntscre: common forward: 5' ATA GGC TGC TGA ACC AGG AA, Cre reverse: 5' CCA AAA GAC GGC AAT ATG GT and WT reverse: 5’ CAA TCA CAA TCA CAG GTC AAG AA. Rosa26EGFP-L10a : mutant forward: 5’ TCT ACA AAT GTG GTA GAT CCA GGC, WT forward: 5’ GAG GGG AGT GTT GCA ATA CC and common reverse: 5’ CAG ATG ACT ACC TAT CCT CCC. Adult male and female NtsCre;GFP mice were used for all studies. Surgery: Adult NtsCre;GFP mice received a pre-surgical injection of carprofen (5mg/kg s.c.) and were anesthetized with 3-4% isoflurane/O2 in an induction chamber before being placed in a stereotaxic frame (Kopf). Under 1-2% isoflurane, access holes were drilled in the skull and a guide cannula with stylet (PlasticsOne) was used to deliver 15 nL of FG unilaterally into the VTA in accordance with the atlas of Paxinos and Franklin156 (A/P: -3.2mm, M/L: 0.48mm, D/V: -4.6mm from bregma). Animals were included in the study if 1) the FG injection was targeted to and confined within the VTA and 2) FG-labeled neurons were observed in a pattern consistent with mouse VTA afferents based on the work of Watabe-Uchida et al. and Faget et al.93,96. Out of 10 FG-injected NtsCre;GFP mice, 5 were excluded from analysis because the injection was targeted lateral or dorsal to the VTA and/or the pattern of FG-labeled neurons was more consistent with SN rather than VTA afferents (see Fig. 3 for verification). Perfusions and Immunohistochemistry: Seven days after surgery, FG-injected Ntscre;GFP mice were deeply anesthetized with sodium pentobarbital, transcardially perfused with 10% formalin and brains were post-fixed for 24 hours. After dehydration with 30% sucrose the brains were coronally sectioned (30 µm) using a freezing microtome (Leica). Each brain was sectioned into four separate but equally represented series. Immunofluorescence was 75 performed as previously described88. Sections were incubated with chicken anti-GFP (1:2000, Abcam) and mouse anti-tyrosine hydroxylase (TH) (1:1000, Millipore) or rabbit anti-fluorogold (1:500, Fluorochrome). Imaging and Quantification: Brain sections were analyzed using an Olympus BX53 fluorescence microscope outfitted with FITC and Texas Red filters. Microscope images were collected using Cell Sens software and a Qi-Click 12 Bit cooled camera, and images were analyzed using Photoshop software (Adobe). Each image was assigned a bregma coordinate based on the mouse brain atlas of Paxinos and Franklin156. For each brain area of interest, images from the 5 well-targeted animals were organized by bregma coordinate and placed sideby-side with images of Nts ISH from the Allen Brain Project100. For quantification of Nts neurons in the VTA, four representative coronal sections across similar bregma coordinates from each animal were assessed for total number of Nts neurons and colocalization with TH. Images were viewed and quantified in Photoshop CS6 (Adobe). To estimate the total number of Nts neurons that project to the VTA, 10x images of every section from FG-injected NtsCre;GFP mice were analyzed for the number the neurons that colocalized with Nts and FG. To improve visualization of FG+ neurons, a contrast mask of +100 was applied to each image for quantification. To ensure that each area was sufficiently represented in each animal, the total span over which a brain area was counted in mm (determined by difference between the two furthest bregma coordinates) and the distance was divided by the total number of sections analyzed. Because 30µm brain sections were collected in four separate series, each section represented approximately 0.12 mm of tissue; thus the average mm per section analyzed for each brain area was kept as close to 0.12 as possible to ensure that the entire region was accounted for in the rostrocaudal axis. Animals that did not have enough sections to meet these criteria for a given region were excluded from quantification. To facilitate clarity of neuronal features, all figures 76 depicting FG+ neurons have been digitally enhanced in Photoshop to improve contrast and, in some cases, brightness. 77 APPENDIX 78 Figure 2. Local Nts expression in the VTA. NtsCre;GFP mice (n = 6) were used to determine the number and distribution of VTA Nts neurons and whether or not they co-express TH. A,B,C) Representative images of GFP-identified Nts neurons across three different bregma coordinates in the VTA of NtsCre;GFP mice (scale bar = 200um). Insets highlight individual Nts neurons and the presence or absence of colocalization with TH (gray arrows=TH+ Nts neurons, white arrows=TH-negative Nts neurons). ml=medial lemniscus, ip=interpeduncular nucleus, fr=fasciculus retroflexus, mt=mammillothalamic tract. 79 Figure 3. Validation of VTA targeting in NtsCre;GFP mice injected unilaterally with FG. A) NtsCre;GFP reporter mice were injected unilaterally in the VTA with FG allowing for simultaneous visualization of neurons that project to the VTA (FG positive neurons), Nts neurons (GFP positive neurons) and Nts afferents to the VTA (GFP/FG+ double-positive neurons). B) Expected afferent patterns to the VTA or SN based on Watabe-Uchida et al.93. C) Midbrain images of the five VTA-targeted NtsCre;GFP mice included in the final analysis. Red dotted-line circles indicate the injection sites while the spread of FG is shown in white. D) FG-labeled cell bodies of VTA-targeted animals are confined to the LHA, ZI, CeA, and absent from the DS. Scale bar=200µm. E) FG-labeled neurons of VTA-targeted animals are located preferentially in the ventral, not dorsal striatum. Scale bar=100µm. 80 Figure 4. The NA shell contains clusters of Nts neurons that project to the VTA. A) Many Nts/FG+ neurons were found in the NAsh (yellow arrows), but not NAc. B) Nts expression in the OFT, an area that does not provide significant VTA input and hence few FG+ neurons were observed. C) Representative expression of Nts and FG in the CPu shows little to no colocalization. D) Nts neurons were numerous in the LS, an area that does not provide substantial input to the VTA. Green or purple arrows represent non-colocalized GFP or FG+ neurons in the neighboring CPu. Scale bar=100uM. NAc=nucleus accumbens core, NAsh=nucleus accumbens shell, OFT=olfactory tubercle, CPu=caudate/putamen, LS=lateral septum, MS=medial septum, lv=lateral ventricle, cc=corpus callosum, aca=anterior commissure. 81 Figure 5. Cortical Nts inputs to the VTA originate in the CeA but not neocortex. Representative images show the absence of FG and Nts co-labeling in the A) PFC and B) LO. C) A dense population of Nts neurons resided in the lateral division of the caudal CeA, but these neurons were unlikely to project to the VTA. D) Some Nts neurons in the medial CeA colocalized with FG (yellow arrows). Scale bar=100uM. PFC= prefrontal cortex, CeA=central amygdala, CeM=medial division of CeA, CeL=lateral division of CeA, CeC=central division of central amygdala, BLA=basolateral amygdala, ot=optic tract, EA=extended amygdala, PrL=prelimbic cortex, IL=infralimbic cortex, LO=lateral orbital cortex, ec=external capsule. 82 Figure. 6. Nts neurons in the IPAC, pallidum, and ST that project to the VTA. A) Nts/FG+ neurons were observed in the lateral IPAC (yellow arrows) while the adjacent VP contained low numbers of Nts neurons. B) Nts/FG+ neurons were observed in the ST (yellow arrows), but not GP. Scale bar=100uM. STlp=lateral posterior division of stria terminalis, STlv= lateral ventral division of stria terminalis, stm=medial portion of bed of stria terminalis, sti=intermediate division of bed of stria terminalis, stl=lateral division of bed of stria terminalis, ic=internal capsule, GP=globus pallidus, VP=ventral pallidum, LPO=lateral preoptic area, aca=anterior commissure. 83 Figure 7. Sub-regions of the hypothalamus provide Nts afferents to the VTA. 84 Figure 7 (cont’d). A) Caudal section of the hypothalamus showing Nts/FG+ neurons in the ZI (top panel, yellow arrows) but not the STN (bottom panel). B) The LHA contained a large number of Nts/FG+ neurons (yellow arrows). C) The VMH (top panel) and ARC (bottom panels) did not provide many Nts afferents to the VTA. Purple arrows indicate VTA afferents that did not express GFP. D) The PVH contained minimal Nts neurons and was not a major source of Nts input to the VTA. E) A dense Nts population resided in the caudal MPO and many of these were Nts/FG+ neurons that projected to the VTA (yellow arrows). F) Numerous Nts/FG+ neurons were also observed in the rostral LPO (yellow arrows). Scale bars=100uM. ZI=zona incerta, pSTN=para subthalamic nucleus, cp=cerebral peduncle, mt=mammillothalamic tract, STN=subthalamic nucleus, LHA=lateral hypothalamic area, DMH=dorsomedial hypothalamic area, 3V=third ventricle, f=fornix, VMH=ventromedial hypothalamus, ARC=arcuate nucleus, PVH=paraventricular nucleus, AHA=anterior hypothalamic area, LPO=lateral preoptic area, MPO=medial preoptic area, MPN=medial preoptic nucleus. 85 Figure 8. Brainstem Nts inputs to the VTA. A small number of Nts/FG+ neurons were detected in the A) superior colliculus (SC) and B) anterior pretectal nucleus (APTN). 86 Figure 8 (cont’d). Representative images show occasional FG/Nts+ neurons observed in the C) PAG, D) PTg, E) DR, F) PBN, and G) LDTg. Scale bar=100uM. SC=superior colliculus, iSC=inferior superior colliculus, dSC=dorsal superior colliculus, pc=posterior commissure, aq=cerebral aqueduct, APT=anterior pretectal nucleus, PAG=periaqueductal gray, DLPAG=dorsolateral PAG, LPAG=lateral PAG, VLPAG=ventrolateral PAG, 3N=oculomotor nucleus, DR=dorsal raphe, PTg=peduncolopontine tegmentum, SPTg=subpenduncular tegmental nucleus, xscp=decussation of superior cerebellar peduncle, ATg=anterior tegmental nucleus, LDTg=laterodorsal tegmental nucleus, scp=superior cerebellar peduncle, DTg=dorsal tegmental nucleus, MPBN=medial parabrachial nucleus, LPBN=lateral parabrachial nucleus. 87 Figure 9. Nts-ergic VTA inputs from the LHb and RMTg. A) The LHb contained only occasional colocalized neurons (yellow arrow). B) Some Nts/FG+ neurons were detected in the caudal (yellow arrows). C, D) Images of the rostral RMTg in uninjected NtsCre;GFP mice showing TH-negative Nts neurons (white arrows). D3V=dorsal third ventricle, MHb=medial habenula, LHb=lateral habenula, xscp=decussation of cerbellar peduncle, ts=tectospinal tract, RLi=raphe linear nucleus, ip=interpeduncular nucleus, ml=medial lemniscus. 88 Figure 10. Quantification of Nts-expressing afferents to the VTA by anatomical subregion. Cell bodies colocalizing with GFP and FG were quantified by sampling every fourth coronal section, thus the bars represent approximately one quarter of the total number of neurons in each region. Areas that contained ≤1 colocalized neuron per section were not included in the quantification. 89 Figure 11. Schematic illustration of Nts afferents to the VTA. Distribution patterns of Nts neurons and VTA afferents at different bregma coordinates. Green ovals represent GFPlabeled neurons, pink ovals represent FG-labeled neurons, and white stars indicate colocalized Nts/FG+ neurons. Only the sub-regions included in analysis are outlined in black and labeled, and blank regions do not necessarily indicate lack of GFP or FG. 90 REFERENCES 91 REFERENCES 1 Carraway, R. & Leeman, S. E. The isolation of a new hypotensive peptide, neurotensin, from bovine hypothalami. The Journal of biological chemistry 248, 6854-6861 (1973). 2 Carraway, R. & Leeman, S. E. Characterization of radioimmunoassayable neurotensin in the rat. Its differential distribution in the central nervous system, small intestine, and stomach. The Journal of biological chemistry 251, 7045-7052 (1976). 3 Uhl, G. R., Goodman, R. R. & Snyder, S. H. Neurotensin-containing cell bodies, fibers and nerve terminals in the brain stem of the rat: immunohistochemical mapping. Brain research 167, 77-91 (1979). 4 Jennes, L., Stumpf, W. E. & Kalivas, P. W. Neurotensin: topographical distribution in rat brain by immunohistochemistry. The Journal of comparative neurology 210, 211-224, doi:10.1002/cne.902100302 (1982). 5 Brown, J. A., Woodworth, H. L. & Leinninger, G. M. To ingest or rest? Specialized roles of lateral hypothalamic area neurons in coordinating energy balance. Frontiers in systems neuroscience 9, 9, doi:10.3389/fnsys.2015.00009 (2015). 6 Fitzpatrick, K. et al. Altered sleep and affect in the neurotensin receptor 1 knockout mouse. Sleep 35, 949-956, doi:10.5665/sleep.1958 (2012). 7 Cape, E. G., Manns, I. D., Alonso, A., Beaudet, A. & Jones, B. E. Neurotensin-induced bursting of cholinergic basal forebrain neurons promotes gamma and theta cortical activity together with waking and paradoxical sleep. The Journal of neuroscience : the official journal of the Society for Neuroscience 20, 8452-8461 (2000). 8 Ferraro, L. et al. Neurotensin: A role in substance use disorder? Journal of psychopharmacology (Oxford, England) 30, 112-127, doi:10.1177/0269881115622240 (2016). 9 Merullo, D. P., Cordes, M. A., Susan DeVries, M., Stevenson, S. A. & Riters, L. V. Neurotensin neural mRNA expression correlates with vocal communication and other highly-motivated social behaviors in male European starlings. Physiology & behavior 151, 155-161, doi:10.1016/j.physbeh.2015.07.019 (2015). 10 Gammie, S. C., D'Anna, K. L., Gerstein, H. & Stevenson, S. A. Neurotensin inversely modulates maternal aggression. Neuroscience 158, 1215-1223, doi:10.1016/j.neuroscience.2008.11.045 (2009). 11 Lammel, S. et al. Input-specific control of reward and aversion in the ventral tegmental area. Nature 491, 212-217, doi:10.1038/nature11527 (2012). 12 van Zessen, R., Phillips, J. L., Budygin, E. A. & Stuber, G. D. Activation of VTA GABA neurons disrupts reward consumption. Neuron 73, 1184-1194, doi:10.1016/j.neuron.2012.02.016 (2012). 92 13 Stamatakis, A. M. et al. A unique population of ventral tegmental area neurons inhibits the lateral habenula to promote reward. Neuron 80, 1039-1053, doi:10.1016/j.neuron.2013.08.023 (2013). 14 Salamone, J. D. & Correa, M. The mysterious motivational functions of mesolimbic dopamine. Neuron 76, 470-485, doi:10.1016/j.neuron.2012.10.021 (2012). 15 Bromberg-Martin, E. S., Matsumoto, M. & Hikosaka, O. Dopamine in motivational control: rewarding, aversive, and alerting. Neuron 68, 815-834, doi:10.1016/j.neuron.2010.11.022 (2010). 16 Carr, D. B. & Sesack, S. R. GABA-containing neurons in the rat ventral tegmental area project to the prefrontal cortex. Synapse (New York, N.Y.) 38, 114-123, doi:10.1002/1098-2396(200011)38:2<114::aid-syn2>3.0.co;2-r (2000). 17 Tan, K. R. et al. GABA neurons of the VTA drive conditioned place aversion. Neuron 73, 1173-1183, doi:10.1016/j.neuron.2012.02.015 (2012). 18 Margolis, E. B., Toy, B., Himmels, P., Morales, M. & Fields, H. L. Identification of rat ventral tegmental area GABAergic neurons. PloS one 7, e42365, doi:10.1371/journal.pone.0042365 (2012). 19 Kim, E. R. & Mizuno, T. M. Role of neurotensin receptor 1 in the regulation of food intake by neuromedins and neuromedin-related peptides. Neuroscience letters 468, 64-67, doi:10.1016/j.neulet.2009.10.064 (2010). 20 Ellenbroek, B. A., Angelucci, F., Husum, H. & Mathe, A. A. Gene-environment interactions in a rat model of depression. Maternal separation affects neurotensin in selected brain regions. Neuropeptides 59, 83-88, doi:10.1016/j.npep.2016.05.001 (2016). 21 Theoharides, T. C., Tsilioni, I., Patel, A. B. & Doyle, R. Atopic diseases and inflammation of the brain in the pathogenesis of autism spectrum disorders. Translational psychiatry 6, e844, doi:10.1038/tp.2016.77 (2016). 22 Caceda, R., Binder, E. B., Kinkead, B. & Nemeroff, C. B. The role of endogenous neurotensin in psychostimulant-induced disruption of prepulse inhibition and locomotion. Schizophrenia research 136, 88-95, doi:10.1016/j.schres.2011.10.013 (2012). 23 Boules, M. M., Fredrickson, P., Muehlmann, A. M. & Richelson, E. Elucidating the role of neurotensin in the pathophysiology and management of major mental disorders. Behavioral sciences (Basel, Switzerland) 4, 125-153, doi:10.3390/bs4020125 (2014). 24 Li, J. et al. An obligatory role for neurotensin in high-fat-diet-induced obesity. Nature 533, 411-415, doi:10.1038/nature17662 (2016). 25 Volkow, N. D., Wang, G. J., Tomasi, D. & Baler, R. D. The addictive dimensionality of obesity. Biological psychiatry 73, 811-818, doi:10.1016/j.biopsych.2012.12.020 (2013). 93 26 Rothwell, P. E. Autism Spectrum Disorders and Drug Addiction: Common Pathways, Common Molecules, Distinct Disorders? Frontiers in neuroscience 10, 20, doi:10.3389/fnins.2016.00020 (2016). 27 Howes, O. D., McCutcheon, R., Owen, M. J. & Murray, R. M. The Role of Genes, Stress, and Dopamine in the Development of Schizophrenia. Biological psychiatry 81, 9-20, doi:10.1016/j.biopsych.2016.07.014 (2017). 28 Nestler, E. J. & Carlezon, W. A., Jr. The mesolimbic dopamine reward circuit in depression. Biological psychiatry 59, 1151-1159, doi:10.1016/j.biopsych.2005.09.018 (2006). 29 Fassio, A. et al. Distribution of the neurotensin receptor NTS1 in the rat CNS studied using an amino-terminal directed antibody. Neuropharmacology 39, 1430-1442 (2000). 30 Lepee-Lorgeoux, I., Betancur, C., Rostene, W. & Pelaprat, D. Differential ontogenetic patterns of levocabastine-sensitive neurotensin NT2 receptors and of NT1 receptors in the rat brain revealed by in situ hybridization. Brain research. Developmental brain research 113, 115-131 (1999). 31 Nicot, A., Rostene, W. & Berod, A. Differential expression of neurotensin receptor mRNA in the dopaminergic cell groups of the rat diencephalon and mesencephalon. Journal of neuroscience research 40, 667-674, doi:10.1002/jnr.490400512 (1995). 32 Palacios, J. M. & Kuhar, M. J. Neurotensin receptors are located on dopaminecontaining neurones in rat midbrain. Nature 294, 587-589 (1981). 33 Sarrieau, A. et al. Characterization and autoradiographic distribution of neurotensin binding sites in the human brain. Brain research 348, 375-380 (1985). 34 Quirion, R., Welner, S., Gauthier, S. & Bedard, P. Neurotensin receptor binding sites in monkey and human brain: autoradiographic distribution and effects of 1-methyl-4phenyl-1,2,3,6-tetrahydropyridine treatment. Synapse (New York, N.Y.) 1, 559-566, doi:10.1002/syn.890010608 (1987). 35 Boudin, H., Pelaprat, D., Rostene, W. & Beaudet, A. Cellular distribution of neurotensin receptors in rat brain: immunohistochemical study using an antipeptide antibody against the cloned high affinity receptor. The Journal of comparative neurology 373, 76-89, doi:10.1002/(SICI)1096-9861(19960909)373:1<76::AID-CNE7>3.0.CO;2-A (1996). 36 Leonetti, M. et al. The neurotensin receptor antagonist SR 142948A blocks the efflux of dopamine evoked in nucleus accumbens by neurotensin ejection into the ventral tegmental area. Naunyn-Schmiedeberg's archives of pharmacology 365, 427-433, doi:10.1007/s00210-002-0574-6 (2002). 37 St-Gelais, F., Legault, M., Bourque, M. J., Rompre, P. P. & Trudeau, L. E. Role of calcium in neurotensin-evoked enhancement in firing in mesencephalic dopamine neurons. The Journal of neuroscience : the official journal of the Society for Neuroscience 24, 2566-2574, doi:10.1523/jneurosci.5376-03.2004 (2004). 94 38 Szigethy, E. & Beaudet, A. Correspondence between high affinity 125I-neurotensin binding sites and dopaminergic neurons in the rat substantia nigra and ventral tegmental area: a combined radioautographic and immunohistochemical light microscopic study. The Journal of comparative neurology 279, 128-137, doi:10.1002/cne.902790111 (1989). 39 Opland, D. et al. Loss of neurotensin receptor-1 disrupts the control of the mesolimbic dopamine system by leptin and promotes hedonic feeding and obesity. Molecular metabolism 2, 423-434, doi:10.1016/j.molmet.2013.07.008 (2013). 40 Beaudet, A. & Woulfe, J. Morphological substrate for neurotensin-dopamine interactions in the rat midbrain tegmentum. Annals of the New York Academy of Sciences 668, 173185 (1992). 41 Brouard, A. et al. Mesencephalic dopaminergic neurons in primary cultures express functional neurotensin receptors. The Journal of neuroscience : the official journal of the Society for Neuroscience 12, 1409-1415 (1992). 42 Legault, M., Congar, P., Michel, F. J. & Trudeau, L. E. Presynaptic action of neurotensin on cultured ventral tegmental area dopaminergic neurones. Neuroscience 111, 177-187 (2002). 43 Sotty, F. et al. Differential effects of neurotensin on dopamine release in the caudal and rostral nucleus accumbens: a combined in vivo electrochemical and electrophysiological study. Neuroscience 85, 1173-1182 (1998). 44 Sotty, F. et al. Comparative effects of neurotensin, neurotensin(8-13) and [DTyr(11)]neurotensin applied into the ventral tegmental area on extracellular dopamine in the rat prefrontal cortex and nucleus accumbens. Neuroscience 98, 485-492 (2000). 45 Seutin, V., Massotte, L. & Dresse, A. Electrophysiological effects of neurotensin on dopaminergic neurones of the ventral tegmental area of the rat in vitro. Neuropharmacology 28, 949-954 (1989). 46 Werkman, T. R., Kruse, C. G., Nievelstein, H., Long, S. K. & Wadman, W. J. Neurotensin attenuates the quinpirole-induced inhibition of the firing rate of dopamine neurons in the rat substantia nigra pars compacta and the ventral tegmental area. Neuroscience 95, 417-423 (2000). 47 Kalivas, P. W. & Duffy, P. Effect of acute and daily neurotensin and enkephalin treatments on extracellular dopamine in the nucleus accumbens. The Journal of neuroscience : the official journal of the Society for Neuroscience 10, 2940-2949 (1990). 48 Kalivas, P. W., Burgess, S. K., Nemeroff, C. B. & Prange, A. J., Jr. Behavioral and neurochemical effects of neurotensin microinjection into the ventral tegmental area of the rat. Neuroscience 8, 495-505 (1983). 49 Bose, P., Rompre, P. P. & Warren, R. A. Neurotensin enhances glutamatergic EPSCs in VTA neurons by acting on different neurotensin receptors. Peptides 73, 43-50, doi:10.1016/j.peptides.2015.08.008 (2015). 95 50 Kempadoo, K. A. et al. Hypothalamic neurotensin projections promote reward by enhancing glutamate transmission in the VTA. The Journal of neuroscience : the official journal of the Society for Neuroscience 33, 7618-7626, doi:10.1523/jneurosci.258812.2013 (2013). 51 Stuhrman, K. & Roseberry, A. G. Neurotensin inhibits both dopamine- and GABAmediated inhibition of ventral tegmental area dopamine neurons. Journal of neurophysiology 114, 1734-1745, doi:10.1152/jn.00279.2015 (2015). 52 Cador, M., Kelley, A. E., Le Moal, M. & Stinus, L. Ventral tegmental area infusion of substance P, neurotensin and enkephalin: differential effects on feeding behavior. Neuroscience 18, 659-669 (1986). 53 Kelley, A. E., Cador, M., Stinus, L. & Le Moal, M. Neurotensin, substance P, neurokininalpha, and enkephalin: injection into ventral tegmental area in the rat produces differential effects on operant responding. Psychopharmacology 97, 243-252 (1989). 54 Kalivas, P. W., Nemeroff, C. B. & Prange, A. J., Jr. Increase in spontaneous motor activity following infusion of neurotensin into the ventral tegmental area. Brain research 229, 525-529 (1981). 55 Steinberg, R. et al. SR 48692, a non-peptide neurotensin receptor antagonist differentially affects neurotensin-induced behaviour and changes in dopaminergic transmission. Neuroscience 59, 921-929 (1994). 56 Elliott, P. J. & Nemeroff, C. B. Repeated neurotensin administration in the ventral tegmental area: effects on baseline and D-amphetamine-induced locomotor activity. Neuroscience letters 68, 239-244 (1986). 57 Feifel, D. & Reza, T. L. Effects of neurotensin administered into the ventral tegmental area on prepulse inhibition of startle. Behavioural brain research 106, 189-193 (1999). 58 Panayi, F. et al. Endogenous neurotensin in the ventral tegmental area contributes to amphetamine behavioral sensitization. Neuropsychopharmacology : official publication of the American College of Neuropsychopharmacology 30, 871-879, doi:10.1038/sj.npp.1300638 (2005). 59 Glimcher, P. W., Giovino, A. A. & Hoebel, B. G. Neurotensin self-injection in the ventral tegmental area. Brain research 403, 147-150 (1987). 60 Rompre, P. P. & Gratton, A. Mesencephalic microinjections of neurotensin-(1-13) and its C-terminal fragment, neurotensin-(8-13), potentiate brain stimulation reward. Brain research 616, 154-162 (1993). 61 Rouibi, K., Bose, P., Rompre, P. P. & Warren, R. A. Ventral Midbrain NTS1 Receptors Mediate Conditioned Reward Induced by the Neurotensin Analog, D-Tyr[11]neurotensin. Frontiers in neuroscience 9, 470, doi:10.3389/fnins.2015.00470 (2015). 62 Glimcher, P. W., Margolin, D. H., Giovino, A. A. & Hoebel, B. G. Neurotensin: a new 'reward peptide'. Brain research 291, 119-124 (1984). 96 63 Kalivas, P. W. & Taylor, S. Behavioral and neurochemical effect of daily injection with neurotensin into the ventral tegmental area. Brain research 358, 70-76 (1985). 64 Voyer, D., Levesque, D. & Rompre, P. P. Repeated ventral midbrain neurotensin injections sensitize to amphetamine-induced locomotion and ERK activation: A role for NMDA receptors. Neuropharmacology 112, 150-163, doi:10.1016/j.neuropharm.2016.06.005 (2017). 65 Panayi, F. et al. Chronic blockade of neurotensin receptors strongly reduces sensitized, but not acute, behavioral response to D-amphetamine. Neuropsychopharmacology : official publication of the American College of Neuropsychopharmacology 26, 64-74, doi:10.1016/s0893-133x(01)00354-2 (2002). 66 Felszeghy, K. et al. Neurotensin receptor antagonist administered during cocaine withdrawal decreases locomotor sensitization and conditioned place preference. Neuropsychopharmacology : official publication of the American College of Neuropsychopharmacology 32, 2601-2610, doi:10.1038/sj.npp.1301382 (2007). 67 Levesque, K., Lamarche, C. & Rompre, P. P. Evidence for a role of endogenous neurotensin in the development of sensitization to the locomotor stimulant effect of morphine. European journal of pharmacology 594, 132-138, doi:10.1016/j.ejphar.2008.07.048 (2008). 68 Kalivas, P. W., Nemeroff, C. B. & Prange, A. J., Jr. Neurotensin microinjection into the nucleus accumbens antagonizes dopamine-induced increase in locomotion and rearing. Neuroscience 11, 919-930 (1984). 69 van Wimersma Greidanus, T. B. et al. Neurotensin and bombesin, a relationship between their effects on body temperature and locomotor activity? Pharmacology, biochemistry, and behavior 21, 197-202 (1984). 70 Vadnie, C. A. et al. Activation of neurotensin receptor type 1 attenuates locomotor activity. Neuropharmacology 85, 482-492, doi:10.1016/j.neuropharm.2014.05.046 (2014). 71 Meisenberg, G. & Simmons, W. H. Motor hypoactivity induced by neurotensin and related peptides in mice. Pharmacology, biochemistry, and behavior 22, 189-193 (1985). 72 Elliott, P. J., Chan, J., Parker, Y. M. & Nemeroff, C. B. Behavioral effects of neurotensin in the open field: structure-activity studies. Brain research 381, 259-265 (1986). 73 Skoog, K. M., Cain, S. T. & Nemeroff, C. B. Centrally administered neurotensin suppresses locomotor hyperactivity induced by d-amphetamine but not by scopolamine or caffeine. Neuropharmacology 25, 777-782 (1986). 74 Nemeroff, C. B. et al. Interactions of neurotensin with brain dopamine systems: biochemical and behavioral studies. The Journal of pharmacology and experimental therapeutics 225, 337-345 (1983). 75 Ervin, G. N., Birkemo, L. S., Nemeroff, C. B. & Prange, A. J., Jr. Neurotensin blocks certain amphetamine-induced behaviours. Nature 291, 73-76 (1981). 97 76 Sarhan, S., Hitchcock, J. M., Grauffel, C. A. & Wettstein, J. G. Comparative antipsychotic profiles of neurotensin and a related systemically active peptide agonist. Peptides 18, 1223-1227 (1997). 77 Luttinger, D. et al. The effect of neurotensin on food consumption in the rat. European journal of pharmacology 81, 499-503 (1982). 78 Cooke, J. H. et al. Peripheral and central administration of xenin and neurotensin suppress food intake in rodents. Obesity (Silver Spring, Md.) 17, 1135-1143, doi:10.1038/oby.2008.652 (2009). 79 Hokfelt, T., Everitt, B. J., Theodorsson-Norheim, E. & Goldstein, M. Occurrence of neurotensinlike immunoreactivity in subpopulations of hypothalamic, mesencephalic, and medullary catecholamine neurons. The Journal of comparative neurology 222, 543559, doi:10.1002/cne.902220407 (1984). 80 Woulfe, J. & Beaudet, A. Immunocytochemical evidence for direct connections between neurotensin-containing axons and dopaminergic neurons in the rat ventral midbrain tegmentum. Brain research 479, 402-406 (1989). 81 Zahm, D. S., Grosu, S., Williams, E. A., Qin, S. & Berod, A. Neurons of origin of the neurotensinergic plexus enmeshing the ventral tegmental area in rat: retrograde labeling and in situ hybridization combined. Neuroscience 104, 841-851 (2001). 82 Geisler, S. & Zahm, D. S. On the retention of neurotensin in the ventral tegmental area (VTA) despite destruction of the main neurotensinergic afferents of the VTA--implications for the organization of forebrain projections to the VTA. Brain research 1087, 87-104, doi:10.1016/j.brainres.2006.02.108 (2006). 83 Geisler, S. & Zahm, D. S. Neurotensin afferents of the ventral tegmental area in the rat: [1] re-examination of their origins and [2] responses to acute psychostimulant and antipsychotic drug administration. The European journal of neuroscience 24, 116-134, doi:10.1111/j.1460-9568.2006.04928.x (2006). 84 Louis, G. W., Leinninger, G. M., Rhodes, C. J. & Myers, M. G., Jr. Direct innervation and modulation of orexin neurons by lateral hypothalamic LepRb neurons. The Journal of neuroscience : the official journal of the Society for Neuroscience 30, 11278-11287, doi:10.1523/jneurosci.1340-10.2010 (2010). 85 Hanno, Y., Nakahira, M., Jishage, K., Noda, T. & Yoshihara, Y. Tracking mouse visual pathways with WGA transgene. The European journal of neuroscience 18, 2910-2914 (2003). 86 Brown, J. A. et al. Loss of Action via Neurotensin-Leptin Receptor Neurons Disrupts Leptin and Ghrelin-Mediated Control of Energy Balance. Endocrinology, doi:10.1210/en.2017-00122 (2017). 87 Patterson, C. M. et al. Ventral tegmental area neurotensin signaling links the lateral hypothalamus to locomotor activity and striatal dopamine efflux in male mice. Endocrinology 156, 1692-1700, doi:10.1210/en.2014-1986 (2015). 98 88 Leinninger, G. M. et al. Leptin action via neurotensin neurons controls orexin, the mesolimbic dopamine system and energy balance. Cell metabolism 14, 313-323, doi:10.1016/j.cmet.2011.06.016 (2011). 89 Smits, S. M., Terwisscha van Scheltinga, A. F., van der Linden, A. J., Burbach, J. P. & Smidt, M. P. Species differences in brain pre-pro-neurotensin/neuromedin N mRNA distribution: the expression pattern in mice resembles more closely that of primates than rats. Brain research. Molecular brain research 125, 22-28, doi:10.1016/j.molbrainres.2004.03.001 (2004). 90 Krashes, M. J. et al. An excitatory paraventricular nucleus to AgRP neuron circuit that drives hunger. Nature 507, 238-242, doi:10.1038/nature12956 (2014). 91 Omelchenko, N. & Sesack, S. R. Ultrastructural analysis of local collaterals of rat ventral tegmental area neurons: GABA phenotype and synapses onto dopamine and GABA cells. Synapse (New York, N.Y.) 63, 895-906, doi:10.1002/syn.20668 (2009). 92 Dobi, A., Margolis, E. B., Wang, H. L., Harvey, B. K. & Morales, M. Glutamatergic and nonglutamatergic neurons of the ventral tegmental area establish local synaptic contacts with dopaminergic and nondopaminergic neurons. The Journal of neuroscience : the official journal of the Society for Neuroscience 30, 218-229, doi:10.1523/jneurosci.388409.2010 (2010). 93 Watabe-Uchida, M., Zhu, L., Ogawa, S. K., Vamanrao, A. & Uchida, N. Whole-brain mapping of direct inputs to midbrain dopamine neurons. Neuron 74, 858-873, doi:10.1016/j.neuron.2012.03.017 (2012). 94 Richard, J. M., Castro, D. C., Difeliceantonio, A. G., Robinson, M. J. & Berridge, K. C. Mapping brain circuits of reward and motivation: in the footsteps of Ann Kelley. Neuroscience and biobehavioral reviews 37, 1919-1931, doi:10.1016/j.neubiorev.2012.12.008 (2013). 95 Ikemoto, S. Dopamine reward circuitry: two projection systems from the ventral midbrain to the nucleus accumbens-olfactory tubercle complex. Brain research reviews 56, 27-78, doi:10.1016/j.brainresrev.2007.05.004 (2007). 96 Faget, L. et al. Afferent Inputs to Neurotransmitter-Defined Cell Types in the Ventral Tegmental Area. Cell reports 15, 2796-2808, doi:10.1016/j.celrep.2016.05.057 (2016). 97 Lee, H. J., Gallagher, M. & Holland, P. C. The central amygdala projection to the substantia nigra reflects prediction error information in appetitive conditioning. Learning & memory (Cold Spring Harbor, N.Y.) 17, 531-538, doi:10.1101/lm.1889510 (2010). 98 Shammah-Lagnado, S. J., Alheid, G. F. & Heimer, L. Striatal and central extended amygdala parts of the interstitial nucleus of the posterior limb of the anterior commissure: evidence from tract-tracing techniques in the rat. The Journal of comparative neurology 439, 104-126, doi:10.1002/cne.1999 (2001). 99 Gartner, U., Hartig, W., Riedel, A., Brauer, K. & Arendt, T. Immunocytochemical evidence for the striatal nature of the rat lateral part of interstitial nucleus of the posterior 99 limb of the anterior commissure (IPAC). Journal of chemical neuroanatomy 24, 117-125 (2002). 100 Lein, E. S. et al. Genome-wide atlas of gene expression in the adult mouse brain. Nature 445, 168-176, doi:10.1038/nature05453 (2007). 101 Beier, K. T. et al. Circuit Architecture of VTA Dopamine Neurons Revealed by Systematic Input-Output Mapping. Cell 162, 622-634, doi:10.1016/j.cell.2015.07.015 (2015). 102 Navarro, M. et al. Lateral Hypothalamus GABAergic Neurons Modulate Consummatory Behaviors Regardless of the Caloric Content or Biological Relevance of the Consumed Stimuli. Neuropsychopharmacology : official publication of the American College of Neuropsychopharmacology 41, 1505-1512, doi:10.1038/npp.2015.304 (2016). 103 McHenry, J. A. et al. Hormonal gain control of a medial preoptic area social reward circuit. Nature neuroscience, doi:10.1038/nn.4487 (2017). 104 Reynolds, S. M., Geisler, S., Berod, A. & Zahm, D. S. Neurotensin antagonist acutely and robustly attenuates locomotion that accompanies stimulation of a neurotensincontaining pathway from rostrobasal forebrain to the ventral tegmental area. The European journal of neuroscience 24, 188-196, doi:10.1111/j.1460-9568.2006.04791.x (2006). 105 Jassik-Gerschenfeld, D. Somesthetic and visual responses of superior colliculus neurones. Nature 208, 898-900 (1965). 106 Rees, H. & Roberts, M. H. The anterior pretectal nucleus: a proposed role in sensory processing. Pain 53, 121-135 (1993). 107 Mena-Segovia, J., Winn, P. & Bolam, J. P. Cholinergic modulation of midbrain dopaminergic systems. Brain research reviews 58, 265-271, doi:10.1016/j.brainresrev.2008.02.003 (2008). 108 Lecca, S., Melis, M., Luchicchi, A., Muntoni, A. L. & Pistis, M. Inhibitory inputs from rostromedial tegmental neurons regulate spontaneous activity of midbrain dopamine cells and their responses to drugs of abuse. Neuropsychopharmacology : official publication of the American College of Neuropsychopharmacology 37, 1164-1176, doi:10.1038/npp.2011.302 (2012). 109 Quina, L. A. et al. Efferent pathways of the mouse lateral habenula. The Journal of comparative neurology 523, 32-60, doi:10.1002/cne.23662 (2015). 110 Geisler, S. & Zahm, D. S. Afferents of the ventral tegmental area in the rat-anatomical substratum for integrative functions. The Journal of comparative neurology 490, 270294, doi:10.1002/cne.20668 (2005). 111 Phillipson, O. T. Afferent projections to the ventral tegmental area of Tsai and interfascicular nucleus: a horseradish peroxidase study in the rat. The Journal of comparative neurology 187, 117-143, doi:10.1002/cne.901870108 (1979). 100 112 Yetnikoff, L., Cheng, A. Y., Lavezzi, H. N., Parsley, K. P. & Zahm, D. S. Sources of input to the rostromedial tegmental nucleus, ventral tegmental area, and lateral habenula compared: A study in rat. The Journal of comparative neurology 523, 2426-2456, doi:10.1002/cne.23797 (2015). 113 Hoebel, B. G. & Teitelbaum, P. Hypothalamic control of feeding and self-stimulation. Science (New York, N.Y.) 135, 375-377 (1962). 114 Caggiula, A. R. & Hoebel, B. G. "Copulation-reward site" in the posterior hypothalamus. Science (New York, N.Y.) 153, 1284-1285 (1966). 115 Garris, P. A. et al. Dissociation of dopamine release in the nucleus accumbens from intracranial self-stimulation. Nature 398, 67-69, doi:10.1038/18019 (1999). 116 Stamatakis, A. M. & Stuber, G. D. Activation of lateral habenula inputs to the ventral midbrain promotes behavioral avoidance. Nature neuroscience 15, 1105-1107, doi:10.1038/nn.3145 (2012). 117 Sanchez-Catalan, M. J. et al. Response of the Tail of the Ventral Tegmental Area to Aversive Stimuli. Neuropsychopharmacology : official publication of the American College of Neuropsychopharmacology, doi:10.1038/npp.2016.139 (2016). 118 Alexander, M. J. et al. Distribution of neurotensin/neuromedin N mRNA in rat forebrain: unexpected abundance in hippocampus and subiculum. Proceedings of the National Academy of Sciences of the United States of America 86, 5202-5206 (1989). 119 Sato, M., Kiyama, H., Yoshida, S., Saika, T. & Tohyama, M. Postnatal ontogeny of cells expressing prepro-neurotensin/neuromedin N mRNA in the rat forebrain and midbrain: a hybridization histochemical study involving isotope-labeled and enzyme-labeled probes. The Journal of comparative neurology 310, 300-315, doi:10.1002/cne.903100303 (1991). 120 Bayer, V. E., Towle, A. C. & Pickel, V. M. Vesicular and cytoplasmic localization of neurotensin-like immunoreactivity (NTLI) in neurons postsynaptic to terminals containing NTLI and/or tyrosine hydroxylase in the rat central nucleus of the amygdala. Journal of neuroscience research 30, 398-413, doi:10.1002/jnr.490300216 (1991). 121 Mazei-Robison, M. S. & Nestler, E. J. Opiate-induced molecular and cellular plasticity of ventral tegmental area and locus coeruleus catecholamine neurons. Cold Spring Harbor perspectives in medicine 2, a012070, doi:10.1101/cshperspect.a012070 (2012). 122 Raisman, G. & Field, P. M. Sexual dimorphism in the preoptic area of the rat. Science (New York, N.Y.) 173, 731-733 (1971). 123 Shah, N. M. et al. Visualizing sexual dimorphism in the brain. Neuron 43, 313-319, doi:10.1016/j.neuron.2004.07.008 (2004). 124 Stolzenberg, D. S. et al. Dopamine D1 receptor stimulation of the nucleus accumbens or the medial preoptic area promotes the onset of maternal behavior in pregnancyterminated rats. Behavioral neuroscience 121, 907-919, doi:10.1037/07357044.121.5.907 (2007). 101 125 Numan, M. & Smith, H. G. Maternal behavior in rats: evidence for the involvement of preoptic projections to the ventral tegmental area. Behavioral neuroscience 98, 712-727 (1984). 126 Stolzenberg, D. S. & Numan, M. Hypothalamic interaction with the mesolimbic DA system in the control of the maternal and sexual behaviors in rats. Neuroscience and biobehavioral reviews 35, 826-847, doi:10.1016/j.neubiorev.2010.10.003 (2011). 127 Driessen, T. M., Zhao, C., Whittlinger, A., Williams, H. & Gammie, S. C. Endogenous CNS expression of neurotensin and neurotensin receptors is altered during the postpartum period in outbred mice. PloS one 9, e83098, doi:10.1371/journal.pone.0083098 (2014). 128 Gammie, S. C. et al. Altered gene expression in mice selected for high maternal aggression. Genes, brain, and behavior 6, 432-443, doi:10.1111/j.1601183X.2006.00271.x (2007). 129 Tsuneoka, Y. et al. Functional, anatomical, and neurochemical differentiation of medial preoptic area subregions in relation to maternal behavior in the mouse. The Journal of comparative neurology 521, 1633-1663, doi:10.1002/cne.23251 (2013). 130 Merullo, D. P., Cordes, M. A., Stevenson, S. A. & Riters, L. V. Neurotensin immunolabeling relates to sexually-motivated song and other social behaviors in male European starlings (Sturnus vulgaris). Behavioural brain research 282, 133-143, doi:10.1016/j.bbr.2015.01.007 (2015). 131 Blass, E. M. & Epstein, A. N. A lateral preoptic osmosensitive zone for thirst in the rat. Journal of comparative and physiological psychology 76, 378-394 (1971). 132 Hawkins, M. F., Baker, J. D. & Baumeister, A. A. Neurotensin-induced polydipsia: a structure-activity study. Brain research 487, 188-191 (1989). 133 Quirk, W. S., Wright, J. W. & Harding, J. W. Tachyphylaxis of dipsogenic activity to intracerebroventricular administration of angiotensins. Brain research 452, 73-78 (1988). 134 Baker, J. D., Hawkins, M. F., Baumeister, A. A. & Nagy, M. Microinjection of neurotensin into the CNS induces hyperdipsia in the rat. Pharmacology, biochemistry, and behavior 33, 7-10 (1989). 135 Eban-Rothschild, A., Rothschild, G., Giardino, W. J., Jones, J. R. & de Lecea, L. VTA dopaminergic neurons regulate ethologically relevant sleep-wake behaviors. Nature neuroscience 19, 1356-1366, doi:10.1038/nn.4377 (2016). 136 Luppi, P. H., Peyron, C. & Fort, P. Not a single but multiple populations of GABAergic neurons control sleep. Sleep medicine reviews, doi:10.1016/j.smrv.2016.03.002 (2016). 137 Gvilia, I., Turner, A., McGinty, D. & Szymusiak, R. Preoptic area neurons and the homeostatic regulation of rapid eye movement sleep. The Journal of neuroscience : the official journal of the Society for Neuroscience 26, 3037-3044, doi:10.1523/jneurosci.4827-05.2006 (2006). 102 138 Sherin, J. E., Shiromani, P. J., McCarley, R. W. & Saper, C. B. Activation of ventrolateral preoptic neurons during sleep. Science (New York, N.Y.) 271, 216-219 (1996). 139 Saito, Y. C. et al. GABAergic neurons in the preoptic area send direct inhibitory projections to orexin neurons. Frontiers in neural circuits 7, 192, doi:10.3389/fncir.2013.00192 (2013). 140 Castel, M. N., Stutzmann, J. M., Lucas, M., Lafforgue, J. & Blanchard, J. C. Effects of ICV administration of neurotensin and analogs on EEG in rats. Peptides 10, 95-101 (1989). 141 Leinninger, G. M. et al. Leptin acts via leptin receptor-expressing lateral hypothalamic neurons to modulate the mesolimbic dopamine system and suppress feeding. Cell metabolism 10, 89-98, doi:10.1016/j.cmet.2009.06.011 (2009). 142 Feifel, D., Goldenberg, J., Melendez, G. & Shilling, P. D. The acute and subchronic effects of a brain-penetrating, neurotensin-1 receptor agonist on feeding, body weight and temperature. Neuropharmacology 58, 195-198, doi:10.1016/j.neuropharm.2009.07.001 (2010). 143 Boules, M. et al. A novel neurotensin peptide analog given extracranially decreases food intake and weight in rodents. Brain research 865, 35-44 (2000). 144 Remaury, A. et al. Targeted inactivation of the neurotensin type 1 receptor reveals its role in body temperature control and feeding behavior but not in analgesia. Brain research 953, 63-72 (2002). 145 Grossberg, A. J. et al. Inflammation-induced lethargy is mediated by suppression of orexin neuron activity. The Journal of neuroscience : the official journal of the Society for Neuroscience 31, 11376-11386, doi:10.1523/jneurosci.2311-11.2011 (2011). 146 Watts, A. G., Sanchez-Watts, G. & Kelly, A. B. Distinct patterns of neuropeptide gene expression in the lateral hypothalamic area and arcuate nucleus are associated with dehydration-induced anorexia. The Journal of neuroscience : the official journal of the Society for Neuroscience 19, 6111-6121 (1999). 147 Nieh, E. H. et al. Decoding neural circuits that control compulsive sucrose seeking. Cell 160, 528-541, doi:10.1016/j.cell.2015.01.003 (2015). 148 Colussi-Mas, J., Geisler, S., Zimmer, L., Zahm, D. S. & Berod, A. Activation of afferents to the ventral tegmental area in response to acute amphetamine: a double-labelling study. The European journal of neuroscience 26, 1011-1025, doi:10.1111/j.14609568.2007.05738.x (2007). 149 Richardson, K. A. & Aston-Jones, G. Lateral hypothalamic orexin/hypocretin neurons that project to ventral tegmental area are differentially activated with morphine preference. The Journal of neuroscience : the official journal of the Society for Neuroscience 32, 3809-3817, doi:10.1523/jneurosci.3917-11.2012 (2012). 150 Mahler, S. V. & Aston-Jones, G. S. Fos activation of selective afferents to ventral tegmental area during cue-induced reinstatement of cocaine seeking in rats. The Journal 103 of neuroscience : the official journal of the Society for Neuroscience 32, 13309-13326, doi:10.1523/jneurosci.2277-12.2012 (2012). 151 Nauta, W. J., Smith, G. P., Faull, R. L. & Domesick, V. B. Efferent connections and nigral afferents of the nucleus accumbens septi in the rat. Neuroscience 3, 385-401 (1978). 152 Kalivas, P. W., Churchill, L. & Klitenick, M. A. GABA and enkephalin projection from the nucleus accumbens and ventral pallidum to the ventral tegmental area. Neuroscience 57, 1047-1060 (1993). 153 Einhorn, L. C., Johansen, P. A. & White, F. J. Electrophysiological effects of cocaine in the mesoaccumbens dopamine system: studies in the ventral tegmental area. The Journal of neuroscience : the official journal of the Society for Neuroscience 8, 100-112 (1988). 154 Rahman, S. & McBride, W. J. Feedback control of mesolimbic somatodendritic dopamine release in rat brain. Journal of neurochemistry 74, 684-692 (2000). 155 Bocklisch, C. et al. Cocaine disinhibits dopamine neurons by potentiation of GABA transmission in the ventral tegmental area. Science (New York, N.Y.) 341, 1521-1525, doi:10.1126/science.1237059 (2013). 156 Keith B.J. Franklin, G. P. The Mouse Brain In Stereotaxic Coordinates. Third edn, (Elsevier, 2007). 157 Kiyama, H., Emson, P. C. & Ruth, J. Distribution of Tyrosine Hydroxylase mRNA in the Rat Central Nervous System Visualized by Alkaline Phosphatase in situ Hybridization Histochemistry. The European journal of neuroscience 2, 512-524 (1990). 104 Chapter 3. Identification of Neurotensin Receptor Expressing Cells in the Ventral Tegmental Area Across the Lifespan Authors: Hillary L. Woodworth, Bethany G. Beekly, Trevor J. Lewis, and Gina M. Leinninger The is a modified version of a manuscript under revision. 3.1 Abstract Neurotensin (Nts) promotes activation of dopamine (DA) neurons in the ventral tegmental area (VTA) via incompletely understood mechanisms. Nts can signal via the G-protein coupled neurotensin receptors 1 and 2 (NtsR1 and NtsR2), but the lack of methods to detect NtsR1 and NtsR2-expressing cells has limited mechanistic understanding of Nts action. To overcome this challenge, we generated dual recombinase mice that express FlpO-dependent Cre recombinase in NtsR1 or NtsR2 cells. This strategy permitted temporal control over recombination, such that we could identify NtsR1 or NtsR2-expressing cells and determine if their distributions differed between the developing and adult brain. Using this system we found that NtsR1 is transiently expressed in nearly all DA neurons and in many non-DA neurons in the VTA during development. However, NtsR1 expression is more restricted within the adult brain, where only two thirds of VTA DA neurons expressed NtsR1. By contrast, NtsR2 expression remains constant throughout lifespan but it is predominantly expressed within glia. Anterograde tract tracing revealed that NtsR1 is selectively expressed by mesolimbic, not mesocortical DA neurons, thus VTA NtsR1 neurons are a projection-specified and functionally unique subset of VTA DA neurons. Collectively, this work reveals a cellular mechanism by which Nts can directly engage NtsR1-expressing DA neurons to modify DA signaling. Going forward, the dual recombinase strategy developed here will be useful to selectively modulate NtsR1 and NtsR2expressing cells, and to parse their contributions to Nts-mediated behaviors. 105 3.2 Introduction The neuropeptide, neurotensin (Nts), is expressed throughout the central nervous system and has been linked to numerous physiologic and behavioral processes including regulation of feeding, body temperature, nociception, emotional state, and sleep1-5. While the precise neural circuits by which Nts acts to modify behavior remain undetermined, it is wellestablished that Nts modulates dopamine (DA) transmission. Nts signaling may therefore be a target of interest for the treatment of diseases linked with altered DA signaling, including Parkinson’s disease, schizophrenia, drug addiction, and obesity5,6. Nts signals via the G-protein coupled neurotensin receptors 1 and 2 (NtsR1 and NtsR2) which are both expressed in the brain and have been implicated in Nts-mediated behaviors7-10. Nts also binds the intracellular receptor, NtsR3/sortilin, which may regulate recycling and sorting of Nts3,11. Studies using pharmacological antagonists and mice null for NtsR1 or NtsR2 suggest that each receptor isoform mediates distinct aspects of physiology, and hence that they may be expressed on different cell types. The lack of methods to detect and manipulate NtsR1 or NtsR2-expressing cells, however, has limited understanding of the cellular mechanisms by which Nts regulates behavior. NtsR1 was the first NtsR identified and has a 10-30 fold higher affinity for Nts compared to NtsR27-10. Nts binding to NtsR1 activates Gq-coupled induction of phospholipase C and intracellular Ca++ release7,8,12. The signaling mechanism of NtsR2 is less clear, and both Gq and Gi-coupled mechanisms have been reported9,13. In situ hybridization (ISH) and autoradiography methods to detect NtsR1 indicate that it is expressed robustly within the ventral tegmental area (VTA) and substantia (SN) of adult animals14-17. Similar techniques revealed diffuse expression of NtsR2 throughout the brain that may be within both neurons and glia18-20, with the highest levels of expression restricted to the cerebellum, hippocampus, and periaqueductal gray (PAG)9,19,21. Interestingly, the expression patterns of NtsR1 and NtsR2 in 106 the brain may vary with age. For example, central NtsR1 is transiently upregulated during gestation and peaks during the first week of life, but is subsequently downregulated as animals reach maturity, with high levels persisting in select brain regions including the VTA and SN22,23. By contrast, NtsR2 expression is low at birth but gradually increases with age, peaking in adulthood21,22. Taken together, these data indicate that NtsR1 and NtsR2 have distinct expression patterns that vary across the lifespan, and may be found on different cell types within the nervous system. Given the different sites and distribution of NtsR1 and NtsR2 expression throughout lifespan, each receptor isoform may regulate distinct aspects of developmental and adult physiology. Indeed, previous work demonstrates that central Nts promotes DA release, locomotor activity, hypothermia, anorexia, and reward via NtsR11,24-29, whereas NtsR2 may confer the pain-reducing effects of Nts1,3,30,31. However, the evidence for distinct roles of NtsR1 and NtsR2 is not entirely consistent and interpretation of the data is complicated by methodological limitations. For example, the commonly used NtsR1-selective antagonist SR48692 also acts as an agonist at NtsR213,32,33, while a potential compound to selectively antagonize NtsR2 has only recently been developed3,34. NtsR1 and NtsR2 knockout mice have also been used to examine the specific roles of each receptor, but developmental deletion in these models may lead to compensatory changes that mask normal action of the Nts system. For example, it has been suggested that NtsR1 or NtsR2 null mice have increased expression of the remaining receptor isoform26,35 and exhibit alterations in DA signaling and locomotor activity that misrepresent the functional roles of the NtsR receptor signaling1,24,26,35,36. Finally, the lack of immunohistochemical reagents to detect NtsR1 or NtsR2 has impeded determination of the cells and circuits that mediate Nts action. Thus, while NtsR-selective pharmacologic agents and knockout models have added to understanding of central Nts action, developing 107 methods to visualize and manipulate select NtsR1 or NtsR2 populations in vivo is essential to deciphering the neural circuits and physiology regulated by each receptor. To address this challenge, we developed dual recombinase knock-in mouse models in which FlpO is required to induce IRES-Cre in cells that express NtsR1 or NtsR2. Cre-mediated recombination can then be used to induce fluorescent reporters in these cells to permit their detection, and indeed Cre-driver lines have proven to be reliable reagents to identify genetically specified cell populations37-39. A limitation of mouse models that constitutively express Cre is that recombination-mediated labeling occurs whenever Cre is expressed and persists thereafter. This may confound interpretation in cases where gene-induced Cre expression changes over lifespan, as has been suggested to occur for NtsRs (with extensive expression during development but more restricted expression in the adult brain22,23). We thus engineered FlpOdependent Cre expression in NtsR1 and NtsR2 cells, allowing for temporal control over recombination by inducing FlpO expression at defined time points (either embryogenesis or adulthood). Given the well-established description of Nts as a modulator of DA signaling, but the lack of understanding of which VTA cells mediate it, we used these mice to define the cellular distribution of NtsR1 and NtsR2 within the VTA. 3.3 Results Dual Recombinase Strategy to Label NtsR1 and NtsR2 Expressing Cells: To identify NtsR1 or NtsR2 cells “on command”, we generated knock-in mouse models that express Cre in NtsR1 or NtsR2 cells only after FlpO-mediated recombination. To do this, we inserted an frtflanked NEO cassette upstream of an IRES-Cre sequence and cloned it into the non-coding region of the Ntsr1 or Ntsr2 genomic sequences; we refer to these as 108 NtsR1NEO-Cre and NtsR2NEO-Cre mice (Fig. 12). The frt-flanked NEO cassette blocks Cre expression unless NEO is removed, thus NtsR1NEO-Cre and NtsR2NEO-Cre mice lack Cre expression until exposure to FlpO. To reveal cells that express NtsR1 and NtsR2 during development, we crossed NtsR1NEO-Cre and NtsR2NEO-Cre mice to a FlpO deleter line, producing progeny that lack the frtflanked NEO cassette and thus produce Cre whenever NtsR1 or NtsR2 is transcribed throughout the lifespan (NtsR1∆NEO-Cre and NtsR2∆NEO-Cre mice). The NtsR1∆NEO-Cre and NtsR2∆NEO-Cre mice were subsequently bred to a Cre-inducible eGFP-L10a reporter line; in progeny heterozygous for each allele, any cell that expresses NtsR1 or NtsR2 will undergo Cremediated recombination to express GFP. Notably, the recombination is permanent, so GFP labeling persists even in cells that cease to express NtsR1 or NtsR2. This model enables us to visualize any cells that expressed NtsR1 or NtsR2 from conception onward, and we refer to these as NtsR1Dev;GFP and NtsR2Dev;GFP mice (Fig. 12A). To study the adult expression pattern of NtsR1 and NtsR2, we bred NtsR1NEO-Cre and NtsR2NEO-Cre mice to the Cre-inducible eGFP-L10a reporter. In this case, the progeny (NtsR1NEO-Cre;GFP and NtsR2NEO-Cre;GFP) carry the GFP allele, but no reporter is expressed because the frt-flanked NEO cassette suppresses Cre expression. These mice were maintained in our colony until they reached adulthood, then were injected with an adenovirus expressing FlpO recombinase into the lateral ventricles to permit FlpO expression throughout the adult brain. FlpO excises the frt-flanked NEO, which permits Cre-mediated GFP expression only in cells that actively express NtsR1 or NtsR2, allowing us to visualize the adult expression pattern of NtsR1 and NtsR2 (NtsR1Adult;GFP and NtsR2Adult;GFP) (Fig. 12B). 109 Distribution and Morphology of NtsR1 and NtsR2 in the VTA: Previously, identification of NtsR1 and NtsR2 neurons has been challenging given the lack of reliable immunohistochemical reagents to visualize them. The dual recombinase mouse models we generated, however, enabled us to visualize the distributions of NtsR1 and NtsR2 in the VTA, and whether they differed between developmental and adult stages. First, analysis of NtsR1Dev;GFP mice revealed a wide-spread, dense population of GFP+ neurons in the VTA (Fig. 13A) and throughout the brain (data not shown). Despite many GFP-labeled neurons, we ruled out the possibility of ectopic expression via lack of GFP in regions that are thought to have minimal NtsR1 expression at any point in life, including the interpeduncular nucleus (ip)40 (see ip in Fig. 13A). By contrast, NtsR1Adult;GFP mice had far fewer GFP+ neurons in general and were primarily restricted to the VTA and SN (Fig. 13B). These findings are consistent with previous work showing that central NtsR1 expression peaks during gestation, but subsequently decreases and only remains prominent within the midbrain of adult rodents22,23. Analysis of NtsR2-reporter mice revealed two morphologically distinct populations of GFP-labeled cells in the VTA. A few GFP+ cells displayed clear neuronal features (Fig. 13C, white arrows) but the vast majority of GFP-labeled cells in both developmental and adult models were detected in cells with diffuse, stellate morphology indicative of glial cells (Fig. 13C). This finding supports previous work suggesting that NtsR2 is expressed on astrocytes20,41,42. Neuronal GFP expression was sparse in both NtsR2Dev;GFP and NtsR2Adult;GFP models but slightly more neurons were detected in NtsR2Dev;GFP mice (Fig. 13C,D). When quantified, we found that NtsR1-GFP+ neurons outnumber NtsR2-GFP+ neurons approximately 18 to 1 (Fig. 13E). While the total number of NtsR1-GFP+ neurons in adult animals is about half of that observed in the developmental model, adult NtsR1-GFP+ neurons outnumber NtsR2-GFP+ neurons by 60 to 1 (Fig. 13F). Collectively, these data indicate that NtsR1 is the predominant isoform expressed on VTA neurons in development and adulthood, while NtsR2 is primarily 110 expressed on cells with glial morphology. Furthermore, our findings reveal that many cells express NtsR1 and NtsR2 at some stage of development, but in the adult brain, NtsR1 expression is confined to a subset of VTA cells. NtsR1 is the Predominant Isoform on VTA DA Neurons: To define the neurochemical phenotype of NtsR1 and NtsR2 neurons, we examined VTA sections from the adult and developmental mouse models via immunofluorescence for tyrosine-hydroxylase (TH), the ratelimiting enzyme in catecholamine synthesis and a marker of DA neurons. This analysis revealed that nearly all (~98%) of VTA DA neurons express NtsR1 at some point in development (Fig. 14A, E), while about two thirds (~70%) of DA neurons actively co-express NtsR1 in adulthood (Fig. 14B, F). Interestingly, many TH-negative VTA neurons co-expressed NtsR1 in the developmental model (~30% of NtsR1-GFP+ neurons) but not the adult model (Fig. 14G, H). Thus during development NtsR1 is expressed in, and presumably regulates, both DA and other VTA neurons, but only mediates Nts actions in the adult brain via DA neurons. By contrast, the majority of the small population of NtsR2-GFP+ neurons in the developmental and adult models did not colocalize with TH (Fig. 14C, D, E, F) and NtsR2-GFP expression was found on only 6.5% of DA neurons in NtsR2Dev;GFP mice and <1% of DA neurons in NtsR2Adult;GFP mice (Fig. 14G, H). Taken together, these data support the hypothesis that NtsR1 is the dominant isoform regulating VTA DA neurons in both development and adulthood, and suggests that NtsR1 may also modulate non-DA neurons in the VTA during development. Projections of VTA NtsR1 Neurons: Given that NtsR1 was expressed on some, but not all VTA DA neurons in adult mice, we investigated whether NtsR1 defined the subset of VTA DA neurons that project to the NA or the prefrontal cortex (PFC)43, two major outputs of VTA DA neurons. To do this, we injected adult NtsR1∆NEO-Cre;GFP mice in the VTA with the Cremediated tract tracer Ad-syn-mCherry27 (Fig. 15A). In these brains, GFP identifies any cell that 111 expressed NtsR1 throughout lifespan, but only cells that actively express NtsR1 can undergo Cre-mediated recombination to express both GFP and the synaptophysin-mCherry fusion protein within cell bodies and terminals27. Thus, this method allows us to discriminate cells that only expressed NtsR1 during development from the adult cells that currently express NtsR1 and can mediate DA release. Visualization of the VTA revealed that nearly all mCherry+ neurons co-expressed GFP, however many GFP+ neurons within the injection site did not co-express mCherry (Fig. 15B, yellow arrows). These findings are consistent with our detection of almost twice as many NtsR1-GFP+ neurons in NtsR1Dev;GFP mice compared to NtsR1Adult;GFP mice (Fig. 12E, F) and confirms that NtsR1 expression, and hence induction of Cre, is confined to a limited set of VTA neurons in the adult brain. Out of 5 injected mice, 3 had mCherry-labeling confined to the VTA and were used for analysis (Fig. 15C). We also injected Ad-syn-mCherry into the VTA of NtsR2∆NEO-Cre;GFP mice to verify the minor population NtsR2 neurons in the adult brain and define their projections. In contrast to the numerous mCherry-labeled neurons observed in NtsR1∆NEO-Cre;GFP mice (Fig. 15D), examination of NtsR2∆NEO-Cre;GFP mice revealed mCherry+ cells with predominantly stellate morphology in the VTA. This small population of VTA NtsR2 neurons provided <10 single terminals observed throughout the entire brain, including a minor projection to the NA and IPAC (Fig. 15 E, white arrows). These data confirm the dearth of NtsR2-expressing neurons in the adult VTA (Fig. 12F) and support a predominant role for NtsR1 in directly modifying VTA DA signaling. Analysis of the projections from the well-targeted NtsR1∆NEO-Cre;GFP mice is summarized in Fig. 16M. VTA NtsR1 neurons most densely project to sub-regions of the ventral striatum, including the NA core (NAc), NA shell (NAsh), and olfactory tubercle (OFT) (Fig. 16A, B, C). By contrast, the lack of terminals in the PFC implicates VTA NtsR1 neurons in regulating mesolimbic, but not mesocortical, DA signaling (Fig. 16D)43. We also observed dense VTA NtsR1 terminals within the interstitial nucleus of the posterior limb of the anterior commissure 112 (IPAC) and modest terminals in the neighboring stria terminals (ST) (Fig. 16F). The caudate putamen (CPu) contained sparse VTA NtsR1 terminals (Fig. 16E), consistent with the VTA providing less input to the dorsal striatum compared to SN43,44. A few mCherry-labeled terminals were observed in the ventral pallidum (VP) (Fig. 16G) and the central amygdala (CeA) (Fig. 16H), where DA release is associated with emotional learning and feeding45-48. While we detected few terminals within the lateral hypothalamus (LHA), the dense patch of mCherry+ fibers in the neighboring nigrostriatal tract likely represents VTA NtsR1 axons traversing the brain en route to the striatum (Fig. 16J, see nigrostriatal tract (ns)). Intriguingly, the lateral habenula (LHb) also contained many mCherry-labeled terminals (Fig. 16I), although VTA neurons projecting to the LHb do not release DA and instead inhibit the LHb to indirectly promote reward49. We also observed low to medium terminal density in the hindbrain, namely in the laterodorsal tegmental nucleus (LDTg) (Fig. 16K), the parabrachial nucleus (PBN) (Fig. 16L), and the dorsal raphe (DR) (Fig. 16M). 3.4 Discussion Here, we used dual recombinase mice to identify NtsR1 and NtsR2 cells in the VTA to establish the precise cellular mediators of Nts action. We demonstrate that NtsR2 is predominately expressed on glial cells and only a small number of both DA and non-DAergic neurons throughout life. By contrast, NtsR1 is expressed on many VTA DA neurons during both development and adulthood. Furthermore, VTA NtsR1 neurons project to the ventral striatum, not the PFC, and hence are positioned to specifically modify mesolimbic DA signaling. Thus, our data are the first to demonstrate NtsR1 as the predominant receptor isoform by which Nts can directly engage DA neurons to modify DA-dependent signaling and physiology. Our findings provide cellular resolution that will be essential to understand how Nts mechanistically acts in the VTA. Consistent with our finding of NtsR1 expression on some VTA 113 DA neurons, the application of Nts or Nts analogues into the VTA activates DA neurons, increases NA DA release, restrains food intake, and promotes reward behaviors, and these effects are reduced by NtsR1 antagonists or developmental deletion of NtsR11,25,26,28,29. Similarly, Bose et al. suggested that Nts enhances excitatory input onto VTA DA neurons via NtsR150. However, not all studies using pharmacological agents or receptor null mice have supported a predominant role for NtsR1 in directly modulating VTA DA signaling. For example, Kempadoo et al. found that low doses of intra-VTA Nts increased excitatory transmission to DA neurons via NtsR1, while high doses of Nts reduced excitatory input through an NtsR1independent mechanism, implying involvement of NtsR229. Similarly, Rouibi et al. reported that Nts action in the VTA reduces excitatory input to DA neurons via an NtsR1-independent mechanism, again suggesting involvement of NtsR228. Our finding that <1% of adult VTA DA neurons co-express NtsR2 suggest it is unlikely that Nts directly modulates most DA neurons via NtsR2. One possible mechanism to resolve these discrepancies arises from our finding that NtsR2 is predominantly expressed on astrocytes, which might act via tripartite synapses to mediate local regulation of DA neurons51. Going forward, the ability to identify NtsR1 and NtsR2 cells will permit direct testing of how Nts engages these cells to orchestrate DA signaling. Previously, NtsR expression was characterized using in situ hybridization (ISH), which labels mRNA but does not always provide sufficient signal to visualize morphologic features of the cell in which the transcript is expressed. An advantage of the dual recombinase strategy we used to identify NtsR-expressing cells is that Cre-induced GFP fills the entire cell, which allowed us to clearly distinguish between neuronal and glial morphologies. We were thus able to discern that most NtsR2 is expressed by glial cells throughout the brain, consistent with the diffuse, low intensity signal observed via ISH for Ntsr29,19,21,52. Similarly, cultured astrocytes express NtsR220,41,42 and Nts modulates their activity53,54. Interestingly, because astrocytes also express endopeptidases that catabolize Nts55-57, they might act as a sink to internalize and 114 degrade residual Nts at the synapse via NtsR2. Loss of NtsR2 signaling, as in NtsR2-null mice, could therefore lead to excess synaptic Nts and enhanced Nts-NtsR1 activation of VTA DA neurons, which may explain their increased striatal DA levels and hyperactivity35. VTA DA neurons are heterogeneous and have been defined by their projection targets (see 58,59 for review). We found that VTA NtsR1 neurons comprise ~70% of VTA DA neurons and project primarily to the NA, OFT, and IPAC, but not to other efferent targets of VTA NA neurons such as the PFC or hippocampus43. Given that VTA NtsR1 neurons are mesolimbic, they are likely to modulate the reinforcing properties of natural and pharmacologic rewards, which depend upon DA release to the NA60-62 63,64. Furthermore, our current data provides a cellular and circuit mechanism to explain how Nts signaling directly via VTA NtsR1 neurons can increase NA DA release and mediate conditioned reward12,28,29,65-69. The functional implications of the VTA NtsR1 projections to the IPAC are less clear. The IPAC is structurally related to the amygdala70, and DA release in the amygdala has been implicated in emotional learning46-48 and in regulation of food intake45. Finally, we noted several VTA NtsR1 projections to the LHb, a structure that mediates aversion. However, TH+ neurons that project to the LHb do not release DA, but rather inhibit the LHb, which then disinhibits VTA DA neurons49. Thus, the TH+ VTA projections to the LHb, including NtsR1-expressing neurons, may indirectly support reward rather than aversion. Overall, VTA NtsR1 neurons project to regions that can modify reward intake, suggesting that Nts action via the VTA may be necessary to drive pursuit of pharmacological and natural rewards, such as food. A limitation of the developmental NtsR1∆NEO-Cre;GFP and NtsR2∆NEO-Cre;GFP models we developed is that GFP expression does not necessarily reflect cells with active NtsR1 or NtsR2 expression. Inducing Cre expression from the beginning of embryonic development in the developmental models leads to permanent GFP expression in all cells that expressed NtsR1 or 115 NtsR2 at any point in life. Comparing findings from developmental to adult models however, can inform how NtsR expression patterns change through life. For example, essentially all VTA DA neurons colocalize with GFP in NtsR1Dev;GFP mice whereas only ~70% colocalize in NtsR1Adult;GFP mice. These data imply that at some point in development ~30% of DA neurons transiently expressed NtsR1, but do not identify when they cease expressing NtsR1. Inducing FlpO expression at discrete time points across development will be required to define the temporal dynamics of NtsR1 expression. This could be accomplished by breeding mice to a tamoxifen-inducible FlpO deleter line, whereby injection of tamoxifen would induce Cre expression at any desired point in development71 to label cells actively expressing NtsR1. While NtsR1 is robustly expressed on VTA DA neurons throughout life, our NtsR1NEO-Cre; GFP models do not distinguish between neurons that express NtsR1 on the soma/dendrites vs. synaptic terminals. The subcellular localization of NtsR1 is functionally important because NtsNtsR1 signaling elicits different behavioral effects via either pre- or post-synaptic mechanisms. Nts-NtsR1 signaling on DA cell bodies in the VTA increases DA release and locomotor activity similar to psychostimulants12,65-67,72,73. By contrast, Nts-NtsR1 action in the NA suppresses locomotor activity induced by AMPH, cocaine, and DA itself, similar to antipsychotic medications74-76, and is thought to be mediated by NtsR1 expressed pre-synaptically on DA terminals in the striatum. Studies using immunohistochemistry and autoradiography suggest that NtsR1 binding sites are present on DA terminals in the ventral and dorsal striatum77-79. Furthermore, radiolabeled Nts injected in to the striatum accumulates in the cell bodies of VTA DA neurons80-82, indicating that Nts binds DA neurons at presynaptic terminals and is internalized. Collectively, VTA Nts-NtsR1 signaling appears to have opposing behavioral outcomes depending on whether it occurs pre- or post-synaptically, and this should be taken into account when interpreting future data from NtsR1NEO-Cre mice. 116 Our work indicates vast expression of NtsR1 throughout the brain in NtsR1Dev;GFP mice, and a heretofore unappreciated potential role for Nts to mediate development of the central nervous system. Within the VTA, NtsR1Dev;GFP mice showed nearly 100% of VTA DA neurons colocalizing with GFP, which represented ~65% of all GFP+ neurons. Curiously, this is similar to the overall distribution of DA and non-DA neurons in adults (~65% TH+, 35% TH-)58 and suggests that at some point in life, almost every VTA neuron, both DA and non-DAergic, expresses NtsR1. Nts-NtsR1 signaling may thus play a critical role in establishing VTA circuits during development. Mice null for NtsR1 may therefore suffer from abnormal formation of the VTA DA system that causes aberrant DA signaling, locomotor activity, body weight, anxiety, sleep, and exaggerated response to psychostimulants that has been reported to occur in this line1,2,27,35,36. Our data also identify a substantial population of VTA NtsR1 neurons that persist in the adult brain, and hence signify roles for Nts signaling via NtsR1 in regulating normal physiology. Going forward, using NtsR1NEO-Cre mice will permit selective manipulation of the VTA NtsR1 neurons in the adult brain to discern how they contribute to DA signaling, physiology and behavior. 3.5 Methods Generation of NtsR1NEO-Cre and NtsR2NEO-Cre Knock-In Mice: NtsR1NEO-Cre and NtsR2NEO-Cre targeting vectors were generated by inserting an IRES-Cre between the stop codon and the polyadenylation site of the sequence encoding the 3’ end of the mouse Ntsr1 gene, with an frt-flanked NEO cassette placed upstream of the IRES-Cre. The linearized targeting vector was electroporated into mouse R1 embryonic stem (ES) cells (129sv background) and cells were selected with G418. DNA from ES cell clones was analyzed via qPCR for loss of homozygosity using Taqman primer and probes for the genomic Ntsr1 or Ntsr2 insertion sites (NtsR1: Forward: TCTGATGTTGGACTTGGGTTC, Reverse: TCTGATGTTGGACTTGGGTTC, Probe: TCTGATGTTGGACTTGGGTTC. NtsR2: Forward: 117 ACCCATCAGATAAGCCATGC, Reverse: GTGGGAAGTTGAGGGCAG, Probe: GTCTAAGCGGACCTACTGACCCA). NGF was used as a copy number control83. Putative positive ES clones were expanded, confirmed for homologous recombination by Southern blot and injected into mouse C57BL/6 blastocysts to generate chimeras. Chimeric males were mated with C57BL/6 females (Jackson Laboratory), and germline transmission was determined initially via progeny coat color, then confirmed via conventional PCR for IRES-Cre. Breeding and Genotyping: Mice were bred and housed in a 12h light/12h dark cycle with ad libitum access to water and food (Harlan Teklad #7913). All procedures were approved by the Institutional Animal Care and Use Committee (IACUC) at Michigan State University. Heterozygous NtsR1NEO-Cre and NtsR2NEO-Cre mice were bred to either FlpO deleter mice (Jax Stock #012930), a Cre-inducible RosaeGFP-L10a reporter84, or C57/Bl6 wild-type mice to maintain the lines. Developmental Model: NtsR1NEO-Cre and NtsR2NEO-Cre mice were bred to a FlpO deleter line and progeny that inherited the FlpO allele (NtsR1∆NEO-Cre and NtsR2∆NEO-Cre) were subsequently mated to a Cre-inducible RosaeGFP-L10a reporter, generating NtsR1Dev;GFP and NtsR2Dev;GFP mice. Progeny that were heterozygous for both IRES-Cre and GFP alleles were used for analysis. Adult Model: NtsR1NEO-Cre and NtsR2NEO-Cre mice were mated directly to Creinducible RosaeGFP-L10a animals, producing heterozygous NtsR1NEO-Cre;GFP and NtsR2NEOCre ;GFP progeny. These animals were then injected with FlpO adenovirus in adulthood (see below) to generate NtsR1Adult;GFP and NtsR2Adult;GFP study mice. All animals were genotyped by standard PCR using the following primer sequences: IRES-Cre: Forward: GGACGTGGTTTTCCTTTGAA. Reverse: AGGCAAATTTTGGTGTACGG. Rosa26EGFP-L10a : Mutant forward: TCTACAAATGTGGTAGATCCAGGC, WT forward: GAGGGGAGTGTTGCAATACC, Common: CAGATGACTACCTATCCTCCC. FlpO: Mutant: GCGAAGAGTTTGTCCTCAACC, Common: GCG AAG AGT TTG TCC TCA ACC, Wild-type: GGAGCGGGAGAAATGGATATG. Adult male and female mice of each model were studied. 118 Surgery: Adult NtsR1Adult;GFP and NtsR2Adult;GFP mice received a pre-surgical injection of carprofen (5mg/kg s.c.) and were anesthetized with 3-4% isoflurane/O2 in an induction chamber before being placed in a stereotaxic frame (Kopf). Under 1-2% isoflurane, access holes were drilled in the skull allowing a guide cannula with stylet (PlasticsOne) to be lowered into the lateral ventricles (A/P: -0.22, M/L: +/- 1.0, D/V: -2.0). Mice were bilaterally injected with 1µL FlpO adenovirus (Vector Biolabs), which was infused at a rate of 1µL/minute. The animals recovered for 10 days prior to perfusion to permit sufficient time for FlpO-mediated excision of the frt-flanked NEO cassette and GFP expression. For tracing studies, NtsR1∆NEO-Cre;GFP mice were injected unilaterally in the VTA (A/P: -3.2, M/L: +/-0.48, D/V: -4.65) with 75-100nL of Adsyn-mCherry, an adenovirus expressing a Cre-dependent synaptophysin-mCherry fusion protein27 (provided by Martin Myers, University of Michigan). Mice recovered for 7-10 days after surgery to allow for Cre-mediated recombination and synaptophysin-mCherry expression at presynaptic terminals. Perfusion and Immunofluorescence: Mice were treated with a lethal dose of i.p. pentobarbital followed by transcardial perfusion with 10% neutral-buffered formalin (Fisher Scientific). Brains were removed, post-fixed in 10% formalin overnight at 4˚C, dehydrated with 30% sucrose in PBS for 2-3 days, and sectioned into 30 µm slices using a sliding microtome (Leica). Brain sections were then analyzed by immunofluorescence or immunohistochemistry as previously described27,37. For characterization of NtsR1 and NtsR2 expression, sections were exposed to chicken anti-GFP (1:2000, Abcam) and mouse anti-TH (1:1000, Millipore), followed by incubation with species-specific secondary antibodies conjugated to AlexaFluor 488 or 568 fluorophores (Life Technologies or Jackson ImmunoResearch). Brains were analyzed using an Olympus BX53 fluorescence microscope outfitted with FITC and Texas Red filters. Microscope images were collected using Cell Sens software and a Qi-Click 12 Bit cooled camera, and images were analyzed using Photoshop software (Adobe). Five representative 119 sections spanning the entire VTA were selected from each mouse and were used to determine the number and colocalization of GFP and/or TH positive neurons. Statistics. All data were analyzed in Prism 6 (GraphPad) using unpaired t-tests. Bar graphs represent mean ± SEM. 120 APPENDIX 121 Figure 12. Generation of mouse models to identify developmental vs. adult expression patterns of NtsR1 and NtsR2. 122 Figure 12 (cont’d). A) Schematic demonstrating how constitutive Cre expression is induced in NtsR1Neo-Cre mice, resulting in GFP labeling of any neuron that expresses NtsR1 throughout the lifespan (NtsR1Dev model). B) Schematic depiction of how Cre expression is suppressed until adulthood in NtsR1Neo-Cre mice, resulting in GFP labeling restricted to cells actively expressing NtsR1 in adult animals (NtsR1Adult model). *The same strategies depicted in A-B were used in NtsR2Neo-Cre mice. Figure 13. Developmental vs. adult expression of NtsR1 and NtsR2 in the VTA. NtsR1NEO-Cre and NtsR2NEO-Cre mice were bred to a Cre-inducible GFP reporter line and Cre expression was induced constitutively from conception (developmental expression) or only in adulthood. A-B) Developmental and adult distributions of NtsR1 neurons. C-D) Developmental and adult distributions of NtsR2 neurons. GFP+ cells have either neuronal (white arrows) or glial morphology. ip=interpeduncular nucleus, ml=medial lemniscus. Scale bars represent 200 µm. Insets are digital enlargements of the areas within dashed boxes. E, F) Total number of GFP+ NtsR1 or NtsR2 neurons in developmental and adult models in the VTA (NtsR1Dev n=3, NtsR2Dev n=3, NtsR1Adult n=3, NtsR2Adult n=6). Each bar represents mean ± SEM and data were analyzed by unpaired t-tests. 123 Figure 14. NtsR1 and NtsR2 colocalization with TH in development compared to adulthood. 124 Figure 14 (cont’d). Representative images showing TH co-expression in the VTA with A) developmental NtsR1-GFP+ neurons, B) adult NtsR1-GFP+ neurons, C) developmental NtsR2GFP+ neurons D) adult NtsR2-GFP+ neurons. Yellow arrows = TH/GFP+ colocalized neurons, green arrows = GFP+ only neurons, pink arrows = TH+ only neurons, green asterisks = GFP+ glia. Scale bars=100µm. ml=medial lemniscus, ip=interpeduncular nucleus. Insets are digital enlargements of the indicated boxed areas. Percentage of NtsR1 and NtsR2 neurons that colocalize with TH in E) developmental and F) adult expression models. Percentage of all VTA TH+ neurons that colocalize with NtsR1 or NtsR2 in G) developmental and H) adult models. (NtsR1Dev n=3; NtsR2Dev n=3; NtsR1Adult n=3; NtsR2Adult n=6). Bars represent mean ± SEM and data were analyzed by unpaired t-tests. 125 Figure 15. Ad-syn-mCherry reveals projections of VTA NtsR1 neurons. A) Schematic showing NtsR1∆NEO-Cre;GFP mice injected in the VTA with Ad-syn-mCherry, which labels VTA NtsR1 projections by expression of a Cre-inducible synaptophysin-mCherry fusion protein. B) VTA of NtsR1∆NEO-Cre;GFP mouse injected with Ad-syn-mCherry showing GFP+ neurons that do not co-express mCherry (yellow arrows), representing neurons that transiently expressed NtsR1 during development. C) Distribution of mCherry+ cell bodies within the VTA from 3 individual NtsR1∆NEO-Cre;GFP mice, confirming selective labeling of VTA NtsR1 neurons. Comparison of VTA injection site and mCherry labeled terminals in the NAc and IPAC of D) NtsR1∆NEO-Cre;GFP mice vs. E) NtsR2∆NEO-Cre;GFP mice. Scale bars represent 100µm. ml=medial lemniscus, ip=interpeduncular nucleus, aca=anterior commissure. 126 Figure 16. Projections of VTA NtsR1 neurons. 127 Figure 16 (cont’d). Ad-syn-mCherry-labeled terminals in the A) nucleus accumbens core (NAc), B) nucleus accumbens shell (NAsh), C) olfactory tubercle (OFT), D) absence of terminals in the prefrontal cortex (PFC), E) caudate/putamen (CPu), F) IPAC and stria terminals (ST), G) ventral pallidum (VP), H) central amygdala (CeA), I) lateral habenula (LHb), J) lateral hypothalamic area (LHA), K) laterodorsal tegmentum (LDTg), L) parabrachial nucleus (PBN). M) Schematic summarizing projections of VTA NtsR1 neurons (purple stars). Scale bars represent 100µm. aca=anterior commissure, cc=corpus callous, BLA=basolateral amygdala, 3V= third ventricle, ns= nigrostriatal tract, f=fornix, 4V=fourth ventricle, scp=superior cerebellar peduncle. 128 REFERENCES 129 REFERENCES 1 Remaury, A. et al. Targeted inactivation of the neurotensin type 1 receptor reveals its role in body temperature control and feeding behavior but not in analgesia. Brain research 953, 63-72 (2002). 2 Fitzpatrick, K. et al. Altered sleep and affect in the neurotensin receptor 1 knockout mouse. Sleep 35, 949-956, doi:10.5665/sleep.1958 (2012). 3 Kleczkowska, P. & Lipkowski, A. W. Neurotensin and neurotensin receptors: characteristic, structure-activity relationship and pain modulation--a review. European journal of pharmacology 716, 54-60, doi:10.1016/j.ejphar.2013.03.004 (2013). 4 Bissette, G., Luttinger, D., Mason, G. A., Hernandez, D. E. & Loosen, P. T. Neurotensin and thermoregulation. Annals of the New York Academy of Sciences 400, 268-282 (1982). 5 Boules, M. M., Fredrickson, P., Muehlmann, A. M. & Richelson, E. Elucidating the role of neurotensin in the pathophysiology and management of major mental disorders. Behavioral sciences (Basel, Switzerland) 4, 125-153, doi:10.3390/bs4020125 (2014). 6 Caceda, R., Kinkead, B. & Nemeroff, C. B. Neurotensin: role in psychiatric and neurological diseases. Peptides 27, 2385-2404, doi:10.1016/j.peptides.2006.04.024 (2006). 7 Tanaka, K., Masu, M. & Nakanishi, S. Structure and functional expression of the cloned rat neurotensin receptor. Neuron 4, 847-854 (1990). 8 Vita, N. et al. Cloning and expression of a complementary DNA encoding a high affinity human neurotensin receptor. FEBS letters 317, 139-142 (1993). 9 Mazella, J. et al. Structure, functional expression, and cerebral localization of the levocabastine-sensitive neurotensin/neuromedin N receptor from mouse brain. The Journal of neuroscience : the official journal of the Society for Neuroscience 16, 56135620 (1996). 10 Chalon, P. et al. Molecular cloning of a levocabastine-sensitive neurotensin binding site. FEBS letters 386, 91-94 (1996). 11 Mazella, J. et al. The 100-kDa neurotensin receptor is gp95/sortilin, a non-G-proteincoupled receptor. The Journal of biological chemistry 273, 26273-26276 (1998). 12 St-Gelais, F., Legault, M., Bourque, M. J., Rompre, P. P. & Trudeau, L. E. Role of calcium in neurotensin-evoked enhancement in firing in mesencephalic dopamine neurons. The Journal of neuroscience : the official journal of the Society for Neuroscience 24, 2566-2574, doi:10.1523/jneurosci.5376-03.2004 (2004). 130 13 Vita, N. et al. Neurotensin is an antagonist of the human neurotensin NT2 receptor expressed in Chinese hamster ovary cells. European journal of pharmacology 360, 265272 (1998). 14 Alexander, M. J. & Leeman, S. E. Widespread expression in adult rat forebrain of mRNA encoding high-affinity neurotensin receptor. The Journal of comparative neurology 402, 475-500 (1998). 15 Elde, R., Schalling, M., Ceccatelli, S., Nakanishi, S. & Hokfelt, T. Localization of neuropeptide receptor mRNA in rat brain: initial observations using probes for neurotensin and substance P receptors. Neuroscience letters 120, 134-138 (1990). 16 Nicot, A., Rostene, W. & Berod, A. Neurotensin receptor expression in the rat forebrain and midbrain: a combined analysis by in situ hybridization and receptor autoradiography. The Journal of comparative neurology 341, 407-419, doi:10.1002/cne.903410310 (1994). 17 Lein, E. S. et al. Genome-wide atlas of gene expression in the adult mouse brain. Nature 445, 168-176, doi:10.1038/nature05453 (2007). 18 Sarret, P., Perron, A., Stroh, T. & Beaudet, A. Immunohistochemical distribution of NTS2 neurotensin receptors in the rat central nervous system. The Journal of comparative neurology 461, 520-538, doi:10.1002/cne.10718 (2003). 19 Walker, N. et al. Tissue distribution and cellular localization of the levocabastinesensitive neurotensin receptor mRNA in adult rat brain. Brain research. Molecular brain research 57, 193-200 (1998). 20 Nouel, D. et al. Differential binding profile and internalization process of neurotensin via neuronal and glial receptors. The Journal of neuroscience : the official journal of the Society for Neuroscience 17, 1795-1803 (1997). 21 Sarret, P., Beaudet, A., Vincent, J. P. & Mazella, J. Regional and cellular distribution of low affinity neurotensin receptor mRNA in adult and developing mouse brain. The Journal of comparative neurology 394, 344-356 (1998). 22 Lepee-Lorgeoux, I., Betancur, C., Rostene, W. & Pelaprat, D. Differential ontogenetic patterns of levocabastine-sensitive neurotensin NT2 receptors and of NT1 receptors in the rat brain revealed by in situ hybridization. Brain research. Developmental brain research 113, 115-131 (1999). 23 Palacios, J. M., Pazos, A., Dietl, M. M., Schlumpf, M. & Lichtensteiger, W. The ontogeny of brain neurotensin receptors studied by autoradiography. Neuroscience 25, 307-317 (1988). 24 Pettibone, D. J. et al. The effects of deleting the mouse neurotensin receptor NTR1 on central and peripheral responses to neurotensin. The Journal of pharmacology and experimental therapeutics 300, 305-313 (2002). 131 25 Leonetti, M. et al. Specific involvement of neurotensin type 1 receptor in the neurotensinmediated in vivo dopamine efflux using knock-out mice. Journal of neurochemistry 89, 16, doi:10.1046/j.1471-4159.2003.02231.x (2004). 26 Kim, E. R., Leckstrom, A. & Mizuno, T. M. Impaired anorectic effect of leptin in neurotensin receptor 1-deficient mice. Behavioural brain research 194, 66-71, doi:10.1016/j.bbr.2008.06.024 (2008). 27 Opland, D. et al. Loss of neurotensin receptor-1 disrupts the control of the mesolimbic dopamine system by leptin and promotes hedonic feeding and obesity. Molecular metabolism 2, 423-434, doi:10.1016/j.molmet.2013.07.008 (2013). 28 Rouibi, K., Bose, P., Rompre, P. P. & Warren, R. A. Ventral Midbrain NTS1 Receptors Mediate Conditioned Reward Induced by the Neurotensin Analog, D-Tyr[11]neurotensin. Frontiers in neuroscience 9, 470, doi:10.3389/fnins.2015.00470 (2015). 29 Kempadoo, K. A. et al. Hypothalamic neurotensin projections promote reward by enhancing glutamate transmission in the VTA. The Journal of neuroscience : the official journal of the Society for Neuroscience 33, 7618-7626, doi:10.1523/jneurosci.258812.2013 (2013). 30 Maeno, H. et al. Comparison of mice deficient in the high- or low-affinity neurotensin receptors, Ntsr1 or Ntsr2, reveals a novel function for Ntsr2 in thermal nociception. Brain research 998, 122-129 (2004). 31 Roussy, G. et al. Altered morphine-induced analgesia in neurotensin type 1 receptor null mice. Neuroscience 170, 1286-1294, doi:10.1016/j.neuroscience.2010.08.016 (2010). 32 Botto, J. M., Guillemare, E., Vincent, J. P. & Mazella, J. Effects of SR 48692 on neurotensin-induced calcium-activated chloride currents in the Xenopus oocyte expression system: agonist-like activity on the levocabastine-sensitive neurotensin receptor and absence of antagonist effect on the levocabastine insensitive neurotensin receptor. Neuroscience letters 223, 193-196 (1997). 33 Yamada, M., Yamada, M., Lombet, A., Forgez, P. & Rostene, W. Distinct functional characteristics of levocabastine sensitive rat neurotensin NT2 receptor expressed in Chinese hamster ovary cells. Life sciences 62, PL 375-380 (1998). 34 Thomas, J. B. et al. Identification of 2-(306amino)tricyc lo[3.3.1.13,7]decane-2-carboxylic Acid (NTRC-844) as a Selective Antagonist for the Rat Neurotensin Receptor Type 2. ACS chemical neuroscience 7, 1225-1231, doi:10.1021/acschemneuro.6b00097 (2016). 35 Liang, Y. et al. Hyperactivity of the dopaminergic system in NTS1 and NTS2 null mice. Neuropharmacology 58, 1199-1205, doi:10.1016/j.neuropharm.2010.02.015 (2010). 36 Li, Z., Liang, Y., Boules, M., Gordillo, A. & Richelson, E. Effect of amphetamine on extracellular concentrations of amino acids in striatum in neurotensin subtype 1 and 2 receptor null mice: a possible interaction between neurotensin receptors and amino acid systems for study of schizophrenia. Neuropharmacology 58, 1174-1178, doi:10.1016/j.neuropharm.2010.02.016 (2010). 132 37 Leinninger, G. M. et al. Leptin action via neurotensin neurons controls orexin, the mesolimbic dopamine system and energy balance. Cell metabolism 14, 313-323, doi:10.1016/j.cmet.2011.06.016 (2011). 38 Vong, L. et al. Leptin action on GABAergic neurons prevents obesity and reduces inhibitory tone to POMC neurons. Neuron 71, 142-154, doi:10.1016/j.neuron.2011.05.028 (2011). 39 Krashes, M. J. et al. Rapid, reversible activation of AgRP neurons drives feeding behavior in mice. The Journal of clinical investigation 121, 1424-1428, doi:10.1172/jci46229 (2011). 40 Geisler, S., Berod, A., Zahm, D. S. & Rostene, W. Brain neurotensin, psychostimulants, and stress--emphasis on neuroanatomical substrates. Peptides 27, 2364-2384, doi:10.1016/j.peptides.2006.03.037 (2006). 41 Nouel, D., Sarret, P., Vincent, J. P., Mazella, J. & Beaudet, A. Pharmacological, molecular and functional characterization of glial neurotensin receptors. Neuroscience 94, 1189-1197 (1999). 42 Yamauchi, R. et al. Neurotensin type 2 receptor is involved in fear memory in mice. Journal of neurochemistry 102, 1669-1676, doi:10.1111/j.1471-4159.2007.04805.x (2007). 43 Swanson, L. W. The projections of the ventral tegmental area and adjacent regions: a combined fluorescent retrograde tracer and immunofluorescence study in the rat. Brain research bulletin 9, 321-353 (1982). 44 Bentivoglio, M., van der Kooy, D. & Kuypers, H. G. The organization of the efferent projections of the substantia nigra in the rat. A retrograde fluorescent double labeling study. Brain research 174, 1-17 (1979). 45 Anderberg, R. H., Anefors, C., Bergquist, F., Nissbrandt, H. & Skibicka, K. P. Dopamine signaling in the amygdala, increased by food ingestion and GLP-1, regulates feeding behavior. Physiology & behavior 136, 135-144, doi:10.1016/j.physbeh.2014.02.026 (2014). 46 Darvas, M., Fadok, J. P. & Palmiter, R. D. Requirement of dopamine signaling in the amygdala and striatum for learning and maintenance of a conditioned avoidance response. Learning & memory (Cold Spring Harbor, N.Y.) 18, 136-143, doi:10.1101/lm.2041211 (2011). 47 Rosenkranz, J. A. & Grace, A. A. Dopamine-mediated modulation of odour-evoked amygdala potentials during pavlovian conditioning. Nature 417, 282-287, doi:10.1038/417282a (2002). 48 Guarraci, F. A., Frohardt, R. J., Young, S. L. & Kapp, B. S. A functional role for dopamine transmission in the amygdala during conditioned fear. Annals of the New York Academy of Sciences 877, 732-736 (1999). 133 49 Stamatakis, A. M. et al. A unique population of ventral tegmental area neurons inhibits the lateral habenula to promote reward. Neuron 80, 1039-1053, doi:10.1016/j.neuron.2013.08.023 (2013). 50 Bose, P., Rompre, P. P. & Warren, R. A. Neurotensin enhances glutamatergic EPSCs in VTA neurons by acting on different neurotensin receptors. Peptides 73, 43-50, doi:10.1016/j.peptides.2015.08.008 (2015). 51 Perea, G., Navarrete, M. & Araque, A. Tripartite synapses: astrocytes process and control synaptic information. Trends in neurosciences 32, 421-431, doi:10.1016/j.tins.2009.05.001 (2009). 52 Asselin, M. L., Dubuc, I., Coquerel, A. & Costentin, J. Localization of neurotensin NTS2 receptors in rat brain, using. Neuroreport 12, 1087-1091 (2001). 53 Hosli, E., Stauffer, S. & Hosli, L. Autoradiographic and electrophysiological evidence for the existence of neurotensin receptors on cultured astrocytes. Neuroscience 66, 627633 (1995). 54 Trudeau, L. E. Neurotensin regulates intracellular calcium in ventral tegmental area astrocytes: evidence for the involvement of multiple receptors. Neuroscience 97, 293302 (2000). 55 Woulfe, J., Checler, F. & Beaudet, A. Light and Electron Microscopic Localization of the Neutral Metalloendopeptidase EC 3.4.24.16 in the Mesencephalon of the Rat. The European journal of neuroscience 4, 1309-1319 (1992). 56 Mentlein, R. & Dahms, P. Endopeptidases 24.16 and 24.15 are responsible for the degradation of somatostatin, neurotensin, and other neuropeptides by cultivated rat cortical astrocytes. Journal of neurochemistry 62, 27-36 (1994). 57 Vincent, B., Vincent, J. P. & Checler, F. Neurotensin and neuromedin N undergo distinct catabolic processes in murine astrocytes and primary cultured neurons. European journal of biochemistry 221, 297-306 (1994). 58 Morales, M. & Margolis, E. B. Ventral tegmental area: cellular heterogeneity, connectivity and behaviour. Nature reviews. Neuroscience 18, 73-85, doi:10.1038/nrn.2016.165 (2017). 59 Juarez, B. & Han, M. H. Diversity of Dopaminergic Neural Circuits in Response to Drug Exposure. Neuropsychopharmacology : official publication of the American College of Neuropsychopharmacology 41, 2424-2446, doi:10.1038/npp.2016.32 (2016). 60 Hart, A. S., Rutledge, R. B., Glimcher, P. W. & Phillips, P. E. Phasic dopamine release in the rat nucleus accumbens symmetrically encodes a reward prediction error term. The Journal of neuroscience : the official journal of the Society for Neuroscience 34, 698704, doi:10.1523/jneurosci.2489-13.2014 (2014). 61 Fiorillo, C. D., Tobler, P. N. & Schultz, W. Discrete coding of reward probability and uncertainty by dopamine neurons. Science (New York, N.Y.) 299, 1898-1902, doi:10.1126/science.1077349 (2003). 134 62 Rutledge, R. B., Dean, M., Caplin, A. & Glimcher, P. W. Testing the reward prediction error hypothesis with an axiomatic model. The Journal of neuroscience : the official journal of the Society for Neuroscience 30, 13525-13536, doi:10.1523/jneurosci.174710.2010 (2010). 63 Lammel, S., Ion, D. I., Roeper, J. & Malenka, R. C. Projection-specific modulation of dopamine neuron synapses by aversive and rewarding stimuli. Neuron 70, 855-862, doi:10.1016/j.neuron.2011.03.025 (2011). 64 Lammel, S. et al. Input-specific control of reward and aversion in the ventral tegmental area. Nature 491, 212-217, doi:10.1038/nature11527 (2012). 65 Kalivas, P. W., Burgess, S. K., Nemeroff, C. B. & Prange, A. J., Jr. Behavioral and neurochemical effects of neurotensin microinjection into the ventral tegmental area of the rat. Neuroscience 8, 495-505 (1983). 66 Legault, M., Congar, P., Michel, F. J. & Trudeau, L. E. Presynaptic action of neurotensin on cultured ventral tegmental area dopaminergic neurones. Neuroscience 111, 177-187 (2002). 67 Sotty, F. et al. Comparative effects of neurotensin, neurotensin(8-13) and [DTyr(11)]neurotensin applied into the ventral tegmental area on extracellular dopamine in the rat prefrontal cortex and nucleus accumbens. Neuroscience 98, 485-492 (2000). 68 Glimcher, P. W., Giovino, A. A. & Hoebel, B. G. Neurotensin self-injection in the ventral tegmental area. Brain research 403, 147-150 (1987). 69 Glimcher, P. W., Margolin, D. H., Giovino, A. A. & Hoebel, B. G. Neurotensin: a new 'reward peptide'. Brain research 291, 119-124 (1984). 70 Gartner, U., Hartig, W., Riedel, A., Brauer, K. & Arendt, T. Immunocytochemical evidence for the striatal nature of the rat lateral part of interstitial nucleus of the posterior limb of the anterior commissure (IPAC). Journal of chemical neuroanatomy 24, 117-125 (2002). 71 Lao, Z., Raju, G. P., Bai, C. B. & Joyner, A. L. MASTR: a technique for mosaic mutant analysis with spatial and temporal control of recombination using conditional floxed alleles in mice. Cell reports 2, 386-396, doi:10.1016/j.celrep.2012.07.004 (2012). 72 Kalivas, P. W. & Duffy, P. Effect of acute and daily neurotensin and enkephalin treatments on extracellular dopamine in the nucleus accumbens. The Journal of neuroscience : the official journal of the Society for Neuroscience 10, 2940-2949 (1990). 73 Steinberg, R. et al. SR 48692, a non-peptide neurotensin receptor antagonist differentially affects neurotensin-induced behaviour and changes in dopaminergic transmission. Neuroscience 59, 921-929 (1994). 74 Kalivas, P. W., Nemeroff, C. B. & Prange, A. J., Jr. Neurotensin microinjection into the nucleus accumbens antagonizes dopamine-induced increase in locomotion and rearing. Neuroscience 11, 919-930 (1984). 135 75 Ervin, G. N., Birkemo, L. S., Nemeroff, C. B. & Prange, A. J., Jr. Neurotensin blocks certain amphetamine-induced behaviours. Nature 291, 73-76 (1981). 76 Robledo, P., Maldonado, R. & Koob, G. F. Neurotensin injected into the nucleus accumbens blocks the psychostimulant effects of cocaine but does not attenuate cocaine self-administration in the rat. Brain research 622, 105-112 (1993). 77 Quirion, R., Chiueh, C. C., Everist, H. D. & Pert, A. Comparative localization of neurotensin receptors on nigrostriatal and mesolimbic dopaminergic terminals. Brain research 327, 385-389 (1985). 78 Dilts, R. P. & Kalivas, P. W. Autoradiographic localization of mu-opioid and neurotensin receptors within the mesolimbic dopamine system. Brain research 488, 311-327 (1989). 79 Schotte, A. & Leysen, J. E. Autoradiographic evidence for the localization of high affinity neurotensin binding sites on dopaminergic nerve terminals in the nigrostriatal and mesolimbic pathways in rat brain. Journal of chemical neuroanatomy 2, 253-257 (1989). 80 Castel, M. N., Woulfe, J., Wang, X., Laduron, P. M. & Beaudet, A. Light and electron microscopic localization of retrogradely transported neurotensin in rat nigrostriatal dopaminergic neurons. Neuroscience 50, 269-282 (1992). 81 Castel, M. N., Malgouris, C., Blanchard, J. C. & Laduron, P. M. Retrograde axonal transport of neurotensin in the dopaminergic nigrostriatal pathway in the rat. Neuroscience 36, 425-430 (1990). 82 Faure, M. P., Nouel, D. & Beaudet, A. Axonal and dendritic transport of internalized neurotensin in rat mesostriatal dopaminergic neurons. Neuroscience 68, 519-529 (1995). 83 Soliman, G. A. et al. A simple qPCR-based method to detect correct insertion of homologous targeting vectors in murine ES cells. Transgenic research 16, 665-670, doi:10.1007/s11248-007-9110-2 (2007). 84 Krashes, M. J. et al. An excitatory paraventricular nucleus to AgRP neuron circuit that drives hunger. Nature 507, 238-242, doi:10.1038/nature12956 (2014). 136 Chapter 4. Neurotensin Receptor-1 Identifies a Subset of Ventral Tegmental Dopamine Neurons that Coordinate Energy Balance Authors: Hillary L. Woodworth, Hannah M. Batchelor, Bethany G. Beekly, Raluca Bugescu, Juliette A. Brown, Gizem Kurt, Patrick M. Fuller, and Gina M. Leinninger This chapter is a modified version of a manuscript accepted to Cell Reports. 4.1 Abstract Dopamine (DA) neurons in the ventral tegmental area (VTA) are heterogeneous and differentially direct ingestive and locomotor behaviors that impact energy balance. Identification of which VTA DA neurons mediate behaviors that limit weight gain has been hindered, however, by the lack of molecular markers to distinguish VTA DA populations. Here, we identified a specific subset of VTA DA neurons that express neurotensin receptor-1 (NtsR1) and project to the nucleus accumbens (NA). Genetically targeted ablation of VTA NtsR1 neurons uncouples motivated feeding and physical activity, biasing behavior toward energy expenditure and protecting mice from age-related and diet-induced weight gain. VTA NtsR1 neurons thus represent the first molecularly defined subset of mesolimbic DA neurons that are essential for the coordination of energy balance. Modulation of VTA NtsR1 neurons may therefore be useful to promote behaviors that prevent the development of obesity. 4.2 Introduction Obesity is caused by overconsumption of calorie-dense food combined with insufficient physical activity, and predisposes individuals to myriad chronic conditions that shorten life span. Proper diet and exercise are important for maintaining healthy weight but these lifestyle modifications are difficult to maintain long-term. Consequently, most individuals that lose weight 137 through changes in diet and activity tend to regain it over time1. Therapeutic strategies to suppress feeding and encourage physical activity would therefore be helpful to prevent weight gain and combat the obesity epidemic. The motivated feeding and locomotor behaviors that influence body weight are regulated, in part, by dopamine (DA) neurons in the ventral tegmental area (VTA) that release DA within the nucleus accumbens (NA) and prefrontal cortex (PFC)2,3. DA itself is essential for energy balance, demonstrated by the aphagia, hypolocomotion, reduced body weight and early lethality of mice that genetically lack DA4. Disruptions in DA signaling are also observed in obese rodents and humans, suggesting that inappropriate regulation of DA circuits contributes to weight gain5-7. Yet, the mechanisms by which VTA DA neurons orchestrate energy balance remain poorly understood. One emerging theory is that VTA DA neurons are not homogenous, but in fact consist of subsets of neurons that coordinate distinct aspects of feeding and locomotor activity. Indeed, VTA DA neurons can be differentiated according to their anatomical inputs and projections8,9 or their electrophysiological firing properties10. VTA DA neurons projecting to the NA are primarily activated by “rewarding” stimuli, whereas DA neurons projecting to the PFC are regulated by “aversive” cues8. VTA DA neurons may also contain the classical neurotransmitters GABA or glutamate11,12, whose co-release with DA at distinct projection sites may either promote or suppress feeding13. Collectively, these data suggest the interesting, and as yet untested possibility that distinct VTA DA circuits promote or suppress obesogenic behaviors. Unfortunately, neither projection output nor transmitter content identifies which specific VTA DA neurons may be leveraged to treat energy balance disorders. Certainly, molecular markers to differentiate functionally distinct subsets of VTA DA neurons would be useful in this regard, but to date no such markers have been identified. 138 Neuropeptides such as orexin, corticotropin releasing factor (CRF), glucagon-like peptide-1 (GLP-1), and neurotensin (Nts) differentially modify DA-dependent behaviors14-17, thus we reasoned that VTA DA neurons might be distinguished by their expression of neuropeptide receptors. Of these systems, Nts holds particular promise for modifying obesogenic behaviors, since pharmacologic administration of Nts in the VTA restrains feeding, increases locomotor activity and induces weight loss in obese rodents18-20. Nts signals via neurotensin receptors -1 and -2 (NtsR1 and NtsR2) that are expressed in the VTA21,22, and Nts enhances the activation of VTA DA neurons and DA release in the NA14,23. The lack of reagents to identify cells expressing NtsRs, however, has hampered a more detailed understanding of how Nts mechanistically engages DA neurons. To overcome this challenge, we used Cre-lox technology to selectively visualize and manipulate NtsR1 or NtsR2-expressing cells in mice and in doing so identified a unique subset of VTA DA neurons that express NtsR1. Furthermore, we found that disrupting VTA NtsR1 neurons alters DA-mediated behaviors to prevent weight gain, suggesting that this subset of VTA neurons may represent a novel cellular target for preventing the development of obesity. 4.3 Results NtsR1 and NtsR2 Identify Distinct VTA Cell Types: We crossed existing transgenic NtsR1Cre mice and newly generated knock-in NtsR2Cre mice onto a Cre-inducible-GFP reporter line, such that progeny expressed GFP selectively in NtsR1 or -2 expressing cells, respectively (Fig. 17). These models revealed striking differences in the morphology, number, and distribution of NtsR1-GFP and NtsR2-GFP cells throughout the rostrocaudal axis of the VTA (Fig. 18A). While a few NtsR2-GFP cells appeared to be neurons (Fig. 18A, white arrows), the vast majority were stellate-shaped cells consistent with glial morphology (Fig. 18A, yellow arrows). By contrast, NtsR1-GFP cells were exclusively neuronal (Fig. 18A) and more numerous than NtsR2-GFP neurons. The NtsR1-GFP neurons were predominantly located in 139 the rostral VTA (bregma-3.40 to -3.05), while the few NtsR2-GFP neurons clustered along the ventral-medial border of the caudal VTA (bregma -3.64), a region enriched in GABA-ergic neurons24. Given the distinct morphology and distribution of NtsR1-GFP and NtsR2-GFP cells, we next used immunolabeling to determine if they represent distinct cellular populations. The NtsR2-GFP cells with stellate morphology co-localized with the astrocyte marker S100, confirming that most NtsR2 cells are glia (Fig. 18B, yellow arrows). By contrast, none of the NtsR1-GFP cells or the few NtsR2-GFP cells with neuronal morphology co-localized with S100 (Fig. 18B, C, white arrows). Instead >90% of NtsR1-GFP neurons co-expressed tyrosine hydroxylase (TH), a marker of DA neurons (Fig. 18E-I), consistent with our previous findings25. Only ~30% of the much smaller population of NtsR2-GFP neurons co-expressed TH, representing ≤1% of all VTA DAergic neurons (Fig. 18D, F-I). Collectively, these data reveal that NtsR1 and NtsR2 are expressed in non-overlapping VTA subpopulations, that NtsR1 is the predominant receptor isoform expressed on VTA neurons, and that VTA NtsR1 neurons are a subset of all VTA DA neurons. VTA NtsR1 Neurons are a Projection-Defined Subset of VTA DA Neurons: We next investigated whether the novel subset of VTA NtsR1 neurons are a projection-specific population of VTA DA neurons by injecting the Cre-mediated anterograde tract tracer Ad-SynmCherry25 into the VTA of transgenic NtsR1Cre mice (Fig. 19A, B). This method revealed that VTA NtsR1 neurons densely innervate the ventral striatum, particularly the NA core and shell (NAc and NAs) and the nearby olfactory tubercle (OFT) (Fig. 19C, E). The second most densely innervated structure was the interstitial nucleus of the posterior limb of the anterior commissure (IPAC), which adjoins the NA (Fig. 19C, E). Moderate terminal density was observed in the ventral caudate/putamen (CPu) just adjacent to the NA (Fig. 19C, E), as well as 140 sparse projections within the caudal CPu and basolateral amygdala (BLA). By contrast, no projections were found in the central amygdala (CeA), hippocampus (Hipp) or the PFC (Fig. 19C-E). Thus, NtsR1 defines a subset of VTA DA neurons that regulate the NA and contiguous subregions, and can be used as a molecular marker to parse mesolimbic vs. mesocortical populations of VTA DA neurons. Loss of VTA NtsR1 Neurons Alters Energy Balance to Promote Leanness: To determine if VTA NtsR1 neurons contribute to the control of energy balance, we specifically ablated VTA NtsR1 neurons in adult mice (Fig. 20A). Adult NtsR1Cre;GFP mice received bilateral injections of an adeno-associated virus (AAV) expressing Cre-dependent Diphtheria Toxin A subunit (AAVDTA) into the VTA, and DTA expression selectively kills VTA NtsR1 neurons (NtsR1DTA mice). Surrounding cells remain intact since DTA is not transmitted outside of Cre-expressing neurons26. Separate NtsR1Cre;GFP mice received bilateral VTA injections of AAV-GFP and retained intact NtsR1 neurons (NtsR1GFP controls). We verified that AAV-DTA sufficiently depleted VTA NtsR1 neurons as early as two weeks post-surgery, but the remaining TH+ DA neurons confirmed that this method does not ablate DA neurons that do not express NtsR1 (Fig. 20B). We therefore compared NtsR1DTA mice and controls to reveal the role of VTA NtsR1 neurons in energy balance. NtsR1DTA mice exhibited significantly reduced body weight compared to NtsR1GFP controls by 4 wk post-surgery and remained leaner (Fig. 20C) with decreased body fat percentage (Fig. 20D). Surprisingly, NtsR1DTA mice consumed increasingly more chow (Fig. 20E) and drank more water over the study compared to NtsR1GFP controls (Fig. 21A), suggesting that loss of VTA NtsR1 neurons promotes ingestive behavior. While the increased energy intake of NtsR1DTA mice was surprising given their decreased body weight (Fig. 20G, I), we wondered if augmented energy expenditure might explain why these mice remain lean. To this end, we noted hyperactivity of NtsR1DTA mice in their home cages and metabolic analysis confirmed increased locomotor activity compared to NtsR1GFP controls (Fig. 141 20F). Further, when energy intake was plotted against energy expenditure and the distributions were tested for differences using linear regression, we found a significantly increased yintercept in the NtsR1DTA group at 4 but not 16 weeks (Fig. 20S, T). This indicates that at a given caloric intake, NtsR1DTA mice use more energy than controls and thus may not be able to appropriately adjust caloric intake to support their elevated activity at least 4 weeks after ablation, which likely contributes to their reduced body weight and adiposity. Lack of differences between groups at 16 weeks (Fig. 20T) suggests that compensatory feeding eventually occurs with time, but is not sufficient to match the normal body weight and adiposity levels of controls (Fig. 20C, D). Increased locomotion generally demands a commensurate increase in respiration to support the activity, and indeed the hyperlocomotive NtsR1DTA mice had increased VO2 and VCO2 (Fig. 21C-G) and exhibited increasingly higher rates of energy expenditure (adjusted for weight) over the course of the study compared to controls (Fig. 20H, J). The respiratory exchange ratio (RER) was unaffected (Fig. 21E-H) indicating that loss of VTA NtsR1 neurons does not alter energy substrate usage. Collectively, these data indicate that loss of VTA NtsR1 neurons biases energy balance toward increased physical activity and energy expenditure to promote leanness. Given the lean phenotype of mice lacking VTA NtsR1 on a normal chow diet, we hypothesized that they would be protected from diet-induced obesity. NtsR1DTA mice gained significantly less weight on high fat (HF) diet and were completely protected from diet-induced increase in body fat percentage compared to control mice (Fig. 20K, L). Although NtsR1DTA mice consumed more HF diet (Fig. 20M) and had higher weight-adjusted energy intake than NtsR1GFP controls (Fig. 20O, Q), they also exhibited escalating increases in locomotor activity (Fig. 20N), VO2, VCO2 (Fig. 21I-N) and energy expenditure (Fig. 20P, R) over the study. Thus, 142 NtsR1DTA mice remain lean in spite of their obesogenic diet, and this was due to chronically increased locomotor activity and energy expenditure. Loss of VTA NtsR1 Neurons Alters Hedonic and Motivated Sucrose Intake: Loss of VTA NtsR1 neurons could also promote leanness by reducing the hedonic value of food, leading to insufficient caloric intake needed to balance increased energy expenditure. Surprisingly, NtsR1DTA mice prefer and overconsume sucrose compared to control mice (Fig. 22A-C). We reasoned that if NtsR1DTA mice drink more sucrose to meet homeostatic need then their intake should be suppressed by treatment with the appetite-suppressing hormone, leptin27. While leptin mildly diminished sucrose preference in NtsR1GFP controls, it did not suppress the sucrose preference of NtsR1DTA mice (Fig. 22D). These data suggest that loss of VTA NtsR1 neurons prevents appropriate response to leptin, compromising the ability to adapt caloric intake and locomotor behaviors as needed to maintain normal weight. We next examined whether loss of VTA NtsR1 neurons might have uncoupled motivated feeding required to offset increased energy expenditure. Despite being hyperactive, NtsR1DTA mice correctly learned to self-administer sucrose in a time frame similar to that of control mice (Fig. 22E, H), confirming that loss of VTA NtsR1 neurons does not compromise learning or attention processes required for motivated intake. Furthermore, NtsR1DTA and NtsR1GFP work equally to obtain sucrose (Fig. 22I, J), thus loss of VTA NtsR1 neurons does not prevent sucrose “wanting”. Given the extreme energy demands of NtsR1DTA mice, however, their lack of increased motivated responding is counterintuitive: these mice should be more motivated than controls to consume calories28. We therefore reasoned that VTA NtsR1 neurons may be required to coordinate signals of energy status with appropriate adaptations in motivated behavior needed to maintain body weight. If this were true, then NtsR1DTA mice lacking VTA NtsR1 neurons should be unable to adjust their motivated intake in response to peripheral 143 energy cues. As expected, NtsR1GFP control mice were less motivated to nose-poke for sucrose in response to cues of energy sufficiency such as sucrose pre-feeding29 or leptin30, but NtsR1DTA mice did not adjust their intake in response to these cues (Fig. 22K, L). Taken together, these findings reveal an important role for VTA NtsR1 neurons in linking metabolic status and motivated feeding necessary to maintain body weight. Loss of VTA NtsR1 Neurons Modifies DA-Mediated Locomotor Activity: The hyperactivity of NtsR1DTA mice suggested that they retained DA-mediated locomotor behavior. Consistent with this, open field locomotor activity of NtsR1DTA and NtsR1GFP mice was not disrupted by PBS-injection stress (Fig. 23A, B, Test A), whereas treatment with the D1 receptor antagonist SCH23390 (0.1mg/kg i.p.) blunted locomotor activity in both groups (Fig. 23A, B, Test B). We therefore used amphetamine (AMPH)-induced locomotor activity to assess the integrity of the mesolimbic DA system in NtsR1DTA mice. As expected, AMPH treatment (4mg/kg i.p.) significantly increased locomotor activity in NtsR1GFP mice, but NtsR1DTA mice did not respond with increased activity (Fig 23D-G). Instead, we anecdotally observed that AMPHtreated NtsR1DTA mice engaged in stationary, stereotypic movements typical with ceiling effects of DA signaling, which account for their apparent decrease in ambulatory activity (Fig. 23D-G). Since NtsR1DTA mice lack a subset of VTA DA neurons that projects to the NA, we examined the ability of AMPH to induce DA release in the NA using AMPH-induced cFos immunoreactivity to identify activated NA neurons. AMPH treatment significantly increased numbers of cFospositive cells in the NA of both NtsR1DTA and control mice, although the response was nonsignificantly blunted in NtsR1DTA mice, consistent with loss of some, but not all NA-projecting VTA DA neurons (Fig. 23H, I). Together, these data indicate that loss of VTA NtsR1 neurons does not completely abolish mesolimbic signaling. In fact, the high baseline locomotor activity and AMPH-induced stereotypy observed in NtsR1DTA mice may indicate enhanced DA signaling, similar to compensatory increases in extracellular DA that occur with partial loss of DA 144 neurons31. Although hyperactivity and excessive DA have been linked with anxiety, there were no differences in anxiety-like behavior between NtsR1GFP and NtsR1DTA mice as assessed via open field center activity (Fig. 23C, F) and elevated plus maze (EPM) (Fig. 24). Loss of VTA NtsR1 Neurons Modifies the Mesolimbic DA System: Lastly, we examined how loss of VTA NtsR1 neurons impacts the integrity of the mesolimbic DA system. NtsR1DTA mice have reduced TH and DAT immune-labeled terminals in the NA and OFT compared to control mice (Fig. 25A, B) but residual immunoreactivity suggests preservation of non-NtsR1 expressing mesolimbic neurons. To assess molecular alterations in the VTA, we administered unilateral VTA injections of AAV-DTA or AAV-GFP to NtsR1Cre;GFP mice and analyzed the fold difference in gene expression between the injected and un-injected sides; we refer to these as NtsR1GFP-Uni and NtsR1DTA-Uni mice (Fig. 25C). As expected, NtsR1 and DA-associated transcripts such as Th, Dat and D2 were significantly reduced in the VTA of NtsR1DTA-Uni mice, consistent with the loss of DAergic VTA NtsR1 cell bodies (Fig. 25D). The slight reduction in vglut2 and vgat in NtsR1DTA-Uni mice suggests that VTA NtsR1 neurons may also contain glutamate and/or GABA in addition to DA. Ntsr2 expression, however, did not differ between NtsR1DTA-Uni mice and controls, consistent with our finding that NtsR1 and NtsR2 are expressed on distinct VTA cell populations (Fig. 18). No differences in NA expression of D1 or D2 were observed between groups (Fig. 25E). In sum, loss of VTA NtsR1 neurons results in structural and molecular adaptations in the mesolimbic DA system but does not abrogate the entirety of DA-mediated signaling that is required for survival. 4.4 Discussion The lack of molecular markers to differentiate functionally heterogeneous and projectionspecified VTA DA neurons has impeded understanding of how these neurons orchestrate behavior and body weight. A recent study reported that a subpopulation of VTA DA neurons 145 that project to the lateral septum can be identified by the transcription factor, Neurod632, however no additional markers of projection-specific VTA subsets have been identified. Here, we show that NtsR1 expression identifies a subset of VTA DA neurons that specifically projects to the ventral striatum and thus is the first molecular marker specific to mesolimbic DA neurons. Loss of this VTA NtsR1 population alters DA-mediated behaviors to protect against weight gain without compromising DA-mediated signaling necessary for survival4. Collectively, our data identify VTA NtsR1 neurons as important coordinators of energy intake and output behaviors that determine body weight. Since Nts actions via NtsR1 and NtsR2 are implicated in control of distinct physiology, we hypothesized that these receptors might identify functionally distinct DA neurons. We therefore used Cre-mediated reporter mice to identify the cells expressing each receptor isoform, confirming that NtsR1 is almost exclusively expressed by DA neurons while NtsR2 is primarily expressed by non-DAergic neurons and astrocytes. Since NtsR1 is a Gq-coupled receptor33, our findings indicate a mechanism for Nts to directly enhance the activation of NtsR1-expressing DA neurons, consistent with the established roles for NtsR1 in promoting DA release to the striatum and DA-mediated locomotor activity34,35, anorexia35, and reward36,37. Our data do not, however, rule out roles for the non-DAergic populations of NtsR2 cells to indirectly promote DA signaling. The TH-negative NtsR2 neurons found primarily within the ventromedial, GABA-rich portion of the VTA may disinhibit local DA neurons and promote their activation37,38. Nts may also act via the numerous NtsR2-expressing astrocytes within the VTA to indirectly modify DA-mediated behavior, consistent with reports of glia mediating NtsR2-dependent fear behavior39. Going forward, it will be important to define the neural circuits by which Nts engages these distinct NtsR1 and NtsR2-expressing cell types and their individual contributions to DAmediated behaviors and energy balance. 146 In situ hybridization predicts a larger population of NtsR1-expressing neurons in the VTA than were identified using transgenic NtsR1Cre-GFP mice (Allen Brain Atlas and Figure 1). This discrepancy may be due to low Cre expression common to transgenic lines, diminishing the intensity of Cre-mediated EGFP expression necessary to identify NtsR1 cells. Increasing the concentration of Cre-inducible transcripts can enhance recombination, and indeed, injecting NtsR1Cre mice with AAV-GFP identifies additional NtsR1 neurons in the VTA. The AAVidentified VTA NtsR1-GFP neurons exclusively co-label with a subpopulation of TH cells, confirming that VTA NtsR1 neurons are a subset of all DA neurons (Fig. 26). In sum, these data suggest that the VTA NtsR1 neuronal population is larger than depicted in Fig. 18, but that subsequent AAV manipulations were able to transduce and modulate most VTA NtsR1 neurons (see further discussion in Chapter 6). Although generalized depletion of VTA DA neurons does not alter locomotor activity or energy intake40-42, ablation of the specific subset of VTA NtsR1 neurons altered both behaviors and prevented age-associated and diet-induced weight gain. Since aging and obesogenic environment are the two most significant instigators of human weight gain, our findings suggest roles for VTA NtsR1 neurons in countering the development of obesity. Intriguingly, the pronounced physical activity and energy expenditure of NtsR1DTA mice appears to be the primary source of protection from weight gain, and is consistent with the established role for mesolimbic DA in regulating locomotor behavior. Although NtsR1DTA mice eat slightly more than controls, their energy intake is only sufficient to compensate for elevated levels of activity and energy expenditure, and did not promote weight gain. Thus, mesolimbic VTA NtsR1 neurons are not essential drivers of feeding per se, but they coordinate energy intake required to support physical activity. 147 While open-circuit indirect calorimetry systems are widely accepted methods to analyze VO2 and energy expenditure53, methodological limitations should be taken into account when interpreting the data. For example, the metabolic chambers record food intake by detecting slight differences in weight of the food hopper as food is removed by the mouse during feeding. However, this does not ensure that all food removed from the hopper was consumed. Further, the analysis does not take into account the possibility of differences in digestion efficiency between NtsR1DTA and control mice as not all ingested food may have been absorbed for energy. Differences in digestion could be addressed by assessing caloric content of fecal matter, which was not performed in this study. These two caveats together pose the possibility that food intake could be slightly overestimated. Another limitation is that O2 and CO2 concentrations were collected once for 3 minutes every 27 minutes over the duration each 4 day experiment. Thus, we can only assume that metabolism during the 3 minute collection period represents the entire 27 minute interval, which could overestimate or underestimate energy expenditure. To examine this, we plotted energy intake vs. expenditure for both groups at 4 and 16 weeks post-ablation (Fig. 20S, T). As long as no weight gain or loss occur, the slope of the distributions between intake and expenditure should be near 1.0. We found however, that the slopes were much less than 1.0 but similar between groups. A slope <1.0 suggests that the food intake exceeds energy expenditure, yet neither group gained weight during the study periods (data not shown), suggesting that either 1) food intake is overestimated due to lack of complete consumption or differences in digestion or 2) energy expenditure is underestimated, perhaps due to the assumption that energy expenditure during the 3 minute recording interval is representative of a larger time period, or a combination of both. However, the fact that the slopes did not vary significantly between ablated and control mice suggests that any error in estimating caloric intake or expenditure applies equally to both groups. Further, we used weight-adjusted ANCOVA analysis to detect significant increases in VO2, VCO2, and energy expenditure with VTA NtsR1 ablation (Fig. 20, 21), which eliminates the potential for erroneous 148 differences that can occur when metabolic parameters are simply divided by body weight53. Thus, although the absolute values for caloric intake and expenditure may vary by system, we firmly believe the differences detected between groups are valid. Given that VTA NtsR1 neurons project to the NA, where DA release regulates the motivation for palatable foods2,3, we were surprised that NtsR1DTA mice did not differ from controls in their willingness to work for sucrose rewards. Furthermore, NtsR1DTA mice demonstrated intact liking and wanting of palatable foods. These findings suggest that VTA NtsR1 neurons are not required for DA-mediated ingestive behavior, and that other non-NtsR1 expressing circuits exert this control. It may be argued, however, that NtsR1DTA mice with low body weight and adiposity should display increased motivation for sucrose28, and thus have a deficit in coordinating intake behavior to meet energy demands. This is consistent with our findings that NtsR1DTA mice also lack the ability to adapt motivated intake in the face of satiety cues such as sucrose pre-feeding and leptin treatment30. Taken together, these data suggest that VTA NtsR1 neurons are not required for the execution of motivated food intake, but rather serve to tune it in response to peripheral energy needs. VTA NtsR1 neurons must receive information about peripheral energy status to coordinate metabolic state and DA-mediated behavior. The adipocyte-derived hormone leptin modulates energy balance in part via the actions of Nts and NtsR125,43, and we demonstrate an obligate role of VTA NtsR1 neurons in this process. It is unlikely that VTA NtsR1 neurons are direct targets of leptin because VTA NtsR1- and leptin receptor (LepRb) expressing populations have distinct anatomical distributions and projection sites (Fig. 19 and 44), suggesting that they comprise distinct neuronal populations. Leptin does, however activate LepRb neurons of the lateral hypothalamic area (LHA) that project to and release Nts into the VTA, promoting NtsR1- 149 dependent release of DA in the NA25,45. LHA LepRb neurons may thus be the neural hubs linking leptin action with Nts and DA signaling. We used NtsR1 as a marker to identify VTA NtsR1 neurons, but it remains to be determined which signals from these neurons are critical for regulation of energy balance. VTA NtsR1 neurons may co-express both GABA and/or glutamate to modify activation of striatal targets. Nts action via NtsR1 expressed on VTA NtsR1 neurons may also contribute to DAmediated behaviors. Indeed, the phenotype of NtsR1DTA mice resembles whole-body NtsR1 knockout mice with hyperactivity, increased sucrose preference, and lack of adaptive response to leptin25. Together, these models argue that chronic disruption of Nts signaling via VTA NtsR1 neurons biases homeostasis toward energy expenditure. Perhaps the most important signal from VTA NtsR1 neurons may be DA itself. Although NtsR1DTA mice exhibit substantial loss of VTA NtsR1 neurons and diminished DAergic projections to the NA, their intact AMPH-induced cFos confirms that at least some DA neurons remain and are functional. If anything, chronic loss of VTA NtsR1 neurons may lead to enhanced DA signaling. Indeed, mice lacking VTA NtsR1 neurons display increased general ambulatory behavior, but decreased AMPH-induced locomotor activity accompanied by stereotypy, a behavioral response typically associated with high-dose AMPH and ceiling-levels of DA efflux46. We thus speculate that loss of VTA NtsR1 neurons leads to increased striatal DA signaling, and AMPH further increases DA to levels that promote stereotypy instead of augmenting ambulatory activity. Numerous mechanisms could enhance DA action, such as altered balance of D1/D2 protein expression or impaired DAT kinetics, and will be important to define in the future. Our data indicate that VTA NtsR1 neurons play an important role in preventing dietinduced obesity, hence VTA NtsR1 neurons may represent a novel cellular target for preventing and treating human obesity. Ablation of neurons is not a feasible therapeutic strategy, thus it 150 will be imperative to identify pharmacological means to selectively modulate VTA NtsR1 neurons to promote behaviors that prevent weight gain. Previous work demonstrates that pharmacologic treatment of Nts or Nts agonists in the VTA suppresses operant-reinforced feeding47 and enhances locomotor activity in an NtsR1-dependent manner19,34. Conversely, other studies showed suppression of locomotor activity when Nts was applied directly to the NA14, underscoring the need for circuit-specific interventions. Although developmental loss of NtsR1 in NtsR1 knockout mice promotes palatable food intake and weight gain25, our current data show that selective ablation VTA NtsR1 neurons in adult mice protects them from obesity, suggesting that pharmacological modulation of Nts signaling may be useful in adults with established VTA NtsR1 DA circuits. Going forward, it will be important to define how Nts and other signals engage VTA NtsR1 neurons to identify strategies that prevent obesity 4.5 Methods Reagents: Recombinant leptin was purchased from the National Hormone and Peptide Program (Torrance, CA). The dopamine receptor-1 antagonist, SCH233990, was purchased from Sigma. Amphetamine hydrochloride was from Cayman Pharmaceuticals. Mice: Transgenic NtsR1Cre mice were purchased from the Mutant Mouse Regional Resource Center at UC Davis (B6.FVB(Cg)-Tg(NtsR1-cre)GN220Gsat/Mmucd, Stcok number 030648-UCD). NtsR2Cre mice were generated via homologous recombination (knock-in) methods described previously48. To visualize NtsR1 and NtsR2-expressing neurons, NtsR1Cre and NtsR2Cre mice were bred to a Cre-inducible RosaeGFP-L10a reporter line, generating mice that express GFP selectively in NtsR1 and NtsR2 neurons (NtsR1Cre;GFP and NtsR2Cre;GFP). Mice were bred and housed in a 12h light/12h dark cycle and all procedures were approved by the Institutional Animal Care and Use Committee (IACUC) at Michigan State University. 151 Generation of NtsR2Cre Knock-In Mice: An NtsR2cre targeting vector was generated by inserting an IRES-Cre sequence between the stop codon and the polyadenylation site of the sequence encoding the 3’ end of the mouse NtsR2 gene, with an frt-flanked Neo cassette inserted upstream of the IRES-Cre. The linearized targeting vector was electroporated into mouse R1 embryonic stem (ES) cells (129sv background) and cells were selected with G418. DNA from ES cell clones was analyzed via qPCR for loss of homozygosity using Taqman primer and probes for the genomic NtsR2 insertion site (Forward: ACCCATCAGATAAGCCATGC, Reverse: GTGGGAAGTTGAGGGCAG, Probe: GTCTAAGCGGACCTACTGACCCA). NGF was used as a copy number control49. Putative positive ES clones were expanded, confirmed for homologous recombination by Southern blot and injected into mouse C57BL/6 blastocysts to generate chimeras. Chimeric males were mated with C57BL/6 females (Jackson Laboratory), and germline transmission was determined initially via progeny coat color, then confirmed via conventional PCR for IRES-Cre. Genotyping: NtsR1Cre and NtsR2Cre mice were bred to C57BL/6J mice to maintain the lines, and progeny were genotyped via standard PCR (NtsR1Cre: F: GACGGCACGCCCCCCTTA, R: CGGCAAACGGACAGAAGCATT, NtsR2Cre: CCGTGTCTTCCTTCAGA, R: CTACACCTTGGTTGCACAGG). Surgery: Mice received a pre-surgical injection of carprofen (5mg/kg s.c.) and were anesthetized with 3-4% isoflurane/O2 in an induction chamber before being placed in a stereotaxic frame (Kopf). Under 1-2% isoflurane, access holes were drilled in the skull allowing a guide cannula with stylet (PlasticsOne) to be lowered into the brain target area. After 7 min to allow for AAV absorption, the injector and cannula were removed from the skull and the incision was closed using Vet Bond. To generate NtsR1GFP and NtsR1DTA study mice, 8-10 wk old NtsR1Cre;GFP males received bilateral injections of either 150 nL of Cre-inducible AAV-GFP 152 (rAAV2/hSvn-DIO-eGFP, University of North Carolina Vector Core) or AAV-DTA (lox-mCherryloxDTA-WPRE-AAV, serotype 10), which expresses the cytotoxic Subunit A of Diptheria Toxin in the presence of Cre, into the VTA (A/P: -3.2, M/L: +/-0.48, D/V: -4.65). NtsR1GFP and NtsR1DTA mice were monitored weekly for body weight after surgery, but did not undergo analysis until 4 wks post-surgery to allow DTA-mediated cell death to occur. We generated and studied NtsR1GFP = 7, NtsR1DTA =15 and NtsR1GFP HF=10, NtsR1DTA HF=13. Mice were only included in the final study data if injections were localized to, and contained within the VTA which was determined by specific lack of GFP expression in the VTA. After post-hoc examination, 2/15 NtsR1DTA and 3/13 NtsR1DTA-HF mice were excluded due to unilateral targeting and/or spread of the injection site to the substantia nigra (SN), leaving NtsR1GFP = 7 and NtsR1DTA =13; NtsR1GFP HF=10 and NtsR1DTA HF=10. To visualize NtsR2 neurons, adult male and female NtsR2Cre;GFP were bilaterally injected with 1uL FlpO adenovirus (Vector Biolabs) into the lateral ventricles in accordance with the atlas of Paxinos and Franklin50: A/P: -0.22, M/L: +/- 1.0, D/V: -2.0. Mice were perfused 10 days after surgery to permit sufficient time for FlpO-mediated excision of the frt-flanked Neo cassette and GFP expression. For tracing studies, NtsR1Cre mice were injected unilaterally in the VTA with 75-100 nL of an anterograde cre-inducible adenovirus expressing a synaptophysin-mCherry fusion protein, Ad-syn-mCherry25, (Martin Myers, University of Michigan) using the same coordinates described above. Mice recovered for 7-10 days after surgery to allow for synaptophysin-mCherry expression at pre-synaptic terminals. Metabolic Profiling: At 4, 8 and 16 wk post-surgery the mice were analyzed for body composition using and NMR-based instrument (Minispec mq7.5, Bruker Optics). At 4 and 16 wk post-surgery, mice were placed in TSE cages for metabolic phenotyping (PhenoMaster, TSE Systems). After 24 hours of acclimation, mice were continuously monitored for food and water 153 intake, locomotor activity, and energy expenditure over 4 days. Ambient temperature was maintained at 20-23˚C and the airflow rate through the chambers was adjusted to maintain an oxygen differential around 0.3% at resting conditions. The O2 and CO2 sensors were calibrated to gas mixtures with verified gas concentrations. All air entering the system was filtered and dried to avoid variation in gas partial pressure due to moisture. O2 and CO2 concentrations and gas flow rates were measured over a 3 minute period from each cage individually, every 27 minutes for the duration of the run. The system consisted of 8 air-tight cages containing singlyhoused study animals and an empty cage from which gas concentrations were used as a reference to calculate oxygen consumption (VO2), carbon dioxide production (VCO2), energy expenditure, and respiratory exchange ratio (RER) as described below. Metabolic Calculations: VO2 = (Flowin * [O2]in) – (Flowout * [O2]out) VCO2 = (Flowin * [CO2]in) – (Flowout * [CO2]out) In these equations, [O2]out, [CO2]out and Flowout are measured from air exiting the experimental cage, while [O2]in and [CO2]in are measured from the reference cage. Note that although gas concentrations from both reference and experimental cages are measured after they exit the cage, no gas exchange has occurred in the reference cage concentrations can be used for the “in” terms for all cages. The only term not measured directly is Flowin, which can be solved for using the Haldane correction, based on the principle that the nitrogen concentration if air going into the cage should equal air exiting the cage as nitrogen is physiologically inert. Taking this into account: Haldane Correction: Flowin * (1-([O2]in +[CO2]in)) = Flowout * (1-([O2]out +[CO2]out)) Thus, Flowin can be solved for and used to calculate VO2 and VCO2. 154 RER= VCO2/VO2 indicates whether glucose or fatty acids are preferentially used as energy substrates. Fatty acid oxidation yields a ratio of 0.7 while glycolysis is 1.0. Energy Expenditure was calculated using the Weir equation: Metabolic rate (kcal/hr) = (3.815+1.232*RER) * VO2 All parameters were calculated by the TSE system software. Sucrose Preference: NtsR1DTA and NtsR1GFP mice were given two identical 50 mL sipper bottles in their home cages for 4 consecutive days. On days 1-2, both bottles were filled with water to acclimate the animals to the test environment. On days 3-4, one bottle was switched to 0.5% sucrose while the other contained water. The bottle positions were rotated daily to avoid position bias, and the volume consumed from each bottle was measured at the same time each day. To assess leptin-induced changes in sucrose preference, animals were re-tested one month later for 6 consecutive days. On days 1-2, animals received once daily sham i.p. injections while both bottles contained water. On day 3, one bottle was switched to 0.5% sucrose. Mice were treated with PBS on days 3-4 and leptin (5mg/kg ip) on days 5-6. All injections were administered at the onset of the dark cycle when most food and water intake naturally occurs. Data are reported as sucrose preference, which is the percentage of sucrose consumed out of total liquid consumed. Operant Testing: Standard Testing: NtsR1DTA and NtsR1GFP were trained to nose-poke for unflavored 20 mg sucrose pellets (TestDiet 1811555) using operant-responding chambers (Med Associates). As previously described30, mice were food restricted to 90% of their body weight and trained on a fixed ratio-1 (FR1) schedule until they achieved 75% response accuracy 155 with ≥20 rewards earned on 3 consecutive days. Training sessions were terminated after 1 hr or when the animal had earned 50 rewards. Mice achieving these criteria were then switched to ad lib food and trained on an FR5 schedule for 3 consecutive days. On test days, mice were subject to a progressive ratio (PR) schedule were PR=[5e(R*0.2)]-5 with R=number food rewards earned+151. Thus, the number of correct responses needed to earn a sucrose reward increases as follows: 1, 2, 4, 6, 9, 12, 15, 20, 25, 32, 40, 50, 62, 77, 95 etc. The PR breakpoint was recorded as the highest ratio completed for each 1 hr test session. Mice were tested until they achieved stable PR which was defined as <10% variation in rewards earned over 3 consecutive sessions. Sucrose pre-feeding: After achieving stable PR, mice were given ad lib access to 3g of sucrose pellets in their home cages overnight. The following morning, they were re-tested on the PR schedule. Leptin Treatment: Mice with stable PR responses were injected with PBS or leptin (5mg/kg, ip) on different days and tested one hour later. Elevated Plus Maze: Mice were assessed for anxiety-like behavior using an elevated plus maze (EPM) as previously described52. Briefly, the EPM apparatus was custom-built based on plans from ANY-maze (www.anymaze.com, Stoelting Co.) and mice were given free access to the open and closed arms for 5 minutes. Their behavior was recorded using a digital CCD camera and the percentages of time spent in the open and closed arms were analyzed using Topscan automated video tracking software (Clever Sys). Open Field Locomotor Activity: Open field locomotor activity was assessed using a digital CCD camera and video-tracking software (Clever Sys)52. Mice were tested on 3 separate days; each day, mice were placed in the boxes to acclimate, followed by an i.p. injection of PBS 30 min later. At 60 min, mice received either a second injection of PBS, or SCH23390 (0.1mg/kg) or amphetamine (4mg/kg) and recorded for another 60 minutes. 156 Immunohistochemistry and Immunofluorescence: Mice were treated with a lethal dose of ip pentobarbital followed by transcardial perfusion with 10% neutral-buffered formalin (Fisher). Brains were removed, post-fixed in 10% formalin overnight at 4˚C, dehydrated with 30% sucrose in PBS for 2-3 days, and sectioned into 30 µm slices using a sliding microtome (Leica). Brain sections were then analyzed by immunofluorescence or immunohistochemistry as previously described25,48. For characterization of NtsR1 and NtsR2 expression, sections from NtsR1Cre;GFP and NtsR2Cre;GFP mice were stained with chicken anti-GFP (1:2000, Abcam), mouse anti-TH (1:1000, Millipore), or rabbit anti-S100 (1:1000, Abcam), followed by incubation with species-specific secondary antibodies conjugated to AlexaFluor 488 or 568 fluorophores (Life Technologies or Jackson ImmunoResearch). For ablation studies, NtsR1DTA and NtsR1GFP mice were treated with PBS or amphetamine (4mg/kg) 90 minutes prior to perfusion, and brain sections were stained for cFos (1:500, goat, Santa Cruz) with secondary detection via DAB (Life Technologies), followed by immunofluorescent detection of GFP, TH, or DAT (1:1000, Millipore) as described above. Brains were analyzed using an Olympus BX53 fluorescence microscope outfitted with transmitted light to analyze DAB-labeled tissue, as well as FITC and Texas Red filters. Microscope images were collected using Cell Sens software and a Qi-Click 12 Bit cooled camera, and images were analyzed using Photoshop software (Adobe). For quantification and colocalization of NtsR1 and NtsR2 neurons with TH, five representative sections spanning the entire VTA were selected from each mouse, from which all GFP and/or TH positive neurons were counted. Gene Expression: At 16 weeks post-surgery, NtsR1DTA-Uni (n=17) and NtsR1GFP-Uni (n=7) were deeply anesthetized with sodium pentobarbital and tissue from the injected and uninjected sides of the VTA and NA were separately microdissected. Tissue was immediately snap frozen on dry ice and stored at -80˚C for later processing. RNA was extracted using Trizol (Invitrogen) and 200 ng samples were converted to cDNA using the Superscript First Strand Synthesis 157 System for RT-PCR (Invitrogen). Sample cDNAs were analyzed in triplicate via quantitative RTPCR for gene expression using TaqMan reagents and an ABI 7900 (Applied Biosystems) at the MSU Genomics Core. With GAPDH expression as an internal control, relative mRNA expression values were calculated by the 2-∆∆Ct method and normalized to the uninjected side of each mouse. To verify targeting, NtsR1DTA-Uni mice were considered sufficiently ablated if the fold change in NtsR1 expression was less than 1 standard deviation below the mean NtsR1 fold change in AAV-GFP-injected mice. By this criterion, 7 of 15 NtsR1DTA-Uni mice were deemed insufficiently ablated and were excluded from the analysis. Statistics: Student’s t-tests and 2-way ANOVA were calculated using Prism 6 (GraphPad). For all metabolic data, analysis of covariance (ANCOVA) was computed in SPSS 22 (IBM). Body weight was analyzed as a covariate to correct for any inherent differences it may have on metabolism53. All data were tested for homogeneity of regression, independence of the covariate (body weight), and linearity of regression prior to running the ANCOVA. For all data, *p<0.05, **p<0.01 and ***p<0.001. 158 APPENDIX 159 Figure 17. Generation of NtsR1Cre;GFP and NtsR2Cre;GFP reporter mice to visualize NtsR1 and NtsR2 neurons in the VTA. A) Commercially-available transgenic NtsR1Cre mice were bred to a Cre-inducible GPP reporter line to generate mice expressing GFP selectively in NtsR1 neurons (NtsR1Cre;GFP mice). B) NtsR2NEO-Cre mice were created by inserting an IRES-Cre sequence downstream of an frt-flanked NEO cassette directly into the Ntsr2 locus using homologous recombination. NtsR2NEO-Cre mice were then crossed to a Cre-inducible GFP reporter line, however GFP is not expressed because the NEO cassette prevents Cre expression. Thus, adult NtsR2NEO-Cre;GFP mice were injected with an adenovirus expressing FlpO recombinase ICV to remove the frt-flanked NEO cassette, resulting in Cre and therefore GFP expression in NtsR2-positive cells (NtsR2NEO-Cre;GFP mice). 160 Figure 18. Distribution and neurochemical phenotype of VTA NtsR1 and NtsR2 neurons. VTA NtsR1and NtsR2 neurons were visualized by crossing NtsR1Cre and NtsR2Cre mice to a Creinducible GFP reporter line. See Fig. 17 for more details. A) Rostrocaudal distribution of VTA NtsR1 and NtsR2 neurons, with bregma positions from the mouse brain atlas of Paxinos and Franklin. White arrows denote neurons, yellow arrows denote glial cells. B) Colocalization of glial cell marker S100 (purple) in the VTA with NtsR2 cells (green) and C) NtsR1 cells (green). D) Colocalization of tyrosine hydroxylase (TH, purple) in the VTA with NtsR2 cells (green) and E) NtsR1 cells (green). F) Percentage of NtsR1 and NtsR2 neurons that colocalize with TH. G) Percentage of TH positive neurons that colocalize with NtsR1 or NtsR2. H) Total number of GFP-identified NtsR1 and NtsR2 neurons in the VTA. I) Schematic depicting the distribution of VTA DA and non-DA neurons that co-express NtsR1 and NtsR2. NtsR1Cre;GFP n=4, NtsR2Cre;GFP n=6. Data represent mean ± SEM, *p < 0.05, ***p < 0.001 determined by unpaired t-test. Scale bars = 100 µm. ml = medial lemniscus; ip = interpeduncular nucleus; fr = fasciculus retroflexus. 161 Figure 19. VTA NtsR1 neurons project to the ventral striatum. A) NtsR1Cre mice were unilaterally injected in the VTA with Ad-Syn-mCherry, resulting in expression of a synaptophysin-mCherry fusion protein in the cell bodies and presynaptic terminals of VTA NtsR1 neurons (n=5). B) VTA injection site (left, asterisk) with mCherry-labeled NtsR1 cell bodies. Scale bar = 200 µm. C) Major efferent targets of VTA NtsR1 neurons. D) Minor efferent targets of VTA NtsR1 neurons. Numbered dashed-boxes in coronal schematic images correspond to microscopy images below. Bregma positions are according to the mouse brain atlas of Paxinos and Franklin. Scale bars in c, d = 100 µm. E) Table summarizing relative NtsR1 terminal density across multiple brain areas with emphasis on previously established VTA efferents. Note: n.p.= not present. aca=anterior commissure, cc=corpus callosum, ic=internal capsule, ml=medial lemniscus, ip=interpeduncular nucleus, CeA= central amygdala, BLA= basolateral amygdala. 162 Figure 20. Loss of VTA NtsR1 neurons disrupts energy balance. A) 8 wk old male NtsR1Cre;GFP mice received bilateral injections of AAV-DTA or AAV-GFP in the VTA. B) Images show representative NtsR1-GFP (green) and TH (purple) expression from NtsR1GFP and NtsR1DTA mice two weeks after surgery. Scale bars =200µm. ml=medial lemniscus, ip=interpeduncular nucleus. 163 Figure 20 (cont’d). C-J) Metabolic assessment of NtsR1GFP and NtsR1DTA mice on chow diet (NtsR1GFP n=7, and NtsR1DTA n=13). C) Percentage change in body weight from starting weight and D) percentage of body fat determined via NMR-based instrument. E) Chow intake and F) total locomotor activity as measured in TSE metabolic cages. G) Weight- adjusted energy intake and H) energy expenditure at 4 wk post-surgery and I) weight-adjusted energy intake and J) energy expenditure measured at 16 wk post-surgery. K-R) Metabolic assessment of NtsR1GFP and NtsR1DTA mice on HF diet (NtsR1GFPHF n=10 and NtsR1DT HF n=10). K) Percentage change in body weight from starting weight, L) percentage of body fat, M) HF diet intake and N) total locomotor activity. O) Energy intake and P) energy expenditure at 4 wk postsurgery, and Q) energy intake and R) energy expenditure measured at 16 wk post-surgery, normalized to body mass. S) Total energy intake vs. estimated expenditure over 48 hours at 4 weeks (NtsR1GFP: y=0.317x+26.43, NtsR1DTA: y= 0.271x+31.74) and T) 16 weeks post-ablation (NtsR1GFP: y=0.456x+16.48, NtsR1DTA: y= 0.695x+9.298) . For B-E and J-M, graphed data represent mean ± SEM, *p < 0.05, **p < 0.01, ****p < 0.0001 analyzed by two-way ANOVA. For scatterplots in G-J and O-R data were analyzed using ANCOVA to account for body weight as a covariate. S-T were analyzed by linear regression. See Fig. 21 for water intake and additional metabolic parameters. 164 Figure 21. Water intake and additional metabolic parameters in mice lacking VTA NtsR1 neurons. Fluid intake and metabolic parameters were assessed over 4 days in TSE metabolic cages in NtsR1GFP and NtsR1DTA mice at 4 and 16 wk post-surgery. 165 Figure 21 (cont’d). Water consumption in A) chow-fed and B) HF diet-fed mice. Data represent mean ± SEM, analyzed by two-way ANOVA. C) oxygen consumption (VO2), D) carbon dioxide production (VCO2) and E) respiratory exchange ratio (RER) at 4 weeks-post ablation in chow-fed animals. F) VO2, G) VCO2, and H) RER in chow-fed mice 16 weeks postablation. I) VO2, J) VCO2, and K) RER in HF diet-fed mice at 4 weeks. L) VO2, M) VCO2 and N) RER in HF-diet fed mice 16 weeks post-ablation. All metabolic data were analyzed via ANCOVA to adjust for body weight as a covariate. (NtsR1GFP n=7, NtsR1DTA n=13, NtsR1GFPHF n=10 and NtsR1DTA HF n=10). Figure 22. VTA NtsR1 neurons modulate hedonic and motivated ingestive behavior. A) Water intake, B) 0.5% sucrose intake and C) sucrose preference in untreated NtsR1GFP and NtsR1DTA mice over 4 days. D) Sucrose preference while mice received PBS or leptin (5mg/kg, i.p.) at the onset of the dark cycle. Graphed data represent the difference in average sucrose preference during PBS and leptin treatments. NtsR1GFP n=7, and NtsR1DTA n=13, n.s.=not significant, *p < 0.05, ***p < 0.001 by unpaired t-test. E-H) NtsR1GFP and NtsR1DTA mice were trained to nose-poke for sucrose pellets and assessed for response accuracy and magazine entries during FR1 training. E) Number of correct responses, F) number of incorrect responses G) number of magazine entries and H) days required to complete FR1 training. I-L) NtsR1GFP and NtsR1DTA mice were tested on a PR schedule until the number of rewards earned varied <10% over 3 consecutive days. I) Average stable PR breakpoint and J) percentage of correct responses over the three stable PR days. K) PR breakpoint was assessed after animals received ad lib access to sucrose rewards the night before testing. L) NtsR1GFP and NtsR1DTA mice were tested one hour after treatment with either PBS or leptin (5mg/kg, i.p.). NtsR1GFP n=6, and NtsR1DTA =10, n.s.=not significant, *p < 0.05, **p < 0.01, analyzed via repeated measures two-way ANOVA with Bonferroni post-tests. All graphed data represent mean ± SEM. 166 Figure 23. Loss of VTA NtsR1 neurons alters response to amphetamine. A) Locomotor activity of NtsR1GFP and NtsR1DTA mice in open field boxes during acclimation (Acc), treatment with PBS, and either PBS (Test A) or D1R antagonist SCH23390 (0.1mg/kg, i.p.) (Test B). Data represent average distance traveled per 5 min interval ± SEM (NtsR1GFP n=6, and NtsR1DTA n=10). B) Mean distance traveled during treatment period ± SEM, analyzed by two-way ANOVA. C) Percentage of distance traveled in the center zone of boxes over each testing period. D) Open field locomotor activity of NtsR1GFP and NtsR1DTA mice after treatment with PBS and AMPH (4mg/kg, i.p.). Data represent average distance traveled per 5 min interval ± SEM (NtsR1GFP n=6, and NtsR1DTA n=10). E) Mean difference between locomotor activity after AMPH treatment relative to PBS, **p=0.003. F) Percentage of distance traveled in the center zone over each period. See Fig. 24 for additional anxiety data. G) Representative heat maps demonstrating mouse location during acclimation, PBS, and AMPH treatment. H) cFos expression in the NA (-1.34 bregma) of NtsR1GFP and NtsR1DTA treated with PBS or AMPH 90 minutes prior to perfusion). aca = anterior commissure; NAc = nucleus accumbens core; NAs = nucleus accumbens shell. I) Number of cFos positive (cFos+) cells in NA. Graphed data represent mean ± SEM, analyzed by two-way ANOVA. ƗƗƗƗ p<0.001, ƗƗƗ p<0.001, indicates significant difference between PBS and amphetamine treatment within each group, (NtsR1GFP PBS n=3, AMPH=3; NtsR1DTA PBS n=3, AMPH n=5). 167 Figure 24. Anxiety-like behavior assessed via elevated-plus maze. No significant differences were found between groups in A) percentage of time spent on open arms or B) number of open arm entries (NtsR1GFP n=4, NtsR1DTA n=7). Figure 25. Ablation of VTA NtsR1 neurons decreases markers of DA signaling. Post-hoc analysis of A) tyrosine hydroxylase (TH) and B) dopamine transporter (DAT) immunoreactivity in the NA and OFT (-1.50 bregma) of NtsR1GFP and NtsR1DTA mice perfused 7-11 months after surgery. Scale bars = 100 µm. White boxes identify digitally enlarged areas from each image. C) NtsR1Cre mice were injected unilaterally with AAV-DTA to ablate VTA NtsR1 neurons or AAVGFP as a control. The VTA and NA were microdissected separately from each side of the brain and expression of DA-related genes was assessed at 16 wk post-surgery in the D) VTA and E) NA. Fold change was determined for each animal relative to the un-injected side. Data are expressed as mean ± SEM (NtsR1GFP n=6-7, NtsR1DTA n=8-9). Significant differences were determined by unpaired t-test *p < 0.5, **p < 0.1, ***p < 0.001. aca = anterior commissure. 168 Figure 26. Virally-induced vs. endogenous GFP reporter expression for identification of VTA NtsR1 neurons. A) GFP expression induced by Cre-inducible AAV-GFP injected into the VTA of NtsR1Cre;GFP transgenic mice. Yellow arrows indicate NtsR1 neurons with reduced GFP expression that do not express endogenous L10-GFP and colocalize with TH. B) Endogenous L10-GFP expression in the VTA of uninjected NtsR1Cre;GFP mice reveals fewer labeled neurons compared to AAV-GFP-injected mice of the same genotype. Scale bar represents 100 µm. ip=interpeduncular nucleus, ml=medial lemniscus. 169 REFERENCES 170 REFERENCES 1 Soleymani, T., Daniel, S. & Garvey, W. T. Weight maintenance: challenges, tools and strategies for primary care physicians. Obesity reviews : an official journal of the International Association for the Study of Obesity 17, 81-93, doi:10.1111/obr.12322 (2016). 2 Day, J. J., Roitman, M. F., Wightman, R. M. & Carelli, R. M. Associative learning mediates dynamic shifts in dopamine signaling in the nucleus accumbens. Nature neuroscience 10, 1020-1028, doi:10.1038/nn1923 (2007). 3 McCullough, L. D., Cousins, M. S. & Salamone, J. D. The role of nucleus accumbens dopamine in responding on a continuous reinforcement operant schedule: a neurochemical and behavioral study. Pharmacology, biochemistry, and behavior 46, 581-586 (1993). 4 Zhou, Q. Y. & Palmiter, R. D. Dopamine-deficient mice are severely hypoactive, adipsic, and aphagic. Cell 83, 1197-1209 (1995). 5 Beeler, J. A., Faust, R. P., Turkson, S., Ye, H. & Zhuang, X. Low Dopamine D2 Receptor Increases Vulnerability to Obesity Via Reduced Physical Activity Not Increased Appetitive Motivation. Biological psychiatry, doi:10.1016/j.biopsych.2015.07.009 (2015). 6 Stice, E., Spoor, S., Bohon, C. & Small, D. M. Relation between obesity and blunted striatal response to food is moderated by TaqIA A1 allele. Science (New York, N.Y.) 322, 449-452, doi:10.1126/science.1161550 (2008). 7 Johnson, P. M. & Kenny, P. J. Dopamine D2 receptors in addiction-like reward dysfunction and compulsive eating in obese rats. Nature neuroscience 13, 635-641, doi:10.1038/nn.2519 (2010). 8 Lammel, S., Ion, D. I., Roeper, J. & Malenka, R. C. Projection-specific modulation of dopamine neuron synapses by aversive and rewarding stimuli. Neuron 70, 855-862, doi:10.1016/j.neuron.2011.03.025 (2011). 9 Lammel, S. et al. Input-specific control of reward and aversion in the ventral tegmental area. Nature 491, 212-217, doi:10.1038/nature11527 (2012). 10 Lammel, S. et al. Unique properties of mesoprefrontal neurons within a dual mesocorticolimbic dopamine system. Neuron 57, 760-773, doi:10.1016/j.neuron.2008.01.022 (2008). 11 Root, D. H. et al. Single rodent mesohabenular axons release glutamate and GABA. Nature neuroscience 17, 1543-1551, doi:10.1038/nn.3823 http://www.nature.com/neuro/journal/v17/n11/abs/nn.3823.html#supplementaryinformation (2014). 171 12 Stamatakis, A. M. et al. A unique population of ventral tegmental area neurons inhibits the lateral habenula to promote reward. Neuron 80, 1039-1053, doi:10.1016/j.neuron.2013.08.023 (2013). 13 Reynolds, S. M. & Berridge, K. C. Positive and negative motivation in nucleus accumbens shell: bivalent rostrocaudal gradients for GABA-elicited eating, taste "liking"/"disliking" reactions, place preference/avoidance, and fear. The Journal of neuroscience : the official journal of the Society for Neuroscience 22, 7308-7320, doi:20026734 (2002). 14 Kalivas, P. W., Burgess, S. K., Nemeroff, C. B. & Prange, A. J., Jr. Behavioral and neurochemical effects of neurotensin microinjection into the ventral tegmental area of the rat. Neuroscience 8, 495-505 (1983). 15 Vittoz, N. M. & Berridge, C. W. Hypocretin/orexin selectively increases dopamine efflux within the prefrontal cortex: involvement of the ventral tegmental area. Neuropsychopharmacology : official publication of the American College of Neuropsychopharmacology 31, 384-395, doi:10.1038/sj.npp.1300807 (2006). 16 Wanat, M. J., Hopf, F. W., Stuber, G. D., Phillips, P. E. & Bonci, A. Corticotropinreleasing factor increases mouse ventral tegmental area dopamine neuron firing through a protein kinase C-dependent enhancement of Ih. The Journal of physiology 586, 21572170, doi:10.1113/jphysiol.2007.150078 (2008). 17 Anderberg, R. H., Anefors, C., Bergquist, F., Nissbrandt, H. & Skibicka, K. P. Dopamine signaling in the amygdala, increased by food ingestion and GLP-1, regulates feeding behavior. Physiology & behavior 136, 135-144, doi:10.1016/j.physbeh.2014.02.026 (2014). 18 Cador, M., Kelley, A. E., Le Moal, M. & Stinus, L. Ventral tegmental area infusion of substance P, neurotensin and enkephalin: differential effects on feeding behavior. Neuroscience 18, 659-669 (1986). 19 Elliott, P. J. & Nemeroff, C. B. Repeated neurotensin administration in the ventral tegmental area: effects on baseline and D-amphetamine-induced locomotor activity. Neuroscience letters 68, 239-244 (1986). 20 Feifel, D., Goldenberg, J., Melendez, G. & Shilling, P. D. The acute and subchronic effects of a brain-penetrating, neurotensin-1 receptor agonist on feeding, body weight and temperature. Neuropharmacology 58, 195-198, doi:10.1016/j.neuropharm.2009.07.001 (2010). 21 Lepee-Lorgeoux, I., Betancur, C., Rostene, W. & Pelaprat, D. Differential ontogenetic patterns of levocabastine-sensitive neurotensin NT2 receptors and of NT1 receptors in the rat brain revealed by in situ hybridization. Brain research. Developmental brain research 113, 115-131 (1999). 22 Palacios, J. M., Pazos, A., Dietl, M. M., Schlumpf, M. & Lichtensteiger, W. The ontogeny of brain neurotensin receptors studied by autoradiography. Neuroscience 25, 307-317 (1988). 172 23 Sotty, F. et al. Differential effects of neurotensin on dopamine release in the caudal and rostral nucleus accumbens: a combined in vivo electrochemical and electrophysiological study. Neuroscience 85, 1173-1182 (1998). 24 Mazei-Robison, M. S. & Nestler, E. J. Opiate-induced molecular and cellular plasticity of ventral tegmental area and locus coeruleus catecholamine neurons. Cold Spring Harbor perspectives in medicine 2, a012070, doi:10.1101/cshperspect.a012070 (2012). 25 Opland, D. et al. Loss of neurotensin receptor-1 disrupts the control of the mesolimbic dopamine system by leptin and promotes hedonic feeding and obesity. Molecular metabolism 2, 423-434, doi:10.1016/j.molmet.2013.07.008 (2013). 26 Pei, H., Sutton, A. K., Burnett, K. H., Fuller, P. M. & Olson, D. P. AVP neurons in the paraventricular nucleus of the hypothalamus regulate feeding. Molecular metabolism 3, 209-215, doi:10.1016/j.molmet.2013.12.006 (2014). 27 Halaas, J. L. et al. Weight-reducing effects of the plasma protein encoded by the obese gene. Science (New York, N.Y.) 269, 543-546 (1995). 28 Fulton, S., Woodside, B. & Shizgal, P. Modulation of brain reward circuitry by leptin. Science (New York, N.Y.) 287, 125-128 (2000). 29 Scheggi, S., Secci, M. E., Marchese, G., De Montis, M. G. & Gambarana, C. Influence of palatability on motivation to operate for caloric and non-caloric food in non food-deprived and food-deprived rats. Neuroscience 236, 320-331, doi:10.1016/j.neuroscience.2013.01.027 (2013). 30 Sharma, S., Hryhorczuk, C. & Fulton, S. Progressive-ratio responding for palatable highfat and high-sugar food in mice. Journal of visualized experiments : JoVE, e3754, doi:10.3791/3754 (2012). 31 Robinson, T. E., Mocsary, Z., Camp, D. M. & Whishaw, I. Q. Time course of recovery of extracellular dopamine following partial damage to the nigrostriatal dopamine system. The Journal of neuroscience : the official journal of the Society for Neuroscience 14, 2687-2696 (1994). 32 Khan, S. et al. Survival of a Novel Subset of Midbrain Dopaminergic Neurons Projecting to the Lateral Septum Is Dependent on NeuroD Proteins. The Journal of neuroscience : the official journal of the Society for Neuroscience 37, 2305-2316, doi:10.1523/jneurosci.2414-16.2016 (2017). 33 Tanaka, K., Masu, M. & Nakanishi, S. Structure and functional expression of the cloned rat neurotensin receptor. Neuron 4, 847-854 (1990). 34 Steinberg, R. et al. SR 48692, a non-peptide neurotensin receptor antagonist differentially affects neurotensin-induced behaviour and changes in dopaminergic transmission. Neuroscience 59, 921-929 (1994). 35 Remaury, A. et al. Targeted inactivation of the neurotensin type 1 receptor reveals its role in body temperature control and feeding behavior but not in analgesia. Brain research 953, 63-72 (2002). 173 36 Rouibi, K., Bose, P., Rompre, P. P. & Warren, R. A. Ventral Midbrain NTS1 Receptors Mediate Conditioned Reward Induced by the Neurotensin Analog, D-Tyr[11]neurotensin. Frontiers in neuroscience 9, 470, doi:10.3389/fnins.2015.00470 (2015). 37 Kempadoo, K. A. et al. Hypothalamic neurotensin projections promote reward by enhancing glutamate transmission in the VTA. The Journal of neuroscience : the official journal of the Society for Neuroscience 33, 7618-7626, doi:10.1523/jneurosci.258812.2013 (2013). 38 Piccart, E., Courtney, N. A., Branch, S. Y., Ford, C. P. & Beckstead, M. J. Neurotensin Induces Presynaptic Depression of D2 Dopamine Autoreceptor-Mediated Neurotransmission in Midbrain Dopaminergic Neurons. The Journal of neuroscience : the official journal of the Society for Neuroscience 35, 11144-11152, doi:10.1523/jneurosci.3816-14.2015 (2015). 39 Yamauchi, R. et al. Neurotensin type 2 receptor is involved in fear memory in mice. Journal of neurochemistry 102, 1669-1676, doi:10.1111/j.1471-4159.2007.04805.x (2007). 40 Ouachikh, O., Dieb, W., Durif, F. & Hafidi, A. Differential behavioral reinforcement effects of dopamine receptor agonists in the rat with bilateral lesion of the posterior ventral tegmental area. Behavioural brain research 252, 24-31, doi:10.1016/j.bbr.2013.05.042 (2013). 41 Drui, G. et al. Loss of dopaminergic nigrostriatal neurons accounts for the motivational and affective deficits in Parkinson's disease. Molecular psychiatry 19, 358-367, doi:10.1038/mp.2013.3 (2014). 42 Pioli, E. Y. et al. Differential behavioral effects of partial bilateral lesions of ventral tegmental area or substantia nigra pars compacta in rats. Neuroscience 153, 1213-1224, doi:10.1016/j.neuroscience.2008.01.084 (2008). 43 Kim, E. R., Leckstrom, A. & Mizuno, T. M. Impaired anorectic effect of leptin in neurotensin receptor 1-deficient mice. Behavioural brain research 194, 66-71, doi:10.1016/j.bbr.2008.06.024 (2008). 44 Leshan, R. L. et al. Ventral tegmental area leptin receptor neurons specifically project to and regulate cocaine- and amphetamine-regulated transcript neurons of the extended central amygdala. The Journal of neuroscience : the official journal of the Society for Neuroscience 30, 5713-5723, doi:10.1523/jneurosci.1001-10.2010 (2010). 45 Patterson, C. M. et al. Ventral tegmental area neurotensin signaling links the lateral hypothalamus to locomotor activity and striatal dopamine efflux in male mice. Endocrinology 156, 1692-1700, doi:10.1210/en.2014-1986 (2015). 46 Yates, J. W., Meij, J. T., Sullivan, J. R., Richtand, N. M. & Yu, L. Bimodal effect of amphetamine on motor behaviors in C57BL/6 mice. Neuroscience letters 427, 66-70, doi:10.1016/j.neulet.2007.09.011 (2007). 174 47 Kelley, A. E., Cador, M., Stinus, L. & Le Moal, M. Neurotensin, substance P, neurokininalpha, and enkephalin: injection into ventral tegmental area in the rat produces differential effects on operant responding. Psychopharmacology 97, 243-252 (1989). 48 Leinninger, G. M. et al. Leptin action via neurotensin neurons controls orexin, the mesolimbic dopamine system and energy balance. Cell metabolism 14, 313-323, doi:10.1016/j.cmet.2011.06.016 (2011). 49 Soliman, G. A. et al. A simple qPCR-based method to detect correct insertion of homologous targeting vectors in murine ES cells. Transgenic research 16, 665-670, doi:10.1007/s11248-007-9110-2 (2007). 50 Keith B.J. Franklin, G. P. The Mouse Brain In Stereotaxic Coordinates. Third edn, (Elsevier, 2007). 51 Richardson, N. R. & Roberts, D. C. Progressive ratio schedules in drug selfadministration studies in rats: a method to evaluate reinforcing efficacy. Journal of neuroscience methods 66, 1-11 (1996). 52 Eagle, A. L. et al. Experience-Dependent Induction of Hippocampal DeltaFosB Controls Learning. The Journal of neuroscience : the official journal of the Society for Neuroscience 35, 13773-13783, doi:10.1523/jneurosci.2083-15.2015 (2015). 53 Tschop, M. H. et al. A guide to analysis of mouse energy metabolism. Nature methods 9, 57-63, doi:10.1038/nmeth.1806 (2012). 175 Chapter 5. A Central Neurotensin Circuit That Coordinates Weight Loss Behaviors Authors: Hillary L. Woodworth, Bethany G. Beekly, Hannah M. Batchelor, Raluca Bugescu, Patricia Perez-Bonilla, Laura E. Schroeder, and Gina M. Leinninger This chapter is a modified version of a manuscript under review 5.1 Abstract The central mechanism by which neurotensin (Nts) potentiates weight loss has remained elusive. We leveraged chemogenetics to reveal that Nts-expressing neurons of the lateral hypothalamic area (LHA) promote weight loss by increasing volitional activity and restraining food intake. Intriguingly, these dual weight loss behaviors are mediated by distinct signaling pathways: Nts action via NtsR1 is required for the anorectic effect of the LHA Nts circuit, but not for regulation of locomotor or drinking behavior. Furthermore, although LHA Nts neurons cannot reduce intake of freely available obesogenic foods, they effectively restrain motivated feeding in hungry, weight-restricted animals. LHA Nts neurons are thus vital mediators of central Nts action, particularly in the face of negative energy balance. Enhanced action via the LHA Nts circuit may be useful to suppress the increased appetite that occurs after lifestylemediated weight and prevent weight regain. 5.2 Introduction Obesity is a formidable health concern yet diet and exercise remain the only effective non-surgical treatments to support weight loss1. Individuals who lose weight via lifestyle 176 modification experience compensatory increases in appetite and diminished metabolic rate, and as a result, most regain weight2,3. Understanding how the brain coordinates feeding and energy expenditure is therefore crucial to identify strategies that support sustained weight loss. Dopamine (DA) released from the ventral tegmental area (VTA) and substantia nigra (SN) is essential for feeding behavior, volitional physical activity, and the maintenance of body weight4. In particular, VTA DA released to the nucleus accumbens (NA) modifies food seeking and “wanting” of palatable foods that can override homeostatic need, leading to overconsumption of calorie-rich food and weight gain5-7. Indeed, DA signaling is altered in obesity8,9 but broad pharmacologic manipulation of DA action (with effects in the NA and other sites of DA release) provides only modest weight loss1. Selective modulation of mesolimbic DA signaling may be useful to modulate feeding and weight without off-target effects. One possible candidate to modify mesolimbic DA action and body weight is the neuropeptide, neurotensin (Nts). Direct application of Nts to the VTA suppresses food intake, increases locomotor activity, and increases DA release to the NA10-12. Nts signals via Nts receptors 1 and 2 (NtsR1 and NtsR2), and NtsR1 in particular is expressed on a subset of VTA DA neurons and has been implicated in Nts-mediated suppression of feeding13-15. Although Nts expression is reduced in the brain of obese animals16, treatment with NtsR1 agonists reduces their food intake and promotes weight loss13. Thus, Nts is a putative anorectic signal, but the neural circuits by which Nts engages the mesolimbic DA system to coordinate energy balance remain to be elucidated. An important source of central Nts originates from the lateral hypothalamic area (LHA), a crucial hub for coordinating peripheral energy cues with DA-mediated motivated behaviors to regulate body weight14,17. The numerous Nts-expressing neurons in the LHA are distinct from orexigenic neurons expressing melanin-concentrating hormone and orexin/hypocretin, project to the VTA and are activated by signals that suppress feeding such as the appetite-suppressing 177 hormone, leptin18. Indeed, loss of leptin signaling via LHA Nts neurons increases adiposity through diminished volitional physical activity, and disrupts coordination between peripheral energy status and motivated feeding behavior18,19. By contrast, experimental activation of LHA Nts neurons increases Nts release in the VTA, promotes NA DA release via NtsR1, and enhances locomotor activity20. We therefore hypothesized that LHA Nts neurons are physiological mediators of Nts-mediated weight loss behaviors. Using a chemogenetic strategy (DREADDs) to selectively activate LHA Nts neurons in the presence and absence of NtsR1, we reveal an important role for these neurons, and specifically for Nts action via NtsR1 in regulating body weight. 5.3 Results To experimentally activate LHA Nts neurons and simultaneously determine the requirement for Nts signaling via NtsR1, we crossed NtsCre mice to a NtsR1 knockout line, generating NtsCre mice that have either intact or absent NtsR1 signaling (NtsCre;++ and NtsCre;NtsR1KOKO mice); these mice are henceforth referred to as WT and NtsR1KO. Adult WT and NtsR1KO mice were injected with Cre-inducible AAV-hM3Dq-mCherry in the LHA to express excitatory DREADD receptors selectively in LHA Nts neurons (Fig. 27A). Treating mice with clozapine-N-oxide (CNO) induced cFos in mCherry-labeled LHA Nts neurons, verifying that this method activates LHA Nts neurons (Fig. 27B). We confirmed that WT and NtsR1KO mice have similar amounts of mCherry-labeled Nts cell bodies in the LHA and terminals in the VTA, indicating that developmental deletion of NtsR1 does not structurally disrupt the LHA Nts circuit (Fig. 27 C, D). Additionally, NtsR1KO mice lack VTA Ntsr1, but they do not exhibit any compensatory increase in the other signaling receptor for Nts, NtsR2, (Fig. 27Ee). Furthermore, although LHA Nts neurons project to the VTA and within the LHA14,19, the dearth of LHA NtsR1 neurons suggests that Nts signaling is predominantly mediated via the dense population of VTA NtsR1 neurons (Fig. 27F). Together these data confirm an intact LHA Nts projection to the 178 NtsR1-rich VTA, and that activating LHA Nts neurons in WT and NtsR1KO mice can reveal the physiological role of Nts action via this circuit Acute Activation of LHA Nts Neurons Promotes Activity and Suppresses Feeding: We first examined how activation of LHA Nts neurons alters energy balance by analyzing WT and NtsR1KO mice in metabolic chambers after single injections of VEH or CNO. Activation of LHA Nts neurons robustly increased water intake over 8 hr in WT mice and (to a lesser extent) in NtsR1KO mice, suggesting that NtsR1 is not essential for LHA Nts-induced drinking (Fig. 28A, B). WT and NtsR1KO mice also exhibited increased locomotor activity and energy expenditure for several hours after CNO-mediated activation of LHA Nts neurons (Fig. 28C-F). CNO-treated mice were less active during the dark cycle, presumably to compensate for increased light cycle activity, yet the locomotor activity over 24 hr was significantly increased in WT but not NtsR1KO mice (Fig. 29A-C). Energy expenditure over 24 hr was elevated in both WT and NtsR1KO mice (Fig. 29D, E, p=0.051 for NtsR1KO group), indicating that that activation of LHA Nts neurons enhances caloric usage in an NtsR1-independent manner. Given the increased energy demands induced by activation of LHA Nts neurons, we examined whether WT or NtsR1KO mice consumed more calories to maintain energy balance. Interestingly, WT mice did not increase feeding after CNO injection, when they are also most active (Fig. 28G vs. C), and reduced feeding during the dark cycle resulting in decreased total food intake over 24 hr (Fig. 29F-H). By contrast, NtsR1KO mice increased their food intake immediately after CNO injection and minimally reduced it throughout the dark cycle, resulting in no feeding difference over 24 hr compared to VEH treated mice (Fig 28H, Fig 29F-H). While activation of LHA Nts neurons in WT mice suppressed feeding, increased energy expenditure and resulted in weight loss, the NtsR1KO mice retained weight due to their compensatory food intake (Fig. 28I). Collectively, these data suggest that activation of LHA Nts neurons promotes locomotor activity and energy 179 expenditure, and that Nts signaling via NtsR1 is required to concomitantly suppress feeding and promote weight loss. To verify the requirement for NtsR1 in LHA Nts-induced anorectic behavior, we pretreated WT and NtsR1KO mice with the NtsR1 antagonist SR48692 (0.3 mg/kg, i.p.), then administered VEH or CNO to activate LHA Nts neurons. The NtsR1 antagonist had no effect on the CNO-induced water intake or locomotor activity of WT and NtsR1KO mice (Fig. 30A, B, D, E), confirming that LHA Nts neurons promote these behaviors by an NtsR1-indpendent mechanism. By contrast, the LHA Nts-mediated reduction of chow intake over 24 hr was abolished in WT mice pre-treated with the NtsR1 antagonist (Fig. 30C). Thus, activation of LHA Nts neurons while NtsR1 signaling was either pharmacologically or genetically disrupted promoted increased feeding during the light cycle, perhaps to compensate for increased energy expenditure (Fig. 31A, B, Fig. 28H). These data suggest that NtsR1 is necessary for LHA Nts neurons to acutely suppress feeding. Since NtsR1 is expressed on VTA DA neurons14 and activation of LHA Nts neurons promotes DA release20, we next investigated whether DA signaling is necessary for LHA Nts-induced weight-loss behaviors. Pre-treatment with the D1 receptor antagonist SCH23390 (0.1 mg/kg i.p.) prior to activating LHA Nts neurons did not block drinking behavior in WT mice (Fig. 30G). Drinking was generally blunted in NtsR1KO mice during this experiment as they acclimated to the lixits in the metabolic cages, but the D1 receptor antagonist did not significantly block their water intake compared to PBS treatment (Fig. 30J). Together these data suggest that DA signaling via D1R is not required LHA Ntsinduced water intake. Consistent with previous work20, D1 receptor antagonism blunted LHA Nts-induced locomotor activity in WT and NtsR1KO mice (Fig. 30H, K, p=0.06 for NtsR1KO mice). D1R blockade also abolished the ability of LHA Nts neurons to suppress feeding in WT mice, similar to the effect of disrupting NtsR1 signaling (Fig. 30I, Fig. 31C, D). D1R blockade did not alter LHA Nts-induced hypophagia in NtsR1KO mice, as lack of NtsR1 on VTA DA 180 neurons likely impedes the ability of LHA Nts neurons to induce DA release to the NA. In sum, these data indicate that NtsR1 and D1 signaling are required for LHA Nts neurons to acutely restrain food intake. LHA Nts Action Promotes Sustained Weight Loss: To determine if LHA Nts action could sustain weight loss, we activated the LHA Nts neurons of WT mice in their home cages for 5 consecutive days. Similar to our previous observations, activating LHA Nts neurons nonsignificantly diminished chow intake on day 1 (Fig. 32A, p=0.14), but did not alter feeding on days 2-5. The increased error of hand-weighing chow from home cages, compared to the previous automated measurements from metabolic cages likely reduced our ability to detect significant feeding differences on day 1. Nonetheless, WT mice lost weight after the first injection of CNO and maintained that reduced body weight throughout the study (Fig. 32B). Given that acute activation of LHA Nts neurons promoted weight loss via both suppressing feeding and increasing physical activity and energy expenditure, we measured the physical activity of mice in locomotor chambers for 1 hr after their final VEH or CNO treatment on day 5. We observed a strong trend towards increased activity in CNO-treated mice (Fig. 32C, p=0.06), suggesting that chronic activation of LHA Nts neurons maintains elevated locomotor activity and presumably increased energy expenditure. Chronic activation of LHA Nts neurons also sustained vigorous water consumption throughout the study (Fig. 32D). Importantly, CNO treatment had no effect on WT or NtsR1KO mice expressing Cre-inducible channel rhodopsin (ChR) in LHA Nts neurons instead of hM3Dq-mCherry, confirming that our results are not due to off-target effects of CNO (Fig. 33). Collectively, these data demonstrate that repeated activation of LHA Nts neurons supports sustained weight loss. Role of LHA Nts Neurons in Palatable Food Intake: Next we investigated whether activating LHA Nts neurons could support weight loss in mice made obese by free access to a 181 palatable high fat (45%), high sucrose diet (HF diet). Obese WT mice expressing hM3DqmCherry in LHA Nts neurons were given daily treatments of VEH or CNO while assessed in metabolic chambers. As in normal weight mice, chronic activation of LHA Nts neurons promoted locomotor activity immediately after CNO injection, followed by a compensatory decrease in activity during the dark cycle (Fig. 34A, B). Despite a trend toward increased total locomotor activity over 5 days in CNO-treated mice (p=0.07, data not shown), chronic activation of LHA Nts neurons did not significantly increase energy expenditure in obese mice as observed in lean animals (Fig. 34C vs. Fig. 28E, F, Fig. 29D, E). Whereas activating LHA Nts neurons suppressed feeding in normal weight mice, activation in obese mice increased intake of the palatable HF diet during the light cycle with no subsequent suppression of feeding during the dark phase (Fig. 34D, E). As a result, CNO-treated mice maintained their obesity while VEHtreated controls slightly lost weight during the experiment due to handling stress (Fig. 34F). Thus, although activation of LHA Nts neurons promoted locomotor activity in obese mice it was insufficient to offset the consumption of freely available palatable food necessary to support weight loss. Prolonged obesity disrupts DA signaling8,9, hence activation of LHA Nts neurons in profoundly obese mice may not be able to appropriately modify mesolimbic DA signaling to promote weight loss behaviors. We therefore tested whether LHA Nts neurons could prevent normal weight mice from HF diet-induced weight gain. Lean chow-fed WT mice were switched to ad lib palatable HF diet followed by daily VEH or CNO treatments to activate LHA Nts neurons. Similar to obese mice, activation of LHA Nts neurons caused lean WT mice to consume significantly more palatable HF diet than VEH-treated controls (Fig. 35A, B). Although activation of LHA Nts neurons also increased locomotor activity (Fig. 35C, measured on day 5), any concomitant increase in energy expenditure was counteracted by increased food intake, 182 leading to weight gain (Fig. 35D). Thus, while activation of LHA Nts neurons suppresses chow intake, it cannot restrain consumption of freely accessible palatable foods that promote obesity. LHA Nts Neurons Suppress Motivated Feeding: VTA DA signaling modifies the motivation to obtain food rewards (e.g. “wanting”) but their hedonic value (“liking”) is mediated via separate systems7. Thus, LHA Nts neurons engaging the VTA are not anatomically positioned to modify palatability-driven intake of freely accessible food, consistent with our findings (Figs. 34 and 35). We reasoned, however, that LHA Nts neurons might suppress motivated feeding via Nts-signaling to VTA NtsR1 neurons. We therefore asked if activating LHA Nts neurons in fasted WT and NtsR1KO mice could blunt their motivation to eat once food was restored21. CNO-mediated activation of LHA Nts neurons restrained re-feeding, which persisted over 24 hr in WT mice (Fig. 36A, B, p=0.10 for NtsR1KOs at 24 hrs) and resulted in less weight regain compared to VEH-treated controls (Fig. 36C). These data demonstrate that LHA Nts neurons can suppress chow feeding and subsequent weight gain even under conditions that increase the drive to feed, such as fasting-induced weight loss22. To determine if activation of LHA Nts neurons suppresses the motivation to obtain palatable food, WT and NtsR1KO mice were trained to nose-poke (e.g. work) for sucrose pellets, and both groups learned the task equally well (Fig. 37A-D). We then examined whether VEH treatment or CNO-mediated activation of LHA Nts neurons alters the progressive ratio (PR) breakpoint for sucrose, a measure of the motivation to obtain rewards that is dependent on mesolimbic DA signaling23. Activation of LHA Nts neurons prior to testing did not alter PR breakpoints in ad lib fed (Sated) WT or NtsR1KO mice (Fig. 36D). Thus, activation of LHA Nts neurons in an energy-replete context does not suppress freely accessible (Fig. 35) or operantreinforced intake of palatable food. By contrast, activation of LHA Nts neurons after overnight fasting suppressed sucrose self-administration in WT but not NtsR1KO mice (Fig. 36D), 183 suggesting that LHA Nts action via NtsR1 reduces motivated sucrose “wanting” in the face of physiologic energy deficit. Importantly, the increased locomotor activity of mice upon activation of LHA Nts neurons did not impair their ability to obtain sucrose, as magazine entries were unaffected by CNO treatment (Fig. 37E). Together, these data provide evidence that enhanced LHA Nts neuronal activity counteracts the drive to consume palatable food in an energyrestricted state22,24, similar to what occurs with diet-induced weight loss. Aversive stimuli or stress25,26 also suppress feeding, so we investigated if activation of LHA Nts neurons promotes these conditions. First we determined whether WT and NtsR1KO mice exhibited any aversion for chambers paired with either VEH or CNO. WT mice demonstrated neither aversion nor preference for the CNO-paired chamber. By contrast, NtsR1KO mice significantly preferred the chamber associated with CNO-mediated activation of LHA Nts neurons (Fig. 38A), but CNO failed to modify place preference in mice expressing ChR in LHA Nts neurons instead of hM3Dq-mCherry (Fig. 38B). Together, these data suggest that intact LHA Nts signaling via NtsR1 does not suppress feeding due to aversion, but loss of action via NtsR1 may underlie the reward-induced feeding observed in NtsR1KO and NtsR1antagonist treated mice. Furthermore, activation of LHA Nts neurons did not increase anxietylike behavior measured via elevated plus maze (EPM), thus this circuit does not mediate stressinduced suppression of feeding (Fig. 38C, D). Lastly, because we anecdotally observed CNOinduced behaviors such as gnawing, grooming, and digging, we examined if these were directed at procuring food or toward any object, in this case a nestlet placed in the cage. Activation of LHA Nts neurons significantly increased nestlet-shredding in both WT and NtsR1KO mice (Fig. 38E, F), consistent with a role for increased DA signaling in mediating arousal behavior27. Collectively, these data indicate that LHA Nts-mediated suppression of food intake is not confounded by aversion or anxiety, and is associated with locomotor behaviors directed toward non-food objects. 184 5.4 Discussion We demonstrate that the LHA Nts circuit increases volitional activity and energy expenditure while suppressing food intake, and hence promotes dual behaviors to support weight loss. Intriguingly LHA Nts neurons modulate these behaviors via distinct mechanisms: Nts-mediated signaling via NtsR1 is required for the anorectic effects of this circuit but the locomotor behavior occurs via an NtsR1-independent mechanism. Although LHA Nts neurons cannot suppress intake of freely accessible palatable foods that promote weight gain, they do restrain feeding in fasted, weight-reduced mice that are highly motivated to eat. Taken together, these data suggest that activation of the LHA Nts circuit might be useful to maintain weight loss in the face of negative energy balance. For example, restraining the increased appetitive drive that occurs after an initial bout of weight loss may prevent weight regain and support further weight loss via continued diet and exercise. Previous work implicated a role for central Nts signaling in regulating feeding and locomotor activity via modifying DA signaling10,11,13 yet the specific neural circuits mediating Nts action remained unclear. Similar to the effects of pharmacologic Nts or NtsR1 agonists in the VTA10-12,28, we demonstrate that activation of LHA Nts neurons that project to the VTA increased physical activity while suppressing food intake in energy-replete mice. However, because LHA Nts-induced locomotor activity did not depend on intact NtsR1, and was only mildly blunted by D1R antagonism, it is likely controlled by a non-Nts signal released from LHA Nts neurons. By contrast, both NtsR1 and D1R antagonism abolished LHA Nts-induced suppression of feeding, implying obligate, overlapping roles for Nts and DA in this process. Given that LHA Nts neurons directly engage VTA NtsR1 neurons to promote DA release14,20, our data suggest that induction of DA signaling via this circuit functionally restrains feeding to potentiate weight loss. While DA release to the NA is most commonly linked to increased feeding rather than suppression, DA signaling is also necessary for the anorectic and weight-reducing effects of some medications29. 185 Thus, discrete DA circuits may differentially modify feeding, and LHA Nts-induced NtsR1 signaling may be a specific pathway to engage mesolimbic DA circuits that suppress appetite. LHA Nts neurons do not provide an absolute brake on feeding, as their activation only suppressed ad lib intake of chow but not HF diet. A possible explanation for this discrepancy is that the palatability and/or caloric value of HF diet overrides the ability of LHA Nts neurons to restrain intake. Mechanistically this may signify that LHA Nts neurons do not modify the opioid signaling systems that encode food “liking” that can drive overconsumption7. Intake of sucrose or fat increases DA release independently of taste, and this post-ingestive DA signaling is important for coordinating caloric intake with energy demands45,46. Thus, it is possible that activating LHA Nts modifies downstream mesolimbic DA circuits in a way that renders them incapable of sensing post-ingestive cues, causing overconsumption of calorie-dense HF diet, whereas post-ingestive DA signaling may play less of a role during intake of less palatable chow. DA signaling also modifies the motivation to work for food rewards7 and hence may be important for coordinating the motivation to consume food necessary for survival. For example, the changes in circulating ghrelin and leptin during food-deprivation modulate VTA DA signaling to increase the motivation to eat and restore energy balance30-32. Anorectic signals thus more profoundly diminish feeding in fasted animals with an elevated drive to eat, similar to the effect of exogenous leptin treatment, which reduces food intake more effectively in fasted state21,22. Similarly, LHA Nts neurons robustly suppress fasting-induced chow intake and weight regain, and may in fact contribute to leptin-mediated suppression of feeding19. Interestingly, activation of LHA Nts neurons only suppressed DA-dependent operant responding for palatable food in fasted WT mice, but not sated mice. Since feeding suppression is more easily detected when levels of food “wanting” are high (as during fasting), our finding in consistent with the possibility that LHA Nts neurons suppress feeding by reducing the “wanting” of food. The failure of LHA 186 Nts neurons to suppress operant palatable intake in the absence of NtsR1, however, supports an essential role for Nts action via NtsR1 in limiting both ad lib and motivated feeding. Some LHA Nts-mediated behaviors, including water intake and locomotor activity were not different between WT and NtsR1KO mice, suggesting that they occur via NtsR1-indpendent mechanisms. Nts may still contribute to these behaviors via engaging NtsR2, which is predominantly expressed by astrocytes in the VTA (Woodworth et al., unpublished). Glia can regulate neuronal signaling33, so the possibility of LHA Nts signaling acting through NtsR2 to alter behavior deserves consideration in the future. Alternately, signals other than Nts that are released from LHA Nts neurons might mediate drinking and locomotor activity. For example, at least some LHA Nts neurons contain GABA34, which is presumably co-released with Nts upon DREADD-induced activation. Similar to LHA Nts action, optogenetic activation of LHA GABA neurons increases NA DA release35, water intake and non-appetitive gnawing36,37. Conversely, optogenetic or pharmacogenetic activation of LHA GABA neurons increases feeding and either does not affect or suppresses locomotor activity25,36,37, whereas LHA Nts activation suppresses chow feeding and increases locomotor activity. Furthermore, LHA GABA neurons that project to the VTA promote conditioned reward35 whereas LHA Nts neurons only induce such reward in the absence of NtsR1. These discrepancies suggest that LHA Nts neurons do not fully overlap with the population of LHA GABA neurons, and that perhaps only a subset of LHA Nts neurons are also GABAergic. Within this putative subset of LHA Nts-GABA neurons, the loss of Nts signaling via NtsR1 might bias GABAergic actions in the VTA, and could explain our observation that activation of LHA Nts neurons in NtsR1KO mice causes conditioned place preference similar to that induced by the activation of LHA GABA neurons. Similarly, hypothalamic Nts is reduced in obesity, and this might also favor GABAergic signaling that potentiates feeding and weight gain, as observed in rodents with experimentally activated LHA GABA neurons25,36,37. Alternately, some LHA Nts neurons may co-express glutamate38 and the phenotype of LHA Nts 187 activation partially overlaps with activation of all LHA glutamatergic neurons as well. In particular, LHA glutamatergic neurons suppress feeding in food-restricted animals39, similar to LHA Nts neurons. By contrast, while LHA Nts activation is neither rewarding nor aversive, LHA glutamatergic stimulation produces aversion39,40, which again supports the idea that Nts is coexpressed on some, but not all LHA glutamatergic neurons. Collectively, we hypothesize that Nts is co-expressed on subsets of glutamatergic and GABAergic LHA neurons, and stimulation of the LHA Nts population as a whole promotes a repertoire of beneficial behaviors that may promote weight loss. LHA Nts neurons may also differentially control locomotor, drinking and feeding behaviors is via distinct projections. Since LHA Nts neurons robustly project to the VTA, where NtsR1 is expressed almost exclusively on DA neurons, the NtsR1-dependent suppression of feeding is likely mediated via LHA Nts inputs to the VTA. Determining whether VTA-projecting LHA Nts neurons co-release glutamate, GABA, or a combination of both may provide insight into the control of non-feeding behaviors. For instance, LHA Nts neurons might co-release GABA onto VTA GABA interneurons, thereby disinhibiting DA neurons and promoting DA signaling35. LHA Nts neurons also project locally to inhibit neighboring orexin (OX) neurons19,41, which could decrease food intake, especially in a food-restricted state. The mechanism of this, however, is likely independent of NtsR1 signaling given the lack of NtsR1 expression in the LHA, and could explain why both WT and NtsR1KO mice have suppressed food intake with LHA Nts activity in a fasted state. Our data thus reveal that LHA Nts neurons are important mediators of central anorectic Nts action via NtsR1. Future work will be key to define the additional signaling mechanisms by which LHA Nts neurons support weight loss behaviors. 188 5.5 Methods Reagents: CNO was obtained from the NIH as part of the Rapid Access to Investigative Drug Program funded by the NINDS. 40x CNO stock solutions were made by diluting with PBS/10% beta-cyclodextrin (Sigma). VEH was PBS. SR48690 was purchased from Sigma and 20x stock solutions were made in 1% TWEEN. SCH23390 was also purchased from Sigma and 50x stock solutions were made in PBS. All stock solutions were aliquoted and stored at -20 until use. Animals: Mice were bred and housed in a 12h light/12h dark cycle and cared for by Campus Animal Resources (CAR) at Michigan State University. Animals had ad lib access to chow (Teklad 7913) and water unless otherwise noted. All animal protocols were approved by the Institutional Animal Care and Use Committee (IACUC) at Michigan State University. Generation of NtsR1IRES-Cre Knock-In Mice: An NtsR1IRES-Cre targeting vector was generated by inserting an IRES-Cre sequence between the stop codon and the polyadenylation site of the sequence encoding the 3’ end of the mouse Ntsr1 gene, with an frt-flanked Neo cassette inserted upstream of the IRES-Cre. The linearized targeting vector was electroporated into mouse R1 embryonic stem (ES) cells (129sv background) and cells were selected with G418. DNA from ES cell clones was analyzed via qPCR for loss of homozygosity using Taqman primer and probes for the genomic Ntsr1 insertion site (Forward: TCT GAT GTT GGA CTT GGG TTC, Reverse: TCT GAT GTT GGA CTT GGG TTC, Probe: TCT GAT GTT GGA CTT GGG TTC). NGF was used as a copy number control42. Putative positive ES clones were expanded, confirmed for homologous recombination by Southern blot and injected into mouse C57BL/6 blastocysts to generate chimeras. Chimeric males were mated with C57BL/6 females (Jackson Laboratory), and germline transmission was determined initially via progeny coat color, then confirmed via conventional PCR for IRES-Cre. 189 Breeding and Genotyping: To generate breeders, heterozygous Ntscre mice19 (Jackson stock #017525) were mated to NtsR1KO+ (Jackson stock #005826) mice, and progeny with the genotypes NtsCre;NtsR1KO+ and ++; NtsR1KO+ were subsequently mated to generate NtsCre;++ and NtsCre;NtsR1KOKO study animals. To maximize animal usage, study animals also arose from the following crosses: NtsCre:++ X WT or NtsCre;NtsR1KOKO X ++;NtsR1KOKO. To visualize NtsR1 neurons, NtsR1IRES-Cre mice were crossed to homozygous Rosa26EGFP-L10a mice and progeny heterozygous for both alleles were studied. Genotyping was performed with standard PCR using the following primer sequences: Ntscre: common forward: 5' ATA GGC TGC TGA ACC AGG AAC, reverse: 5' CCA AAA GAC GGC AAT ATG GT and WT reverse: 5’ CAA TCA CAA TCA CAG GTC AAG AA. Rosa26EGFP-L10a : mutant forward: 5’ TCT ACA AAT GTG GTA GAT CCA GGC, WT forward: 5’ GAG GGG AGT GTT GCA ATA CC and common reverse: 5’ CAG ATG ACT ACC TAT CCT CCC. NtsR1KO: Forward: CTC TAA TGT GCC ACA GCT CAG AGA G, common: CAG CAA CCT GGA CGT GAA CAC TGA C, Reverse: CCA AGC GGC TTC GGC CAG TAA CGT T NtsR1IRES-Cre: Forward: GGA CGT GGT TTT CCT TTG AA, Reverse: AGG CAA ATT TTG GTG TACG G. Surgery: At 8-12 weeks of age, male NtsCre;++ and NtsCre:NtsR1KOKO mice received a pre-surgical injection of carprofen (5mg/kg s.c.) and were anesthetized with 3-4% isoflurane/O2 in an induction chamber before being placed in a stereotaxic frame (Kopf). Under 1-2% isoflurane, access holes were drilled in the skull allowing a guide cannula with stylet (PlasticsOne) to be lowered into the brain target area. Mice were injected bilaterally with either 300 uL of AAV-hM3Dq-mCherry or AAV-ChR-mCherry (UNC vector core) in the LHA (AP: -1.34, ML: -1.05, DV: -5.2 ). After 5 min to allow for virus absorption, the injector and cannula were removed from the skull and the incision was closed using Vet Bond. Analysis began 1-2 weeks after recovery. Mice were only included in the final study data if injections were confined to the 190 LHA on both sides. Approximately 90% of animals included in the study were bilaterally targeted; however in 10% of cases, animals with robust unilateral targeting were included in the study if CNO injection induced >1mL of water consumption, as analysis of several cohorts revealed this as a reliable indicator of LHA Nts targeting. To visualize NtsR1 neurons, adult male and female NtsR1Cre;GFP mice were bilaterally injected with 1uL FlpO adenovirus (Vector Biolabs) into the lateral ventricles (A/P: -0.22, M/L: +/- 1.0, D/V: -2.0). Mice were perfused 10 days after surgery to permit sufficient time for FlpO-mediated excision of the frt-flanked Neo cassette and GFP expression. Metabolic Analysis: At 1-2 weeks post-surgery, mice were placed in TSE cages for metabolic phenotyping (PhenoMaster, TSE Systems). After 24-48 hours of acclimation, mice were i.p. injected 1-2 hr after the onset of the light cycle with PBS, SR48692 (0.3 mg/kg), or SCH23390 (0.1mg/kg), followed by VEH or CNO (0.3 mg/kg) 30 min later. Mice were given a 24 hr washout between PBS and antagonist injections. They were continuously monitored for food and water intake, locomotor activity, and energy expenditure. Ambient temperature was maintained at 20-23˚C and the airflow rate through the chambers was adjusted to maintain an oxygen differential around 0.3% at resting conditions. Metabolic parameters including VO2, respiratory exchange ratio, and energy expenditure were assessed via indirect calorimetry by comparing O2 and CO2 concentrations relative to a reference cage. Gene Expression: Male 12-16wk old WT (++/++) and NtsR1KO mice (NtsR1KOKO) were deeply anesthetized with sodium pentobarbital and tissue from the VTA was microdissected and immediately snap frozen on dry ice and stored at -80˚C (n=10-11 per group). RNA was extracted using Trizol (Invitrogen) and 200 ng samples were converted to cDNA using the Superscript First Strand Synthesis System for RT-PCR (Invitrogen). Sample cDNAs were analyzed in triplicate via quantitative RT-PCR for gene expression using TaqMan reagents and 191 an ABI 7500 (Applied Biosystems). With GAPDH expression as an internal control, relative mRNA expression values are calculated by the 2-∆∆Ct method. Ntsr1 expression levels in 10 of 11 NtsR1KO mice were undetectable and fold change was reported as 0 for those samples. Chronic Activation in Lean Mice: Chow-Experiments: Mice were injected with VEH or CNO between 8-9AM once daily for five consecutive days in home cages while fed ad lib standard chow (Harlan Teklad 7913). Food, water, and body weight were weighed prior to the morning injection and again between 5-6pm. On day 5, locomotor activity was assessed in CPP chambers described below. Animals were randomly assigned to one side of the chamber and were allowed to acclimate for 30 min. Following acclimation, they received VEH or CNO and locomotor activity was measured by laser beam breaks for 1 hr. The study was performed using a cross-over design such that each animal received both VEH and CNO over the course of two separate experiments, with 24 hr of rest between studies. HF-Diet Experiments: As above, mice were injected once daily with VEH or CNO between 89AM in home cages with ad lib access to high fat, high sugar diet (D12451, Research Diets). Mice received 24 hr of access to HF diet prior to the study. On day 5, locomotor activity was assessed as described above, and injections were given for 10 consecutive days to allow time for weight gain. These experiments were also performed using a cross-over design, however mice were given a two-week washout period in between studies with access only to standard chow between experiments. Activation of LHA Nts Circuit in Obese Mice: Four week old male NtsCre;++ mice were weaned onto 45% high fat, high sucrose diet (HF diet, D12451, Research Diets) to induce obesity. After 6-8 months of HF diet consumption, AAV-hM3Dq-mCherry was stereotaxically delivered in the LHA bilaterally as described above. Following a 2 wk recovery period, obese mice were then analyzed in TSE metabolic cages as described above. Following two days of 192 acclimation, mice were divided into separate treatment groups and injected with VEH or CNO for 5 consecutive days in TSE cages at 8AM each morning. Operant Testing: Based on methods by Fulton et al.21, mice were food restricted to 90% of their body weight and trained on a fixed ratio-1 (FR1) schedule until they achieved 75% response accuracy with ≥20 rewards earned on 3 consecutive days. Training sessions were terminated after 1 hr or when the animal had earned 50 rewards. Mice achieving these criteria were then switched to ad lib food and trained on an FR5 schedule for 3 consecutive days. On test days, mice were subject to a progressive ratio (PR) schedule were PR=[5e(R*0.2)]-5 with R=number food rewards earned+1. Thus, the number of correct responses needed to earn a sucrose reward increases as follows: 1, 2, 4, 6, 9, 12, 15, 20, 25, 32, 40, 50, 62, 77, 95 etc. The PR breakpoint was recorded as the highest ratio completed for each 1 hr test session. Mice were tested until they achieved stable PR which was defined as <10% variation in rewards earned over 3 consecutive sessions. Conditioned Place Preference (CPP): CPP boxes (San Diego Instruments) consisted of a box divided into two chambers with different visual and tactile cues (gray wall and smooth floor or striped wall and rough floor) separated by a small center chamber. Day 1 (Pre-test): Mice were allowed to roam freely between chambers for 15 min. After pre-test data were collected, an unbiased, counterbalanced strategy was then used to assign which chamber was paired with either VEH or CNO injection, such that approximately half the mice received CNO pairing in the preferred side and half received CNO in the non-preferred side. Days 2-5 (conditioning): Each morning, mice received an injection of VEH and were immediately placed in the VEH-paired side for 30 min. After the session, mice were returned to their housing environment to facilitate memory consolidation and 4 hours later, they were conditioned with CNO for 30 minutes in the opposite chamber. Day 6 (post-test): Mice were again allowed two 193 roam freely amongst both chambers for 15 minutes. Time-spent and locomotor activity in each side of the box was assessed by laser beam breaks and data were gathered using the manufacturer’s software. One NtsR1KO animal was excluded for displaying strong sidepreference in the pre-test (>75% of time spent on that side). Nestlet-Shredding: Immediately after receiving VEH or CNO injection, mice received a white cotton pillow (Ancare) in their home cages, which was weighed prior to placement in the cage. After 90 min, any intact remnants of the original pillow were removed from the cage and allowed to air-dry overnight. The pillow remnants were weighed the following day. Elevated Plus Maze (EPM): Mice were assessed for anxiety-like behavior using an elevated plus maze (EPM) as previously described43. Briefly, the EPM apparatus was custombuilt based on plans from ANY-maze (www.anymaze.com, Stoelting Co.) and mice were given free access to the open and closed arms for 5 minutes. Their behavior was recorded using a digital CCD camera and the percentages of time spent in the open and closed arms were analyzed using Topscan automated video tracking software (Clever Sys). Fasting-Induced Re-feeding: Chow pellets were removed from home cages around 5PM and mice were given a clean cage bottom to prevent feeding of food that may have fallen into the bedding. Mice had ad lib access to water during food-deprivation. The following morning between 8AM-9AM, fasted mice were given i.p. VEH or CNO and chow pellets were returned to the feeder. Food intake, water intake, and body weight was measured at 1 hr and 24 hr after injection. The study was performed using a cross-over design, such that half the mice received VEH or CNO, and after 3 full days of recovery from fasting, the experiment was repeated with the opposite treatment allowing each animal to serve as its own control. 194 Immunohistochemistry: Mice were treated with a lethal dose of i.p. pentobarbital followed by transcardial perfusion with 10% neutral-buffered formalin (Fisher Scientific). Brains were removed, post-fixed in 10% formalin overnight at 4˚C, dehydrated with 30% sucrose in PBS for 2-3 days, and sectioned into 30 µm slices using a sliding microtome (Leica). Brain sections were then analyzed by immunofluorescence or immunohistochemistry as previously described14,19. For characterization of NtsR1 expression, sections from NtsR1Cre;GFP mice were stained with chicken anti-GFP (1:2000, Abcam) followed by donkey anti-chicken conjugated to AlexaFluor 488 (Jackson ImmunoResearch). For DREADD studies, WT and NtsR1KO mice were treated with VEH or CNO 90 minutes prior to perfusion, and brain sections were stained for cFos (1:500, goat, Santa Cruz) with secondary detection via DAB (Life Technologies), followed by immunofluorescent detection of dsRed (1:1000, Clontech) as described above. Brains were analyzed using an Olympus BX53 fluorescence microscope outfitted with transmitted light to analyze DAB-labeled tissue, as well as FITC and Texas Red filters. Microscope images were collected using Cell Sens software and a Qi-Click 12 Bit cooled camera, and images were analyzed using Photoshop software (Adobe). Statistics: Student’s t-tests and 2-way ANOVA were calculated using Prism 6 (GraphPad). Repeated measures two-way ANOVA with Sidak post-tests was used when each animal was given both VEH and CNO, and when data from the same animals were collected at different time points. For energy expenditure data, analysis of covariance (ANCOVA) was computed in SPSS 22 (IBM). Body weight was analyzed as a covariate to correct for any inherent differences it may have on metabolism44. Data were tested for homogeneity of regression, independence of the covariate (body weight), and linearity of regression prior to running the ANCOVA. For all data, *p<0.05, **p<0.01 and ***p<0.001. (#) is used to indicate comparison to starting time line within a given group. 195 APPENDIX 196 Figure 27. Examination of the LHA NtsVTA circuit in WT and NtsR1KO mice. A) NtsCre mice with either intact or developmentally deleted NtsR1 (WT or NtsR1KO mice) were injected bilaterally in the LHA with cre-inducible AAV-hm3Dq-mCherry, permitting pharmacogenetic activation of LHA Nts neurons by CNO injection. B) i.p CNO treatment induces cFos expression in LHA Nts neurons (white arrows). C) LHA Nts neurons expressing hM3Dq-mCherry in WT and NtsR1KO mice (scale bar=200µm). D) hM3Dq-mCherrry-labeled LHA Nts cell bodies in the LHA and terminals in the VTA of WT mice. E) Ntsr1 and Ntsr2 mRNA expression in the VTA of WT and NtsR1KO mice (n=10-11 per group). Data were analyzed by unpaired-test for each gene. F) NtsR1 expression in NtsR1IRES-Cre;GFP mice in the LHA and VTA. For D and F, scale bar=100µM. f=fornix, ml=medial lemniscus. 197 Figure 28. Acute activation of LHA Nts neurons promotes energy expenditure and suppresses feeding. 198 Figure 28 (cont’d). VEH or CNO-treated WT and NtsR1KO mice were analyzed in TSE metabolic cages to understand how acute LHA Nts neuron activation alters energy balance. In A-H, each point represents 108 minutes and gray boxes denote the dark cycle. A, B) water intake, C, D) locomotor activity, E, F) energy expenditure, and G, H) feeding in WT and NtsR1KO mice. I) total change in body weight 24 hours post-injection. Data were analyzed by repeated measures two-way ANOVA. WT n=10-11; NtsR1KO n=9-10. Figure 29. Additional acute metabolic data in WT and NtsR1KO mice. Light refers to the period 9 hours post VEH or CNO injection while dark refers to data collected during dark cycle (gray background). A) Locomotor activity during the light cycle, B) dark cycle, and C) over 24 hours. D, E) Average energy expenditure over 24 hours with VEH or CNO injection. F) Weightadjusted food intake during the light cycle, G) dark cycle, and H) total over 24 hours. Bar graphs were analyzed by repeated measures two-way ANOVA, while energy expenditure was analyzed by ANCOVA. WT n=10-11; NtsR1KO n=9-10. 199 Figure 30. Acute NtsR1 or D1R blockade abolishes LHA Nts-induced suppression of feeding. Mice were pretreated with PBS, SR48692 (NtsR1 antagonist) or SCH23390 (D1R antagonist) 30 min. prior to VEH or CNO injection. Each bar reflects ingestive behavior or locomotor activity over 24 hours post VEH or CNO injection. A, B, C) Water intake, locomotor activity and feeding in WT mice pretreated with the NtsR1 antagonist. D, E, F) Water intake, locomotor activity and feeding in NtsR1KO mice pretreated with a NtsR1 antagonist. G, H, I, J, K, L) Same parameters in WT or NtsR1KO mice pretreated with a D1R antagonist (SCH23390). Data were analyzed by repeated measures two-way ANOVA. WT n=10-11; NtsR1KO n=9-10. 200 Figure 31. Food intake separated by circadian period with NtsR1 or D1R antagonists. A, B) Food intake in WT mice during light or dark phases with NtsR1 antagonist pre-treatment or C, D) D1 receptor pre-treatment. E, F) Food intake in NtsR1KO mice during light or dark with NtsR1 antagonist pre-treatment or G, H) D1R antagonist pre-treatment. Data were analyzed by repeated measures two-way ANOVA. 201 Figure 32. Chronic activation of LHA Nts neurons induces mild weight loss in chow-fed lean mice. WT and NtsR1KO mice were treated once daily with VEH or CNO for five consecutive days. A) chow intake, B) body weight, C) 1 hr locomotor activity, and D) water intake in WT mice (n=11). Data were analyzed by repeated-measures two-way ANOVA, except for locomotor activity which was analyzed by paired t-test. Figure 33. Chronic VEH or CNO injection in WTChR controls. A) Chow intake, B) body weight, C) 1 hr locomotor activity, and D) water intake in WTChR controls (n=6). Data were analyzed by repeated-measures two way ANOVA, except for locomotor activity which was analyzed by paired t-test. 202 Figure 34. Repeated activation of LHA Nts neurons does not induce weight loss in obesity. LHA Nts neurons were activated each morning for five consecutive days in dietinduced obese mice analyzed in TSE metabolic chambers. A) Daily locomotor activity during the light cycle vs. B) dark cycle. C) Plot showing average rate of energy expenditure vs. body weight. D) Food intake during light cycle compared to E) dark cycle. F) Percent original body weight over five days of chronic VEH or CNO injections. VEH n=6, CNO n=5-6. Data were analyzed by standard two-way ANOVA except for C which was analyzed by ANCOVA. (*) indicates significant difference between VEH and CNO at given time point. In F, (#) represents significant difference in body weight at days 4-5 compared to day 0 in the VEH group. 203 Figure 35. Chronic activation of LHA Nts neurons does not prevent weight gain in lean mice on HF diet. Lean WT and NtsR1KO mice were injected each morning on VEH or CNO for 10 consecutive days with ad lib access to HF diet in home cages. A) HF diet intake during the light cycle, B) total food consumed over 10 days, C) 1 hour locomotor activity, and D) body weight in WT mice (n=10). (*) represents p<0.05 between VEH and CNO from Days 4-10. (#) indicates p<0.05 for CNO time point compared to Day 1. Daily food intake and body weight were analyzed by repeated-measures two-way ANOVA. Total food intake and locomotor activity was analyzed by paired t-test. 204 Figure 36. Activation of LHA Nts neurons suppresses fasting-induced ad lib and motivated food intake. WT and NtsR1KO mice were food-deprived overnight and received VEH or CNO with food restoration the following morning. A, B) Ad lib chow re-feeding 1 and 24 hours after food restoration. C) Body weight gain during 24 hours of ad lib re-feeding after overnight food-deprivation. D) PR breakpoint for sucrose pellets after VEH or CNO injection in both fed and fasted states. WT n=9, NtsR1KO n=12, data were analyzed by repeated measures two-way ANOVA. 205 Figure 37. Learning acquisition during operant training. WT and NtsR1KO mice were trained to nose-poke for sucrose pellets on an FR1 schedule until they earned >20 rewards with 75% accuracy for 3 consecutive days. A, B, C) No differences were detected in total number of active and inactive nose pokes, accuracy, and number of magazine during FR1 training. D) WT and NtsR1KO mice required a similar number of training sessions to learn the task. E) CNO treatment did not significantly increase the number of magazine entries during PR testing. WT n=9; NtsR1KO n=12. Data were analyzed by repeated-measures two-way ANOVA (A,B,C,E) or unpaired t-test (D). 206 Figure 38. Assessment of reward, anxiety, and repetitive behaviors associated with LHA NtsNtsR1 circuit activation. CPP was used assess whether activation of LHA Nts neurons could be either positively or negatively reinforcing. A) Time-spent in the CNO and VEH-paired sides during the pre- and post-tests reveals that activation of LHA Nts neurons is rewarding in NtsR1KO (n=15) but not WT (n=17) mice. B) No differences were detected in WTChR (n=12) and NtsR1KOChR (n=10) controls. Data were analyzed by two-way ANOVA. Anxiety-like behavior was assessed via EPM and no differences were detected in open arm C) entries or D) time spent (WT+VEH n=10, WT+CNO n=10, NtsR1KO+VEH n=6, NtsR1KO+CNO n=8). E, F) Nestlet weight 90 minutes after VEH or CNO injection in WT and NtsR1KO mice (WT+VEH n=8, WT+CNO n=10, NtsR1KO+VEH n=6, NtsR1KO+CNO n=8). Nestlet-shredding and EPM data were analyzed by standard two-way ANOVA while CPP was analyzed by repeated-measures two-way ANOVA to compare pre-test to post-test. 207 REFERENCES 208 REFERENCES 1 Khera, R. et al. Association of Pharmacological Treatments for Obesity With Weight Loss and Adverse Events: A Systematic Review and Meta-analysis. JAMA : the journal of the American Medical Association 315, 2424-2434, doi:10.1001/jama.2016.7602 (2016). 2 Wing, R. R. & Hill, J. O. Successful weight loss maintenance. Annual review of nutrition 21, 323-341, doi:10.1146/annurev.nutr.21.1.323 (2001). 3 Sumithran, P. et al. Long-term persistence of hormonal adaptations to weight loss. The New England journal of medicine 365, 1597-1604, doi:10.1056/NEJMoa1105816 (2011). 4 Zhou, Q. Y. & Palmiter, R. D. Dopamine-deficient mice are severely hypoactive, adipsic, and aphagic. Cell 83, 1197-1209 (1995). 5 Ahn, S. & Phillips, A. G. Dopaminergic correlates of sensory-specific satiety in the medial prefrontal cortex and nucleus accumbens of the rat. The Journal of neuroscience : the official journal of the Society for Neuroscience 19, RC29 (1999). 6 Hernandez, L. & Hoebel, B. G. Feeding and hypothalamic stimulation increase dopamine turnover in the accumbens. Physiology & behavior 44, 599-606 (1988). 7 Berridge, K. C., Ho, C. Y., Richard, J. M. & DiFeliceantonio, A. G. The tempted brain eats: pleasure and desire circuits in obesity and eating disorders. Brain research 1350, 43-64, doi:10.1016/j.brainres.2010.04.003 (2010). 8 Stice, E., Spoor, S., Bohon, C. & Small, D. M. Relation between obesity and blunted striatal response to food is moderated by TaqIA A1 allele. Science (New York, N.Y.) 322, 449-452, doi:10.1126/science.1161550 (2008). 9 Johnson, P. M. & Kenny, P. J. Dopamine D2 receptors in addiction-like reward dysfunction and compulsive eating in obese rats. Nature neuroscience 13, 635-641, doi:10.1038/nn.2519 (2010). 10 Cador, M., Kelley, A. E., Le Moal, M. & Stinus, L. Ventral tegmental area infusion of substance P, neurotensin and enkephalin: differential effects on feeding behavior. Neuroscience 18, 659-669 (1986). 11 Kelley, A. E., Cador, M., Stinus, L. & Le Moal, M. Neurotensin, substance P, neurokininalpha, and enkephalin: injection into ventral tegmental area in the rat produces differential effects on operant responding. Psychopharmacology 97, 243-252 (1989). 12 Kalivas, P. W., Burgess, S. K., Nemeroff, C. B. & Prange, A. J., Jr. Behavioral and neurochemical effects of neurotensin microinjection into the ventral tegmental area of the rat. Neuroscience 8, 495-505 (1983). 209 13 Feifel, D., Goldenberg, J., Melendez, G. & Shilling, P. D. The acute and subchronic effects of a brain-penetrating, neurotensin-1 receptor agonist on feeding, body weight and temperature. Neuropharmacology 58, 195-198, doi:10.1016/j.neuropharm.2009.07.001 (2010). 14 Opland, D. et al. Loss of neurotensin receptor-1 disrupts the control of the mesolimbic dopamine system by leptin and promotes hedonic feeding and obesity. Molecular metabolism 2, 423-434, doi:10.1016/j.molmet.2013.07.008 (2013). 15 Remaury, A. et al. Targeted inactivation of the neurotensin type 1 receptor reveals its role in body temperature control and feeding behavior but not in analgesia. Brain research 953, 63-72 (2002). 16 Beck, B., Stricker-Krongrad, A., Burlet, A., Nicolas, J. P. & Burlet, C. Changes in hypothalamic neurotensin concentrations and food intake in rats fed a high fat diet. International journal of obesity and related metabolic disorders : journal of the International Association for the Study of Obesity 16, 361-366 (1992). 17 Brown, J. A., Woodworth, H. L. & Leinninger, G. M. To ingest or rest? Specialized roles of lateral hypothalamic area neurons in coordinating energy balance. Frontiers in systems neuroscience 9, 9, doi:10.3389/fnsys.2015.00009 (2015). 18 Brown, J. A. et al. Loss of Action via Neurotensin-Leptin Receptor Neurons Disrupts Leptin and Ghrelin-Mediated Control of Energy Balance. Endocrinology, doi:10.1210/en.2017-00122 (2017). 19 Leinninger, G. M. et al. Leptin action via neurotensin neurons controls orexin, the mesolimbic dopamine system and energy balance. Cell metabolism 14, 313-323, doi:10.1016/j.cmet.2011.06.016 (2011). 20 Patterson, C. M. et al. Ventral tegmental area neurotensin signaling links the lateral hypothalamus to locomotor activity and striatal dopamine efflux in male mice. Endocrinology 156, 1692-1700, doi:10.1210/en.2014-1986 (2015). 21 Sharma, S., Hryhorczuk, C. & Fulton, S. Progressive-ratio responding for palatable highfat and high-sugar food in mice. Journal of visualized experiments : JoVE, e3754, doi:10.3791/3754 (2012). 22 Ahima, R. S. et al. Role of leptin in the neuroendocrine response to fasting. Nature 382, 250-252, doi:10.1038/382250a0 (1996). 23 Salamone, J. D., Correa, M., Farrar, A. & Mingote, S. M. Effort-related functions of nucleus accumbens dopamine and associated forebrain circuits. Psychopharmacology 191, 461-482, doi:10.1007/s00213-006-0668-9 (2007). 24 van der Plasse, G. et al. Modulation of cue-induced firing of ventral tegmental area dopamine neurons by leptin and ghrelin. International journal of obesity (2005) 39, 17421749, doi:10.1038/ijo.2015.131 (2015). 25 Jennings, J. H. et al. Visualizing hypothalamic network dynamics for appetitive and consummatory behaviors. Cell 160, 516-527, doi:10.1016/j.cell.2014.12.026 (2015). 210 26 Stamatakis, A. M. et al. A unique population of ventral tegmental area neurons inhibits the lateral habenula to promote reward. Neuron 80, 1039-1053, doi:10.1016/j.neuron.2013.08.023 (2013). 27 Berridge, C. W. Neural substrates of psychostimulant-induced arousal. Neuropsychopharmacology : official publication of the American College of Neuropsychopharmacology 31, 2332-2340, doi:10.1038/sj.npp.1301159 (2006). 28 St-Gelais, F., Legault, M., Bourque, M. J., Rompre, P. P. & Trudeau, L. E. Role of calcium in neurotensin-evoked enhancement in firing in mesencephalic dopamine neurons. The Journal of neuroscience : the official journal of the Society for Neuroscience 24, 2566-2574, doi:10.1523/jneurosci.5376-03.2004 (2004). 29 Kalyanasundar, B. et al. D1 and D2 antagonists reverse the effects of appetite suppressants on weight loss, food intake, locomotion, and rebalance spiking inhibition in the rat NAc shell. Journal of neurophysiology 114, 585-607, doi:10.1152/jn.00012.2015 (2015). 30 Branch, S. Y. et al. Food restriction increases glutamate receptor-mediated burst firing of dopamine neurons. The Journal of neuroscience : the official journal of the Society for Neuroscience 33, 13861-13872, doi:10.1523/jneurosci.5099-12.2013 (2013). 31 Wilson, C., Nomikos, G. G., Collu, M. & Fibiger, H. C. Dopaminergic correlates of motivated behavior: importance of drive. The Journal of neuroscience : the official journal of the Society for Neuroscience 15, 5169-5178 (1995). 32 Larder, R. & O'Rahilly, S. Shedding pounds after going under the knife: guts over glorywhy diets fail. Nature medicine 18, 666-667, doi:10.1038/nm.2747 (2012). 33 Perea, G., Navarrete, M. & Araque, A. Tripartite synapses: astrocytes process and control synaptic information. Trends in neurosciences 32, 421-431, doi:10.1016/j.tins.2009.05.001 (2009). 34 Leinninger, G. M. et al. Leptin acts via leptin receptor-expressing lateral hypothalamic neurons to modulate the mesolimbic dopamine system and suppress feeding. Cell metabolism 10, 89-98, doi:10.1016/j.cmet.2009.06.011 (2009). 35 Nieh, E. H. et al. Inhibitory Input from the Lateral Hypothalamus to the Ventral Tegmental Area Disinhibits Dopamine Neurons and Promotes Behavioral Activation. Neuron 90, 1286-1298, doi:10.1016/j.neuron.2016.04.035 (2016). 36 Navarro, M. et al. Lateral Hypothalamus GABAergic Neurons Modulate Consummatory Behaviors Regardless of the Caloric Content or Biological Relevance of the Consumed Stimuli. Neuropsychopharmacology : official publication of the American College of Neuropsychopharmacology 41, 1505-1512, doi:10.1038/npp.2015.304 (2016). 37 Nieh, E. H. et al. Decoding neural circuits that control compulsive sucrose seeking. Cell 160, 528-541, doi:10.1016/j.cell.2015.01.003 (2015). 38 Kempadoo, K. A. et al. Hypothalamic neurotensin projections promote reward by enhancing glutamate transmission in the VTA. The Journal of neuroscience : the official 211 journal of the Society for Neuroscience 33, 7618-7626, doi:10.1523/jneurosci.258812.2013 (2013). 39 Jennings, J. H., Rizzi, G., Stamatakis, A. M., Ung, R. L. & Stuber, G. D. The inhibitory circuit architecture of the lateral hypothalamus orchestrates feeding. Science (New York, N.Y.) 341, 1517-1521, doi:10.1126/science.1241812 (2013). 40 Stamatakis, A. M. et al. Lateral Hypothalamic Area Glutamatergic Neurons and Their Projections to the Lateral Habenula Regulate Feeding and Reward. The Journal of neuroscience : the official journal of the Society for Neuroscience 36, 302-311, doi:10.1523/jneurosci.1202-15.2016 (2016). 41 Goforth, P. B., Leinninger, G. M., Patterson, C. M., Satin, L. S. & Myers, M. G., Jr. Leptin acts via lateral hypothalamic area neurotensin neurons to inhibit orexin neurons by multiple GABA-independent mechanisms. The Journal of neuroscience : the official journal of the Society for Neuroscience 34, 11405-11415, doi:10.1523/jneurosci.516713.2014 (2014). 42 Soliman, G. A. et al. A simple qPCR-based method to detect correct insertion of homologous targeting vectors in murine ES cells. Transgenic research 16, 665-670, doi:10.1007/s11248-007-9110-2 (2007). 43 Eagle, A. L. et al. Experience-Dependent Induction of Hippocampal DeltaFosB Controls Learning. The Journal of neuroscience : the official journal of the Society for Neuroscience 35, 13773-13783, doi:10.1523/jneurosci.2083-15.2015 (2015). 44 Tschop, M. H. et al. A guide to analysis of mouse energy metabolism. Nature methods 9, 57-63, doi:10.1038/nmeth.1806 (2012). 45 Ferreira, J. G., Tellez, L. A., Ren, X., Yeckel, C. W. & de Araujo, I. E. Regulation of fat intake in the absence of flavour signalling. The Journal of physiology 590, 953-972, doi:10.1113/jphysiol.2011.218289 (2012). 46 de Araujo, I. E. et al. Food reward in the absence of taste receptor signaling. Neuron 57, 930-941, doi:10.1016/j.neuron.2008.01.032 (2008). 212 Chapter 6. Summary, Discussion, and Translational Implications 6.1 Summary of Dissertation The goal of this thesis was to understand how Nts engages mesolimbic DA signaling to modify body weight. In Chapter 2, we established the sources of endogenous Nts input to the VTA, and found that it originates from neurons of the LHA, POA, and NA. This anatomical framework was necessary to reveal which brain regions can coordinate Nts release to the VTA, and hence may utilize Nts signaling to modify energy balance. Next, in Chapter 3, we defined which VTA cells express NtsRs and thus are capable of intercepting Nts to modify DA signaling. We determined that NtsR1 is the predominant Nts receptor isoform on VTA DA neurons, while NtsR2 is instead expressed on glial cells. Furthermore, the subset of VTA DA neurons that express NtsR1 project heavily to the NA and not the PFC. Collectively, these data reveal the cellular mechanism for Nts to engage mesolimbic DA signaling: via binding to NtsR1-expressing VTA DA neurons that exclusively project to, and can release DA into, the NA. We then tested whether these VTA NtsR1-expressing DA neurons are essential for energy balance by genetically ablating them in Chapter 4. Indeed, loss of VTA NtsR1-DA neurons led to chronic hyperactivity and elevated energy expenditure that impaired age- or diet-related weight gain compared to intact controls. Additionally, loss of VTA NtsR1-DA neurons disrupted coordination of motivated feeding behavior with peripheral energy cues, indicating that this VTA DA neuronal subset may serve to adjust feeding behavior in response to peripheral energy status. Finally in Chapter 5, we put these data together to investigate whether LHA Nts neurons that project to the VTA are endogenous mediators of feeding and locomotor behaviors via Nts signaling to NtsR1. Using DREADD technology, we found that activation of LHA Nts neurons suppresses chow feeding in both fed and fasted states via NtsR1 and DA-dependent mechanisms. LHA Nts neurons also promoted increased physical activity and energy expenditure independently of NtsR1. The dual anorectic and activity-enhancing effects of LHA Nts neurons supported both 213 short and long-term weight loss, but palatability/caloric value can subvert the anorectic effects of the LHA Nts circuit. Importantly, LHA Nts action suppressed motivated sucrose selfadministration in hungry animals via an NtsR1-dependent mechanism, which coincides with the ability of Nts to modulate mesolimbic DA signaling via NtsR1, and the specific requirement of mesolimbic DA signaling for reward “wanting”, not ad libitum feeding (discussed in Chapter 1). Collectively, this work defines a neuronal circuit whereby Nts input from the LHA modifies mesolimbic DA via NtsR1 and restrains feeding in the face of negative energy balance, which may help initiate and maintain weight loss (summarized in Fig. 39). 6.2 Discussion 6.2.1 Technical Considerations of Transgenic vs. Knock-in Models to Study VTA NtsR1 Neurons When these studies were initiated, the only Cre-driver line available to study NtsR1 was a commercially-available NtsR1Cre transgenic model. We therefore began using transgenic NtsR1Cre mice to determine the functional requirement for VTA NtsR1 neurons in Chapter 4, while simultaneously working to generate a knock in NtsR1Neo-Cre model. Our rationale for developing a knock-in model was based on their proven utility for reporting endogenous gene expression, particularly for lowly-expressed transcripts such as receptors1,2. Transgenic models, however, come with inherent limitations; due to the transgene (i.e. Cre) being inserted randomly into the genome, expression levels tend to be low and may vary across generations3. Indeed, when bred to a Cre-inducible GFP reporter, the transgenic NtsR1Cre model reported 5% of VTA DA neurons co-expressing NtsR1 (Fig. 18) and the overall number of GFP-labeled neurons was much less than expected compared to Ntsr1 ISH data from the Allen Brain atlas4. By contrast, subsequent examination of the NtsR1Neo-Cre knock-in mice revealed that 70% of VTA DA neurons co-expressed NtsR1 (Fig. 14), more consistent with the Allen Brain distribution of 214 Ntsr1. Naturally, labeling 5% vs. 70% of cells is an appreciable difference, and suggests that NtsR1Neo-Cre knock-in mice are more advantageous for detecting NtsR1 neurons via breeding to Cre-inducible mouse lines. However, we observed labeling of far more than 5% of DA neurons in the transgenic NtsR1Cre model after injection of Cre-inducible viral reporters (Fig. 26). We thus suspect that Cre expression is quite low in most of the NtsR1-expressing cells of the transgenic model and as a result, only a few NtsR1 neurons have sufficient amounts of Cre protein to induce recombination of a Cre-inducible GFP allele necessary to be labeled (5% of TH+ neurons, Fig. 18). By contrast, injection of viral vectors leads to a much higher amount of Cre-inducible reporter in cells, requiring a lower threshold of Cre expression to induce recombination. Thus, transgenic NtsR1Cre mice with very low levels of Cre expression are more likely to exhibit Cre-mediated recombination and be labeled after treatment with viral vectors (Fig. 26). Despite the discrepancy between the transgenic and knock-in models when bred to reporter lines, we suspect that injection of the AAV-DTA targeted a substantial number of the true NtsR1+ population (i.e. closer to 70% rather than 5% of TH+ neurons). This is further supported by the anterograde tracing data, which was largely consistent between the transgenic and knock-in models (Fig. 15 vs. Fig. 19), with the knock-in model revealing some additional minor projections to regions outside the ventral striatum that may not have been appreciable in the transgenic line. Going forward, it will be critical to validate the behavioral data in Chapter 4 using Cre-mediated ablation or inhibitory chemogenetics in NtsR1Neo-Cre knock-in mice, which reliably detect the full extent of NtsR1 neurons. 6.2.2 Phenotype of Mice Lacking VTA NtsR1 Neurons A prominent feature of mice with genetically ablated NtsR1 neurons (Chapter 4) was their persistently elevated volitional locomotor activity. This behavior, and the increased energy expenditure to support it, is likely what protected these mice from weight gain, as their food intake was either unchanged or slightly increased compared to controls with normal activity (Fig. 215 20). A limitation of ablating neuronal populations is that the entire neuron, including all its neurotransmitters and receptors are abolished, thus resultant behaviors cannot be reliably attributed to one particular signal originating from the neuron. Therefore, we cannot conclude that the hyperactivity or uncoupling of energy balance is due to loss of NtsR1, and Nts action, per se. We speculate that the hyperactivity is due to compensatory remodeling of mesolimbic DA circuits that occurs independently of NtsR1. Loss of substantial mesolimbic DA terminals may have potentiated increased DA release and turnover from the remaining intact terminals, causing the DA system to constantly function on overdrive, similar to the action of psychostimulants. In fact, multiple colleagues remarked that the ablated mice look similar to normal mice treated with cocaine, which is characterized by inhibition of DA reuptake and persistently high levels of DA in the synapse. We suspect that if ablated mice were not hyperactive, they may instead have overconsumed palatable food and become overweight due to their impaired ability to respond to satiety cues, an effect that likely is dependent on NtsNtsR1 signaling. In support of this hypothesis, mice constitutively lacking NtsR1 have intact VTA DA neurons, but exhibit an impaired anorectic response to leptin and are prone to weight gain on HF diet, indicating that NtsR1 is required for adjusting feeding in response to satiety cues5,6. Loss of neurons expressing NtsR1 selectively in the VTA produced a similar disruption in coordinated feeding behavior, however the increased energy expenditure may have masked any susceptibility to weight gain. In the future, deletion of NtsR1 itself from the VTA, not the entire neuron, will be essential to verifying the necessity of VTA Nts-NtsR1 signaling in coordination of energy balance. Developing floxed NtsR1 mice, and injecting them in the VTA with Cre-inducible reagents to delete NtsR1 from adult VTA NtsR1 neurons could accomplish this, and holds promise to reveal the physiological role of NtsR1 in the adult brain. 216 6.2.3 Chemogenetic Activation of LHA Nts Neurons: Limitations and Future Directions In Chapter 5, we demonstrated that activation of LHA Nts neurons suppressed feeding via an NtsR1 and DA-dependent mechanism, suggesting this occurred via an LHA NtsVTA NtsR1 circuit. While the VTA is the most likely site of Nts-NtsR1 action, the study was designed such that all LHA Nts neurons were activated by the chemogenetic strategy, including LHA Nts neurons that may not project to the VTA. Future work will be important to provide more circuitspecific resolution to determine whether or not the anorectic effects of LHA Nts neurons are mediated at the level of the VTA or elsewhere. In addition to suppressing food intake via NtsR1, LHA Nts neurons also increased locomotor activity and energy expenditure by an NtsR1independent mechanism. Thus, these effects could be mediated by other neurotransmitters released by LHA Nts neurons such as GABA, glutamate, or other neuropeptides, to the VTA or elsewhere. To determine whether behavioral effects are specifically mediated by the VTA, CNO could be infused directly into the VTA of WT and NtsR1KO mice. Injecting drugs through cannulae would be more technically challenging than i.p. injections, especially for studies requiring mice to be treated multiple times over the course of an experiment. Additionally, generating study animals would be more difficult, as this would require both bilateral targeting of AAV to the LHA and of cannulae to the VTA. An alternative approach would be to design a twostep retrograde Cre-inducible DREADD virus that traveled monosynaptically, whereby injection into the VTA would allow the virus to travel to Nts neurons that project to the VTA and a second injection of a helper virus into the LHA would permit DREADD receptor expression only in LHA Nts neurons that project to the LHA, similar to modified rabies viruses7. While we experimentally activated all LHA Nts neurons, they may not all contribute to the regulation of energy balance. Indeed, LHA Nts neurons can be divided into functionally distinct subpopulations. For example, ~15-30% of LHA Nts neurons are regulated by leptin, while a separate group is selectively regulated by dehydration2,8,9. Thus general activation of all LHA 217 Nts neurons may produce a mixed phenotype due to the simultaneous actions of feedingspecific and drinking-specific subsets. We suspect that the anorectic effects of LHA Nts neurons are likely mediated by the subset of LHA Nts neurons that express LepRb, while the drinking behavior is induced by the dehydration-sensitive LHA Nts neurons. Since all of these neurons are activated by peripheral CNO, some of the behavioral findings may be complicated by desire to drink, since animals did not have access to water in the operant chambers or CPP boxes. Per the sucrose self-administration studies, this is likely not the case as activation of all LHA Nts neurons did not alter baseline PR breakpoint and was only effective at reducing responding in a fasted state (Fig. 36D). Animals had free access to water during the overnight fast and although they may have consumed less, it is unlikely that slight dehydration and/or desire to drink suppressed motivation for sugar pellets; it is much more likely that increased activation of the leptin-sensitive LHA Nts subset is responsible for reduced operant responding in a fasted state, similar to a peripheral injection of leptin10. From a translational perspective, an important finding was that activation of LHA Nts neurons was neither rewarding nor aversive in the presence of NtsR1 (Fig. 38A). Hence, our studies suggest that an NtsR1 agonist, preferentially delivered to the VTA, could restrain appetite in a food-deprived state to help initiate and maintain weight loss but without promoting dependence. It was vital to address the potential addictive nature of activating Nts-NtsR1 signaling, given that it increases mesolimbic DA release11,12 similar to addictive drugs like cocaine and amphetamine. Indeed, prior work suggested that pharmacologic administration of Nts to the VTA is positively reinforcing13-15, but this could have been due to high, nonphysiologic concentrations of Nts tested. A novel anti-obesity agent with abuse potential would not be clinically viable. However, our data suggest that activating LHA Nts neurons, and hence endogenous Nts release, may act via a specific subset of VTA DA neurons that either do not 218 functionally promote positive reinforcement or simply does not cause sufficient DA release to be reinforcing. We found that activation of LHA Nts neurons suppresses feeding in hungry, fasted mice. To clarify the translational potential of this circuit, a series of critical follow-up experiments will be necessary. A key future experiment would be to calorically restrict diet-induced obese mice (i.e. put them on a “diet”) and after initial weight loss, activate LHA Nts neurons in the presence of ad libitum chow or HF diet to see if increased action of the circuit can prevent weight regain in weight-reduced obese mice. This experiment mirrors the metabolic circumstances faced by obese individuals who struggle to maintain weight loss in the face of negative energy balance. If this experiment shows that LHA Nts neurons support sustained weight loss over time, then the circuit holds strong translational potential for maintaining reduced bodyweight in previously overweight individuals. It is possible however, that the LHA Nts circuitry is altered after the onset obesity, as suggested by the inability of LHA Nts neuronal activation to induce weight loss in well-fed obese mice (Chapter 5, Fig. 34). If this is the case, then the LHA Nts circuit may still hold promise in in preventing weight gain or maintain weight loss in younger individuals who have not had long-standing obesity but are at risk for continued weight gain. 6.3 Translational Implications 6.3.1 Neurotensin Circuits as a Translational Target Although this work provides preliminary evidence that LHA NtsNtsR1 signaling may be a candidate anti-obesity target, several questions and challenges need to be addressed to fully understand the function of this circuit in weight control. First, the circuit needs to be tested in the presence of additional variables representative of human populations, including gender, age, and genetic background. All studies in Chapter 5 were performed with male mice with 219 mixed genetic backgrounds (C57/Bl6 and 129 strains) and although testing occurred at a variety of ages, the mice were always injected with DREADD virus between 8-12 wk of age, thus the expression of DREADD receptors would have been similar across the lifespan. It will also be important to know whether LHA Nts or VTA NtsR1 expression varies with age or gender. We did not observe any overt differences in NtsR1 expression across gender or age (Chapter 3), but these studies were not designed to interrogate those variables. Importantly, we must understand whether the NtsNtsR1 system is anatomically present in humans, a question that is difficult to answer given the inability of immunohistochemistry to reliably detect Nts and NtsR1-expressing cell bodies. To date, one study examined Nts immunoreactivity in the human brain and reported Nts expression in both the LHA and VTA16, however more work with human tissue is necessary to confirm similarity in the LHA Nts VTA NtsR1 circuit across species. Nts has been investigated as a drug target for both schizophrenia and Parkinson’s disease. In 2004, the results of a clinical trial testing the efficacy of an NtsR1 antagonist (SR48692 or meclinertant) for schizophrenia found no differences in symptom improvement compared to placebo17. In a letter to the editor, Richelson et al. expressed that they were not surprised by the lack of efficacy of an NtsR1 antagonist and proposed instead that a brain penetrating non-peptide NtsR1 agonist which they had developed and intended to use in clinical trials would be more effective18. However, the fate of this non-peptide agonist is unclear. An Nts analogue (NT69L) and an NtsR1 agonist (PD149163) have been used routinely in animals, however data regarding testing of these agents in clinical trials is either unavailable or nonexistent. Interestingly, a conjugated Nts nanoparticle has been used to deliver neuroprotective genes specifically to DA neurons in animal models of Parkinson’s disease19,20. This approach takes advantage of the internalization of Nts-NtsR1 receptor complexes, and thus genes packaged and tagged to a modified Nts peptide are specially targeted to DA neurons, which have the high levels of NtsR1 expression. While this method proves promising in preclinical 220 studies, the nanoparticle has only been administered directly to the SN, and it is unclear whether or not the particle would have the same effect if administered orally or peripherally. While this approach could be used to enhance Nts-NtsR1 signaling specifically on DA neurons, which could have potential as an anti-obesity agent based on this dissertation, a considerable challenge would be developing a drug with minimal peripheral side effects. As discussed in Chapter 1, peripheral Nts promotes hypotension and increased absorption of dietary fat, which could provide undesirable side effects21,22. Interestingly however, Nts-NtsR1 signaling is a candidate drug target for a wide variety of cancers due to its role in tumor growth, and currently, meclinertant (the NtsR1 antagonist) is being assessed in a clinical trial for lung cancer (ClinicalTrials.gov identifier NCT00290953). Thus, although an NtsR1 agonist would be desirable for weight-loss, perhaps Nts-targeted agents developed through these pipelines may also prove useful in controlling body weight, and underscores the utility of idea-sharing across disciplines among medical professionals and scientists. 6.3.2 Final Thoughts on the Challenges of Addressing Obesity From the perspective of a future scientist and healthcare provider, the challenges posed by the obesity epidemic are humbling. The standard first-line treatment recommended by physicians is diet and exercise, despite overwhelming evidence that diets do not succeed in the long term23. Thus when patients fail to maintain weight loss, many feel tremendous guilt due to perceived lack of will-power to avoid high calorie foods. Patients are often not told that strong biological mechanisms, not personal weakness, are driving them to overconsume food and regain the weight. Biology drives behavior, and unfortunately many people falsely assume that obese individuals are too lazy or unmotivated to lose weight, attributing their excess weight to personal flaw. This is simply not the case. Perhaps better education of both patients and healthcare providers in the science of energy balance and weight loss could help alleviate the negative stereotypes and emotional guilt experienced by obese individuals. While it seems 221 clear that diets are not the answer, this leaves a giant gap in the treatment options for obesity. This is where scientists have enormous opportunity to develop better treatment options, which I hope will be fueled by the advent of optogenetic and pharmacogenetic techniques to decipher neural feeding circuits. Ultimately, addressing the multi-factorial nature of obesity will require concerted effort from medical professionals, scientists, policy-makers, and public health advocates. Although the stakes are high, the future holds promise for a solution to the obesity crisis. Figure 39. LHA Nts neurons restrain feeding and promote locomotor activity, in part via NtsR1 and the mesolimbic DA system. Enhanced activation by the LHA Nts circuit may be useful to support weight-loss in weight-reduced individuals. 222 REFERENCES 223 REFERENCES 1 Leshan, R. L., Bjornholm, M., Munzberg, H. & Myers, M. G., Jr. Leptin receptor signaling and action in the central nervous system. Obesity (Silver Spring, Md.) 14 Suppl 5, 208S212S, doi:10.1038/oby.2006.310 (2006). 2 Leinninger, G. M. et al. Leptin action via neurotensin neurons controls orexin, the mesolimbic dopamine system and energy balance. Cell metabolism 14, 313-323, doi:10.1016/j.cmet.2011.06.016 (2011). 3 Haruyama, N., Cho, A. & Kulkarni, A. B. Overview: engineering transgenic constructs and mice. Current protocols in cell biology Chapter 19, Unit 19 10, doi:10.1002/0471143030.cb1910s42 (2009). 4 Lein, E. S. et al. Genome-wide atlas of gene expression in the adult mouse brain. Nature 445, 168-176, doi:10.1038/nature05453 (2007). 5 Opland, D. et al. Loss of neurotensin receptor-1 disrupts the control of the mesolimbic dopamine system by leptin and promotes hedonic feeding and obesity. Molecular metabolism 2, 423-434, doi:10.1016/j.molmet.2013.07.008 (2013). 6 Kim, E. R., Leckstrom, A. & Mizuno, T. M. Impaired anorectic effect of leptin in neurotensin receptor 1-deficient mice. Behavioural brain research 194, 66-71, doi:10.1016/j.bbr.2008.06.024 (2008). 7 Kim, E. J., Jacobs, M. W., Ito-Cole, T. & Callaway, E. M. Improved Monosynaptic Neural Circuit Tracing Using Engineered Rabies Virus Glycoproteins. Cell reports, doi:10.1016/j.celrep.2016.03.067 (2016). 8 Brown, J. A. et al. Loss of Action via Neurotensin-Leptin Receptor Neurons Disrupts Leptin and Ghrelin-Mediated Control of Energy Balance. Endocrinology, doi:10.1210/en.2017-00122 (2017). 9 Watts, A. G., Sanchez-Watts, G. & Kelly, A. B. Distinct patterns of neuropeptide gene expression in the lateral hypothalamic area and arcuate nucleus are associated with dehydration-induced anorexia. The Journal of neuroscience : the official journal of the Society for Neuroscience 19, 6111-6121 (1999). 10 Ahima, R. S. et al. Role of leptin in the neuroendocrine response to fasting. Nature 382, 250-252, doi:10.1038/382250a0 (1996). 11 Patterson, C. M. et al. Ventral tegmental area neurotensin signaling links the lateral hypothalamus to locomotor activity and striatal dopamine efflux in male mice. Endocrinology 156, 1692-1700, doi:10.1210/en.2014-1986 (2015). 12 Leonetti, M. et al. Specific involvement of neurotensin type 1 receptor in the neurotensinmediated in vivo dopamine efflux using knock-out mice. Journal of neurochemistry 89, 16, doi:10.1046/j.1471-4159.2003.02231.x (2004). 224 13 Rouibi, K., Bose, P., Rompre, P. P. & Warren, R. A. Ventral Midbrain NTS1 Receptors Mediate Conditioned Reward Induced by the Neurotensin Analog, D-Tyr[11]neurotensin. Frontiers in neuroscience 9, 470, doi:10.3389/fnins.2015.00470 (2015). 14 Glimcher, P. W., Margolin, D. H., Giovino, A. A. & Hoebel, B. G. Neurotensin: a new 'reward peptide'. Brain research 291, 119-124 (1984). 15 Glimcher, P. W., Giovino, A. A. & Hoebel, B. G. Neurotensin self-injection in the ventral tegmental area. Brain research 403, 147-150 (1987). 16 Mai, J. K., Triepel, J. & Metz, J. Neurotensin in the human brain. Neuroscience 22, 499524 (1987). 17 Meltzer, H. Y., Arvanitis, L., Bauer, D. & Rein, W. Placebo-controlled evaluation of four novel compounds for the treatment of schizophrenia and schizoaffective disorder. The American journal of psychiatry 161, 975-984, doi:10.1176/appi.ajp.161.6.975 (2004). 18 Richelson, E., Fredrickson, P. A. & Boules, M. M. Neurotensin receptor agonists and antagonists for schizophrenia. The American journal of psychiatry 162, 633-634; author reply 635, doi:10.1176/appi.ajp.162.3.633-b (2005). 19 Espadas-Alvarez, A. J. et al. Regulation of human GDNF gene expression in nigral dopaminergic neurons using a new doxycycline-regulated NTS-polyplex nanoparticle system. Nanomedicine : nanotechnology, biology, and medicine 13, 1363-1375, doi:10.1016/j.nano.2017.02.006 (2017). 20 Gonzalez-Barrios, J. A. et al. Neurotensin polyplex as an efficient carrier for delivering the human GDNF gene into nigral dopamine neurons of hemiparkinsonian rats. Molecular therapy : the journal of the American Society of Gene Therapy 14, 857-865, doi:10.1016/j.ymthe.2006.09.001 (2006). 21 Di Paola, E. D. & Richelson, E. Cardiovascular effects of neurotensin and some analogues on rats. European journal of pharmacology 175, 279-283 (1990). 22 Li, J. et al. An obligatory role for neurotensin in high-fat-diet-induced obesity. Nature 533, 411-415, doi:10.1038/nature17662 (2016). 23 Mann, T. et al. Medicare's search for effective obesity treatments: diets are not the answer. The American psychologist 62, 220-233, doi:10.1037/0003-066x.62.3.220 (2007). 225