NEUROTENSIN-EXPRESSING LATERAL HYPOTHALAMIC NEURONS ALLEVIATE PAIN VIA NEUROTENSIN RECEPTOR SIGNALING By Rabail Khan A DISSERTATION SubmiAed to Michigan State University in parLal fulfillment of the requirements for the degree of Neuroscience - Doctor of Philosophy 2024 ABSTRACT Chronic pain and obesity frequently occur together. An ideal therapy would alleviate pain without weight gain, and most opLmally, could promote weight loss. The neuropepLde neurotensin (Nts) is implicated in reducing weight and pain, but the endogenous mechanisms underlying this physiology were unknown. We previously showed that acLvaLng lateral hypothalamic area neurons expressing Nts (LHANts neurons) suppresses feeding and promotes weight loss. Here we hypothesized that LHANts neurons are an endogenous source of Nts that can provide anLnocicepLon, and hence, that acLvaLng LHANts neurons would alleviate pain dependent on Nts signaling via NtsRs. We first used Designer Receptor Exclusively AcLvated by Designer Drug (DREADDs) to acLvate LHANts neurons in normal weight and diet-induced obese (DIO) NtsCre mice. AcLvaLng LHANts neurons had no effect on thermal pain and mechanical responses in naïve normal weight mice. By contrast, both spared nerve injury- (SNI) and complete Freund’s Adjuvant (CFA)-induced mechanical hypersensiLvity was completely reversed by DREADD ligand clozapine N-oxide (CNO)-mediated sLmulaLon of LHANts neurons compared to control naïve normal weight mice. Similarly, acLvaLng LHANts neurons had no effect on thermal pain responses in DIO mice. By contrast, obesity-induced pain hypersensiLvity as well as CFA- induced inflammatory pain was completely reversed by CNO-mediated acLvaLon of LHANts neurons compared to VEH control. However, pretreatment with the brain permeable Nts receptor pan-antagonist SR142948 (1mg/kg, i.p, 30 min before VEH/CNO) blocked CNO-mediated analgesia, indicaLng that LHANts neurons alleviate chronic pain in an Nts-dependent manner. Furthermore, Nts deleLon from the LHA by injecLng AAV-Cre into the LHA of Ntsflox/flox mice further exacerbated hyperalgesia in DIO mice compare to normal weight mice. Taken together these data suggest that augmenLng signaling via LHANts neurons may be a common acLonable target to treat comorbid obesity pain. To the courageous Afghan women advocaLng for their right to educaLon, and to my younger self for having the stubborn tenacity to see this through. iv ACKNOWLEDGEMENTS CompleLng this dissertaLon would have been impossible without the invaluable guidance of Dr. Leinninger. Since joining her lab, I've been on a conLnuous journey of learning, which has been the greatest gih. Dr. Leinninger has always been supporLve, kind, and excepLonally nurturing. I owe her graLtude for teaching me how to become a beAer mentor and for fostering my academic resilience. Her influence has been crucial in developing my scienLfic career. I extend my sincere thanks to Dr. Greg Swain and Dr. Jim Galligan for interviewing and recruiLng me from Pakistan. Their confidence in me has been deeply meaningful, and I hope I have met their expectaLons. To Dr. Laumet Geoffroy, Dr. Michelle Mazei-Robison, Dr. Alex Johnson, and Dr. Brian Gulbransen—thank you for your parLcipaLon on my dissertaLon commiAee. Your support and guidance have been invaluable. To the former members of the Leinninger lab, Dr. Patricia Perez, my excepLonal undergraduate student Hope Bemis, and current lab member Jariel Ramirez-Virella— thank you for being great companions during the challenges of the pandemic. Special thanks to Raluca Bugescu for being an excepLonally kind colleague and a friend. I also want to acknowledge Beenhwa (Grace) Lee for assisLng me with this project. A hearkelt thanks to the MSU animal care staff and vets, parLcularly BernadeAe Lake, Rebecca Winget, Chase Beard, and Brooke Steele, for their approachability and flexibility. None of this would have been possible without their support. Most importantly, I want to express my deepest graLtude to my husband, Umer Siddiqui. Your unwavering belief in me and your encouragement have been my anchor throughout graduate school. I admire your paLence and your willingness to compromise. Thank you for being an excepLonal partner. v Finally, I must thank my parents, Muhammad Asif Khan and Gulshan Jabeen, and my sisters Shehla Gul, Nadia Daud, and Maleeha Khan. Your belief in me and your encouragement were essenLal in reaching the end of this dissertaLon journey. vi TABLE OF CONTENTS LIST OF ABBREVIATIONS ............................................................................................................... vii CHAPTER 1: HUNGRY FOR RELIEF: POTENTIAL FOR NEUROTENSIN TO ADDRESS COMORBID OBESITY AND PAIN .......................................................................................................................... 1 ABSTRACT .................................................................................................................................. 1 1.1 THE COMORBIDITY OF OBESITY AND CHRONIC PAIN .......................................................... 1 1.2 NEUROTENSIN AND ITS RECEPTORS .................................................................................... 3 1.3 NEUROTENSIN’S PHYSIOLOGIC ROLES ................................................................................. 5 1.4 NEUROTENSIN IN ENERGY BALANCE ................................................................................... 6 1.5 CENTRAL NEUROTENSIN AND PAIN ................................................................................... 17 1.6 FUTURE CHALLENGES: IDENTIFICATION OF COMMON SITES TO TREAT OBESITY AND PAIN ......................................................................................................................................... 26 1.7 GOALS OF DISSERTATION ................................................................................................... 28 ACKNOWLEDGEMENTS ............................................................................................................ 34 REFERENCES ............................................................................................................................. 35 CHAPTER 2: NEUROTENSIN-EXPRESSING LATERAL HYPOTHALAMIC NEURONS ALLEVIATE NEUROPATHIC AND INFLAMMATORY PAIN VIA NEUROTENSIN RECEPTOR SIGNALING ............... 54 ABSTRACT ..................................................................................................................................... 54 2.1 INTRODUCTION .................................................................................................................. 55 2.2 RESEARCH DESIGN AND METHODS .................................................................................... 57 2.3 RESULTS .............................................................................................................................. 63 2.4 DISCUSSION ....................................................................................................................... 69 ACKNOWLEDGEMENTS ............................................................................................................ 80 REFERENCES ............................................................................................................................. 81 APPENDIX ................................................................................................................................. 87 CHAPTER 3: LATERAL HYPOTHALAMIC AREA NEURONS EXPRESSING NEUROTENSIN ALLEVIATE COMORBID OBESITY-PAIN…………………………………………………………………………………………………………91 ABSTRACT ................................................................................................................................ 91 3.1 INTRODUCTION .................................................................................................................. 92 3.2 RESEARCH DESIGN AND METHODS .................................................................................... 94 3.3 RESULTS ............................................................................................................................ 101 3.4 DISCUSSION ..................................................................................................................... 107 ACKNOWLEDGEMENTS .......................................................................................................... 120 REFERENCES ........................................................................................................................... 121 CHAPTER 4: SUMMARY, DISCUSSION, AND TRANSLATIONAL IMPLICATIONS ............................. 126 4.1 SUMMARY OF RESULTS .................................................................................................... 126 4.2 DISCUSSION ..................................................................................................................... 128 4.3 TRANSLATIONAL IMPLICATIONS OF THIS RESEARCH ....................................................... 137 REFERENCES ........................................................................................................................... 140 vii LIST OF ABBREVIATIONS AMPK AMP-AcLvaLng Protein Kinase ANOVA Analysis of Variance β-LT Beta Lactotensin CaMMK2 Ca2+/Calmodulin Kinase II CNO CAR CRH CTCF CGX CFA DIO DRG Clozapine N-Oxide Campus Animal Resources CorLcotropin-Releasing Hormone Corrected Total Cell Fluorescence Cantulakin-G Complete Freund’s Adjuvant Diet-Induced Obese Dorsal Root Ganglion DREADDs Designer Receptors Exclusively AcLvated by Designer Drugs DA FFAR4 GPCR Glp1 HFD Dopamine Free FaAy Acid Receptor 4 G-Protein Coupled Receptor Glucagon-like PepLde Receptor High Fat Diet IACUC InsLtuLonal Animal Care and Use CommiAee ICV Intracerebroventricular viii IPAC KO LHA Posterior Limb of the Anterior Commissure Knock Out Lateral Hypothalamic Area LepRb Long-form LepLn Receptor isoform LS MPP Lateral Septum Medial PreopLc Area MAPK Mitogen-AcLvated Protein Kinase MOR Nts NtsR1 NtsR2 NtsR3 NtsR4 NAc NTS PAG PVN PWL ROI RVM SDH μ-Opioid Receptor Neurotensin Neurotensin Receptor 1 Neurotensin Receptor 2 Neurotensin Receptor 3 Neurotensin Receptor 4 Nucleus Accumbens Nucleus of the Tractus Solitarius Periaqueductal Gray Paraventricular Hypothalamic Nucleus Paw Withdrawal Latency Region of Interest Rostroventral Medulla Spinal Dorsal Horn ix SCN SNc SNI VEH VTA SuprachiasmaLc Nucleus SubstanLa Nigra Pars Compacta SciaLc Nerve Injury Vehicle Ventral Tegmental Area x CHAPTER 1: HUNGRY FOR RELIEF: POTENTIAL FOR NEUROTENSIN TO ADDRESS COMORBID Authors: Rabail Khan, Geoffroy Laumet, and Gina M. Leinninger OBESITY AND PAIN This chapter is a modified version of a review arDcle accepted in 2024 in journal AppeDte. ABSTRACT Chronic pain and obesity frequently occur together. An ideal therapy would alleviate pain without weight gain, and most opLmally, could promote weight loss. The neuropepLde neurotensin (Nts) has been separately implicated in reducing weight and pain but could it be a common acLonable target for both pain and obesity? Here we review the current knowledge of Nts signaling via its receptors in modulaLng body weight and pain processing. EvaluaLng the mechanism by which Nts impacts ingesLve behavior, body weight, and analgesia has potenLal to idenLfy common physiologic mechanisms underlying weight and pain comorbidiLes, and whether Nts may be common acLonable targets for both. 1.1 THE COMORBIDITY OF OBESITY AND CHRONIC PAIN Chronic pain is a major health issue globally1,2. In the United States alone 20.5% of the adult populaLon experiences chronic pain3,4, causing substanLal economic burden5,6 and negaLvely impacLng quality of life7. Acute pain is a short-term unpleasant experience that occurs with injury and is important to prevent further Lssue damage. In contrast, chronic pain is persistent, lasLng beyond the injury healing period8–10, which can negaLvely impact individuals’ producLvity, leading to disability and financial ramificaLons. While chronic pain may arise aher injuries and surgical procedures or can be idiopathic, its prevalence strongly increases with chronic metabolic diseases, including diabetes11 and obesity12–14. This presents a significant public 1 health challenge, given that one-third of Americans are obese and at increased risk to develop chronic pain15–20. Indeed, the InsLtute of Medicine idenLfies obesity as one of five major contributors of chronic pain21. Obesity increases the likelihood of developing various chronic pain condiLons, such as diabeLc neuropathy22,23, osteoarthriLs24, as well as joint and back pain25. Although obesity and pain are comorbid, they have largely been considered as independent problems and treated separately. Few treatments have been evaluated to simultaneously address obesity and pain. Notably, life style changes such as exercise and diet- induced weight loss can stabilize pain26,27. However, sustaining weight reducLon is challenging, and even with maintained diet and exercise most individuals regain lost weight due to metabolic adaptaLons28, making them once again subject to chronic pain. Despite the high and increasing prevalence of obesity there are few effecLve pharmacologic strategies to elicit sustained weight reducLon, as needed to prevent or alleviate pain. Likewise, treatment of pain itself is challenging and the presence of obesity complicates pain treatment. A pikall of current non-opioid pain medicaLons is that they induce weight gain with prolonged use29. For example, long-term use of one such non-opioid pain therapeuLc, pregabalin is reported to increase body weight up to 11 pounds30. Opioids can be useful to treat acute pain but they are less effecLve in most chronic pain condiLons31–35. AddiLonally, their reinforcing side effects have led to over-use and abuse, and the current opioid crisis31,36. The rising prevalence in obesity-related chronic pain has only exacerbated this public health issue, with over 14% of opioids prescribed to individuals classified as obese36. As such, neither opioid nor non-opioid treatments are ideal for individuals with comorbid obesity and pain. Development of alternaLve treatment strategies has been hampered by the lack of basic understanding of the underlying neurobiology mediaLng both energy balance 2 and nocicepLon. However, an ideal therapy would be one that can alleviate pain without weight gain, and most opLmally, could promote weight loss. NeuropepLdes have emerged as important players in regulaLng either obesity or pain, with some promoLng weight loss (anorexigenic), and others providing pain relief (analgesic). Although the field has separately characterized anorexic and analgesic funcLons of certain neuropepLdes, some neuropepLdes may be beneficial in both domains, with potenLal to combat the comorbid chronic pain-obesity epidemic. One such mulLfaceted neuropepLde is neurotensin (Nts), which has been implicated as a regulator of body weight and pain in separate studies. In this review we will focus on the growing understanding of how and where Nts signals within the brain to regulate energy balance and pain, and the potenLal of modulaLng the Nts system to address comorbid obesity and pain simultaneously. 1.2 NEUROTENSIN AND ITS RECEPTORS Neurotensin (Nts) is a 13 amino acid pepLde first isolated from the bovine hypothalamus so it was iniLally characterized as a neuropepLde37–39. In situ hybridizaLon and radio- immunolabelling later revealed that Nts is widely distributed throughout the central nervous system, including the spinal cord and brain regions such as the hindbrain (specifically the nucleus of solitary tract, locus coeruleus, the dorsal raphé nucleus, and parabrachial nucleus), the midbrain (ventral tegmental area, periaqueductal gray area, substanLa nigra)40, limbic system (amygdala, nucleus accumbens, septal area, hippocampus)41–43, and the forebrain (caudate putamen, nucleus accumbens)44,45. Yet Nts is also expressed in peripheral Lssues, including the enteroendocrine N-cells of the intesLnal jejunum and ileum46,47. Nts also binds with varying affinity in the brain vs periphery suggesLng an associaLon with physiologically relevant 3 receptors48,49 and the existence of Nts receptor subtypes50,51. Subsequently four types of Neurotensin Receptors (NtsRs) have been characterized, including Neurotensin Receptors -1, -2, -3, and -4. 1.2.1 Neurotensin Receptor 1 (NtsR1) NtsR1 was discovered first when Barnard and colleagues isolated it from bovine cerebral cortex membrane52. Shortly thereaher Nakanishi and colleagues cloned NtsR1 (then referred to as NtS1) permiwng further structural characterizaLon53. NtsR1 is a seven transmembrane domain receptor that belongs to the G-protein-coupled receptor family. NtsR1 binds Nts with high affinity (Kd = 0.1-0.3 nM), and is widely distributed throughout the central and peripheral system in adult rodents and humans54. NtsR1 is found in diverse brain regions such as the ventral tegmental area (VTA), substanLa nigra (SN), suprachiasmaLc nucleus (SCN), supramammillary area, cortex, striatum, hypothalamus, and periaqueductal gray (PAG)55,56, where it is expressed by neurons. NtsR1 is also present in the small intesLne and colon although the specific cell types uLlizing NtsR1 and its funcLons there remain elusive57. NtsR1 is reported to transduce intracellular signaling via Gα q/11, Gα i/o, Gα s, and/or β-arresLns58–61, which transduce very different downstream signaling cascades. The site-specific acLons and physiological outcomes aAributed to Nts-NtsR1 signaling may vary depending on the coupled pathway, which could differ by the cell type expressing NtsR1 and/or where in the brain the cells reside. However, NtsR1 is well established to couple to Gα q/11 signaling in some specific brain regions, notably the VTA60,62. 1.2.2 Neurotensin Receptor 2 (NtsR2) NtsR2 was idenLfied shortly thereaher, cloned from an adenocarcinoma cell line in 1993. Like NtsR1, NtsR2 belongs to the G protein-coupled receptor superfamily but exhibits lower 4 binding affinity (Kd = 3.7±0.2 nM) for Nts63,64. In contrast to NtsR1’s robust expression on neurons, NtsR2 is predominantly expressed on astrocytes and epithelial cells distributed throughout the brain65,66. NtsR2 is also expressed within the cardiovascular system, brown adipose Lssue, gastrointesLnal tract67,68 and ependymal cells69–72. NtsR2 has been reported to couple to various Gα signaling pathways, which may vary depending on the cell type and region. 1.2.3 Neurotensin Receptor 3 and 4 (NtsR3/4) The receptor characterized as NtsR3 is also known as sorLlin, a receptor sorLng protein73,74. NtsR3 is structurally different from the NtsR1 and NtsR2 GPCRs in that it has single transmembrane domain and an intracellular domain with low affinity Nts binding [Kd = 10nM]73. NtsR4 was idenLfied in 2001 and is an intracellular protein related to the yeast sorLng receptors- NtsR4/sortLA. Like NtsR3, the NtsR4 is a low affinity, single transmembrane domain receptor75. 1.3 NEUROTENSIN’S PHYSIOLOGIC ROLES Neurotensin modulates a diverse array of physiological processes through acLons in different brain regions and the periphery. Pharmacological injecLon of Nts in the brain or periphery induces hypothermia76–78, vasodilaLon, hypotension37,79, hyperglycemia80,81, and muscle contracLon82,83. Nts is also linked to modulaLon of neuroendocrine funcLon, sleep and arousal84–86, body temperature, appeLte, blood pressure, reward behaviors, regulaLon of ingesLve behavior and body weight, and analgesia87–90. The roles of Nts in these aspects of physiology are yet under invesLgaLon and we point the reader to several excellent recent reviews concerning these topics90–94. Here we focus on the role of Nts in the physiology relevant to comorbid obesity and pain: how it modulates energy balance and analgesia. 5 1.4 NEUROTENSIN IN ENERGY BALANCE Living organisms have dedicated physiological systems to preserve energy balance, a criLcal factor for their survival. These systems operate through two crucial mechanisms: first, by sensing the body's energy requirements to gauge its needs, and second, by prompLng relevant behavioral responses from the brain to acquire the necessary energy (seeking and ingesLng food). The ability to adjust behaviors based on energy status is vital to prevent starvaLon during deficiency and to curb excessive intake when reserves are full. Achieving this balanced regulaLon relies on peripheral cues that communicate current energy state to the brain (such as glucose and saLety hormones), which can thereby coordinate appropriate feeding and/or energeLc responses. Hence, a well-funcLoning feedback loop between the body's metabolic condiLon and the brain's behavioral decisions is indispensable for organisms to maintain energy equilibrium and, consequently, to ensure their survival. Here we review the current knowledge on Nts’ role in metabolism, specifically on lipid absorpLon, water and food intake, energy expenditure, and body temperature that impact energy balance and body weight. Given that Nts is produced in both the brain and periphery, with disLnct acLons aAributed to these different pools of Nts, we will separately examine these sources and their effects. 1.4.1 Peripheral Neurotensin in Energy Balance 1.4.1.1 Neurotensin Facilitates Fat AbsorpSon in the Periphery The high level of Nts in plasma originates from chromaffin cells in the adrenal gland95,96 and enteroendocrine N-cells in the distal small intesLne that secrete gut hormones97,98. Given the crucial role of the intesLne for energy absorpLon, the discovery of Nts in the intesLne hinted that it might modulate body weight via modifying nutrient uptake38,99,100. Indeed, there is an increase 6 in circulaLng Nts levels following the ingesLon of fat101, which is thought to be released from the N-enteroendocrine cells102. The elevaLon in Nts is rapid (detectable in the circulaLon within 20 minutes of the meal iniLaLon) and persistent, with a steady secreLon rate persisLng for at least 180 minutes103. The current interpretaLon of these data is that Nts is released from enteroendocrine cells aher a meal to support lipid absorpLon. Indeed, knockout mice with developmental deleLon of Nts have significantly reduced intesLnal fat absorpLon. These Nts- deficient mice are even protected from developing obesity when on a high-fat diet (HFD), and exhibit aAenuated obesity-associated insulin resistance and other metabolic improvements compared to wildtype mice that develop severe diet-induced obesity and metabolic complicaLons57. CollecLvely, these findings support that peripheral Nts signaling mediates fat absorpLon that could potenLate weight gain in individuals consuming modern fat-rich diets103,104. The signaling mechanism by which intesLnal Nts mediates fat absorpLon is an acLve area of inquiry, since it could idenLfy strategies to miLgate fat uptake and weight gain. IntesLnal Nts acts through NtsR1 and NtsR3, enhancing faAy acid absorpLon by suppressing Ca2+/calmodulin kinase II complex CaMKK2-mediated AMP-acLvaLng protein kinase (AMPK) phosphorylaLon57. Conversely, increased AMPK phosphorylaLon, achieved through the inhibiLon of mitogen- acLvated protein kinase (MAPK/ERK1/2), decreases faAy acid-sLmulated Nts release. These results establish an inhibitory crosstalk between MAPK and AMPK signaling pathways downstream of free faAy acid receptor 4 (FFAR4) to regulate Nts secreLon in response to fat consumpLon102. Due to the difficulLes in labeling Nts receptors, the specific locaLon of intesLnal cells expressing NtsR1 and NtsR3 remains unclear. Future work to idenLfy these cells and to 7 understand how they work will provide vital informaLon about the fundamental control of nutrient absorpLon. Recent work has begun to invesLgate the role of peripherally-produced Nts in altered states of energy balance. Given Nts’ role in fat absorpLon, loss of Nts acLon may exacerbate the degree of weight loss in individuals with anorexia nervosa, as it reduces fat absorpLon in addiLonal to self-restricted dieLng105. On the flipside, both rodents and humans with obesity and insulin resistance have elevated circulaLng levels of stable long-form Nts (proNts)106, and lower circulaLng levels of processed (acLve) Nts aher ingesLng faAy meal57,107. A recent study demonstrated that fasLng plasma levels of both mature and pro-Nts collecLvely decrease in obese mice and humans undergoing weight loss through food restricLon. This indicates that the producLon of peripheral Nts is coordinated with body weight108. Moreover, the amount of circulaLng pro-Nts correlates with the degree of obesity and has been suggested as a potenLal diagnosLc marker for risk of developing obesity and metabolic disease57,106. InteresLngly, in obese individuals, circulaLng Nts levels are further elevated aher Roux-en-Y gastric bypass (RYGB) surgery109,110. A recent study also observed increased number of Nts-expressing cells in the small intesLne following RYGB surgery among individuals with type-2 diabetes. This increase in Nts- expressing cells is posiLvely associated with the extent of improvement in their insulin sensiLvity111. However, there is no correlaLon between circulaLng Nts levels and changes in body weight in paLents undergoing gastric banding112. Consistent with clinical findings, a recent rodent study reported elevated Nts levels in plasma and the gastrointesLnal tract aher RYGB surgery113. The simultaneous elevaLon of Nts levels and reducLon in body weight and insulin resistance suggests an important role of Nts in obesity treatment. One might quesLon, if peripheral Nts 8 increases fat absorpLon that in turn increases body weight, then how it can improve the degree of weight loss post-bariatric surgery? While most peripheral studies have focused on how Nts acts within the intesLne to modulate fat absorpLon, Nts may indirectly modulate brain signaling to alter food intake. A potenLal explanaLon for the paradox of elevated Nts post bariatric surgery and weight loss is that when peripheral Nts is present in high concentraLons it signals through the vagal nerve to acLvate hypothalamic regions, inducing a feeling of saLety via hypothalamic anorecLc pathways113. This could support reduced food intake, and accordingly, lower available fat to be absorbed by the intesLne. Indeed, circulaLng Nts can access hypothalamic regions just beyond the blood brain barrier that are important for modulaLng feeding, such as the arcuate nucleus. 1.4.1.2 Effect of Systemic Neurotensin on Food Intake Rodent studies support that peripheral Nts may impact food intake. Acute systemic injecLon of Nts modestly suppresses feeding in rodents, parLcularly during the dark phase when they consume the majority of their daily food114,115 and pretreatment with an Nts receptor antagonist (SR-142948) diminishes the anorecLc effect113. In mice with geneLc obesity (due to deleLon of the anorecLc hormone lepLn or the lepLn receptor), systemic treatment with the Nts analog NT69L reduced food intake and body weight116. Systemic Nts treatment may mimic the endogenous release and acLon of Nts in response to meals (food intake), and coincide with the inducLon of the anorexic hormone lepLn117. This effect is aAributed, in part, to elevated plasma Nts being able to pass through the blood brain barrier to access immediately adjacent hypothalamic regions that modulate feeding113. However, there are other brain areas that contribute to energy balance that lie far from the blood brain barrier, such as extended 9 hypothalamic nuclei and the mesolimbic dopamine (DA) system. Given their locaLon deeper in the brain and the short half-life of Nts in the circulaLon104, it is unlikely that peripherally produced Nts is sLll biologically acLve if and when it reaches these brain regions. Such deeper brain structures are more likely modulated by the central pool of Nts that is produced within the brain and released as a neuropepLde transmiAer to specific brain regions101,104. 1.4.2 Central Neurotensin Signaling in Energy Balance In contrast to Nts acLon in the periphery, where it promotes fat accumulaLon and weight gain, Nts serves as a saLety signal in the brain. In rodents, systemic Nts treatment causes a modest, transient suppression of feeding whereas intracerebroventricular (ICV) administraLon produces a more pronounced anorecLc effect, parLcularly in diet-induced obese mice114,115. These data suggest that at least some of the anorecLc effects of Nts are mediated via brain regions that may not have access to peripherally-produced circulaLng Nts, and perhaps are modulated by brain-produced Nts. Indeed, Nts neurons in the brain synthesize pro-Nts, and pro- hormone convertase cleaves it into a biologically acLve form118. Nts is released via dense core granules such that it can act as a pepLde transmiAer and bind to Nts receptors on other cells to modulate their acLvity and acLons. Thus, idenLfying the specific neurons that produce Nts, which cells they release it to, and their Nts receptors is essenLal to understand the underlying mechanisms governing energy homeostasis and the potent anorecLc effect of Nts in the brain. However, idenLfying endogenous Nts-expressing neurons has been technically challenging. Early studies used immunoreacLve labeling to visualize Nts neurons, yet this method solely labeled Nts fibers and was unable to idenLfy Nts-expressing soma, hindering the idenLficaLon of the neurons' origin. To address this, animals can be pre-treated with the axonal transport inhibitor 10 colchicine, enabling sufficient Nts accumulaLon in the soma for immunohistochemical detecLon40,119–122. However, the drawback of the colchicine method is that it globally impairs neuronal signaling, as it blocks neurotransmiAer release and ulLmately leads to death within less than 48 hours. In situ hybridizaLon labels Nts mRNA but this labor and cost intensive method is not opLmal for rouLne detecLon in physiologic studies. GeneLc strategies to label Nts have eased some of these barriers, including generaLon of knock-in mice that express recombinases in Nts- expressing cells such as NtsCre and NtsFlp mice123,124. These models permit rouLne labeling of Nts- expressing cells and, using recombinase-mediated geneLc tools, dissecLon of their physiological roles. While central Nts does suppress feeding and weight loss, not all Nts-expressing neurons in the brain produce these effects. Instead, disLnct Nts populaLons exhibit diverse and/or overlapping funcLons, ranging from ingesLve behavior, ambulatory acLvity, social behavior, memory, and ethanol consumpLon125–127. Similarly, pharmacological administraLon of Nts into different brain regions elicits site-specific effects. ICV or intra-nucleus accumbens (NAc) treatment with Nts or an NtsR1 agonist aAenuates baseline physical acLvity as well as cocaine- and amphetamine induced-psychosLmulant hyperlocomoLon, promoLng resLng behavior without affecLng feeding. This effect is, in part, mediated through NtsR1/Dopamine receptor 2-expressing glutamatergic terminals in the NAc93,128,129, which acLvate GABAergic medium spiny neurons known to inhibit NAc DA signaling93,130,131. Conversely, injecLng Nts into the VTA reduces feeding but increases locomotor acLvity, at least in part via NtsR1-expressing DA neurons that release DA in the NAc130,132,133. Besides hyperacLvity, microinjecLon of Nts into other parts of the limbic system and hypothalamic nuclei also reduces food intake in both food-deprived and saLated 11 rodents in a dose-dependent manner134–137. These findings underscore the necessity to systemaLcally probe the roles of Nts cells distributed throughout the brain, to idenLfy the endogenously producing Nts neurons that modify energy balance. 1.4.2.1 Neurotensin-Expressing Lateral Hypothalamic Area Neurons Regulate IngesSve Behavior and Physical AcSvity One such populaLon that has emerged in this regard is the Nts-expressing neurons in the Lateral Hypothalamic Area (LHA) or “LHANts neurons”138–142. The LHA, known as a "hub" of energy homeostasis, governs various homeostaLc processes through its heterogeneous populaLons of neurons143,144. The first data implicaLng LHA Nts in suppressing feeding came from dehydraLon- induced anorexia studies, in which water-deprived rodents stop consuming food unLl they have regained normal plasma fluid levels and only then do they resume feeding. During the dehydraLon anorexia there is upregulaLon of Nts expression in the LHA, suggesLng that it may play a role in coordinaLng energy and osmoLc needs123,145,146. Later geneLc studies also supported a role for LHANts neurons in these ingesLve behaviors. GeneLc ablaLon of LHANts neurons in adult mice promotes weight gain147. Conversely, chemogeneLc sLmulaLon of all LHANts neurons non-aversively restrains feeding even in hungry mice and increases their physical acLvity to decrease body weight144. However, acLvaLng all LHANts neurons also invokes a burst of voracious drinking that temporarily increases body weight due to the water consumpLon, although once drinking normalizes the LHANts-mediated feeding suppression and increased energy expenditure leads to weight loss144. Together these data suggest that LHANts neurons can oppositely modulate drinking and feeding. This has prompted speculaLon whether all LHANts neurons uniformly mediate feeding and drinking behaviors, or whether different subsets 12 modulate feeding restraint vs. drinking123. For instance, approximately 15% of LHANts neurons co- express LepRb (long form lepLn receptor isoform) that is necessary to mediate the anorecLc response to lepLn139,141 and maintain proper energy balance (referred as LHANts+LepRb neurons). LepLn acLon via this LepRb-expressing subset of LHANts neurons is important for body weight, as deleLng LepR from these neurons results in weight gain without disrupLng drinking. Intra-LHA infusion of lepLn reduces food intake and body weight148 and loss-of-funcLon mutaLon of lepLn signaling via LHALepRb neurons induces feeding and weight gain124,149. Similarly, sLmulaLon of LHALepRb neurons prevents fasLng-induced refeeding in food deprived mice but also reduces water seeking and water intake behavior150. These data suggest that the LHANts+LepRb neurons may influence feeding and body weight whereas drinking might be regulated via different subsets of LHANts neurons141. Nonetheless, the convergence of Nts and lepLn signaling implies that these two systems act synergisLcally to promote anorecLc behavior. Indeed, the anorecLc effect of lepLn is miLgated in NtsR1 Knockout mice or wild-type mice treated with NtsR1 antagonists117,151. These effects may be mediated partly via the LHANts neuronal projecLons to the VTA, where there are abundant NtsR1-expressing VTADA neurons142 known to modulate mesolimbic DA signaling and moLvated intake. For example, loss of lepLn acLon from the LepRb-expressing subset of LHANts neurons disrupts VTA DA signaling and weight loss behavior148. Similarly, ob/ob mice with a loss-of-funcLon mutaLon in lepLn also have reduced central Nts expression, disrupted DA signaling, increased food consumpLon, and these effects are reversed by lepLn administraLon152– 154. While lepLn acLon via LHALepRb neurons regulates food-directed behaviors, this acLvity is heterogenous, perhaps due to their varied molecular composiLon. Intra-LHA lepLn administraLon was shown to affect LHALepRb acLvity, specifically influencing appeLLve behaviors 13 without impacLng food intake or locomotor acLvity155. In vivo calcium imaging revealed that LHALepRb neurons displayed criLcal acLvity during reward-predicLve and non-predicLve cues during Pavlovian condiLoning, and optogeneLc manipulaLon of these neurons alters acute appeLLve responses via projecLons to the VTA155. These effects might be parLally mediated by overlapping LHANts+LepRb neurons projecLng to the VTA, however this paper did not look into specific subsets of LHALepRb neurons. Together these data suggest that, to some extent, weight loss behavior is mediated by the collaboraLve acLon of lepLn and Nts signaling through LHANts+LepRb neurons. Future work is required to provide resoluLon for how LHANts neurons might be modulated to bias behaviors that support weight loss. 1.4.2.2 Other Neurotensin-Expressing Neuronal PopulaSon Involved in Energy Balance Other populaLons of Nts-expressing neurons can mediate feeding suppression that could support weight loss. Nts injecLon into the paraventricular hypothalamic nucleus (PVN) significantly reduces food intake156 without altering water intake and locomoLon acLvity in food- deprived mice157. Nts acLon via the dorsomedial and ventromedial hypothalamus, as well as the nucleus of the tractus solitarius (NTS) also suppresses feeding156,158. AddiLonally, a populaLon of Nts neurons in the lateral septum (LS) was recently characterized and linked with reducing feeding in stressful situaLons. In vivo calcium imaging revealed that LS Nts neurons are acLve during physical restraint and tail suspension tests, and when sLmulated, these neurons reduce feeding and body weight without changing ambulatory acLvity143. Circuit based optogeneLc sLmulaLon of LS Nts neurons projecLng to the LHA reduces food intake suggesLng that this circuit is acLvely involved in regulaLng food intake under stressful situaLons143. However, it is yet to be elucidated if this anorecLc response is Nts-dependent, as these neurons also co-express glucagon-like 14 pepLde receptor (Glp1R) that is an important player in suppressing feeding and acLvaLon of LS Glp1R neurons leads to aphagia in mice.133 Conversely, a populaLon of Nts-expressing cells in the intersLLal nucleus of the posterior limb of the anterior commissure (IPAC), which is part of the extended amygdala, were recently implicated in promoLng feeding and obesity159. These IPACNts neurons are acLvated by cues of palatable foods to promote intake, whereas inhibiLng the IPACNts neurons protected mice from diet-induced obesity. Here again the role of Nts signaling via these neurons remains unclear, and it remains to be determined if Nts vs. other signals expressed by these neurons mediate their regulaLon of energy balance. SLll other populaLons of Nts neurons do not appear to modify feeding, such as those in the medial preopLc area (MPO) or PAG, although they mediate important roles in social behavior126 and sleep160, respecLvely. These findings underscore the fact that “Nts neurons” across the brain are not homogeneous in funcLon, and the need for populaLon-specific studies to discern their respecLve contribuLons to physiology. 1.4.2.3 Involvement of NtsR1 in Energy Balance Nts-mediated modulaLon of energy balance is currently thought to occur via NtsR1. The support for this comes from pharmacological augmentaLon of Nts-NtsR1 signaling, which promotes weight loss behavior161. Conversely, Nts-mediated feeding suppression is aAenuated by pretreatment with the NtsR1 antagonist (SR48692) and in NtsR1 knockout mice117,144,162. However, systemic and brain-wide administraLon of Nts or first generaLon NtsR1 agonists also causes life-threatening hypothermia76 and vasodepression163, dampening interest in clinical applicaLons for the Nts-NtsR1 system. ExciLngly, new NtsR1 and NtsR2 agonists, specifically β- arresLn-biased NtsR1 agonists like SBI-553, circumvent the adverse side effects associated with 15 standard NtsR1 agonists that acLvate both Gq and β-arresLn pathways164,165, and may have potenLal to safely modulate the Nts-NtsR1 system. Whether β-arresLn biased agonism is sufficient to modulate feeding and body weight is currently under invesLgaLon. However, it is also possible that NtsR1-expressing cells in certain brain regions may induce feeding restraint without eliciLng hypotension and hypothermia. For example, the VTA contains a large populaLon of NtsR1-expressing cells that couple to Gq pathways60,62, where Nts acLon is linked with producing aphagia and hyperlocomoLon but not other physiology associated with systemic Nts treatment. The VTA primarily consists of dopaminergic (DA-ergic) neurons that project to and release DA to the NAc (the mesolimbic system) or the prefrontal cortex (the mesocorLcal system). The majority of the VTA DA neurons co-express NtsR1132,142, which are acLvated by Nts treatment and release DA exclusively to the NAc166,167. This Nts-NtsR1 mediated mesolimbic DA signaling suppresses feeding and increases ambulatory behavior, parLcularly in fasted or highly moLvated mice with an augmented appeLLve drive134,135. BeAer understanding of the mechanisms by which NtsR1 in the VTA, and throughout the brain, modifies feeding may suggest approaches to engage this system to bias for feeding suppression. 1.4.2.4 Involvement of NtsR2 in Energy Balance IniLal findings from NtsR2 knockout mice indicated normal food intake and body weight168, leading to the preliminary conclusion that NtsR2 is not a parLcipant in energy balance. However, concerns have been raised across the neuroscience field about the use of consLtuLve knockout mice, as they may be subject to developmental compensaLon such that they do not accurately reflect normal physiology. Furthermore, early studies presumed NtsR2 was expressed on neurons, when it is now recognized that is primarily expressed on astrocytes65,68,70,72. Recent 16 work has established important roles for astrocytes in modulaLng feeding and body weight169,170, hence it is possible that NtsR2 expression on these cells may contribute to energy balance in yet unappreciated ways. Indeed, NtsR2 is expressed in brain regions where Nts facilitates weight loss behavior, including the VTA132,144. Further studies are needed to thoroughly evaluate the role of NtsR2 before its role in energy balance can be ruled out. This is all the more important given that there are preclinical NtsR2 agonists that do not promote hypotension and hypothermia, and so may be safe opLons to engage the Nts system. 1.5 CENTRAL NEUROTENSIN AND PAIN Nts has been characterized as an analgesic pepLde, exerLng at least part of this effect within the brain. Clineschmidt and colleagues were the first to report an anLnocicepLve response of Nts in rodents that operates independently of μ-opioid receptor (MOR) signaling, iniLaLng excitement about the analgesic potenLal of the Nts system171. Specifically, intracisternal administraLon of Nts caused a prolonged analgesic effect against noxious thermal and chemical sLmuli. Naloxone, a MOR antagonist, did not alter the anLnocicepLve effect induced by Nts171– 173. Excitement for Nts as a new analgesic was quickly dampened when subsequent studies revealed that even a very low dose of intracisternal Nts produced hypothermia that would be disadvantageous in the clinical sewng. This led the field to scruLnize the central mechanisms by which Nts modifies pain processing, and whether there might be specific brain sites and/or Nts receptors to support analgesia without adverse effects. Pain is sensed in the periphery and conveyed to the brain, and in response the descending inhibitory pain circuit controls the percepLon and regulaLon of pain174. The descending inhibitory pain circuit originates in the periaqueductal gray (PAG), which projects to the rostroventral 17 medulla (RVM) that in turn projects to the spinal dorsal horn (SDH)175. Nts and its receptors were idenLfied in each of these areas122. Furthermore, microinjecLons of Nts into the PAG122, RVM176 and spinal cord177 elicit profound analgesic responses. Given the important roles of each of these parts of the descending pain control system it is worth examining how Nts signaling engages them, and if any stand out as potenLal targets for Nts-ergic modulaLon of pain. 1.5.1 Periaqueductal Gray (PAG) The PAG contains Nts-expressing neurons that can directly release Nts to the RVM178, which were recently implicated in modulaLng sleep179. PAG Nts-expressing cells might also contribute to pain processing, supported by findings that visceral and inflammatory pain transiently increase Nts expression in the PAG180. It is conceivable that the endogenous Nts system might be upregulated during persistent pain in an aAempt to provide relief. In addiLon, the PAG also receives indirect Nts via projecLons from hypothalamic nuclei and other brain regions181. These endogenous sources of Nts to the PAG may be important in pain processing, as pharmacological studies clearly demonstrate that Nts in the PAG diminishes pain, at least in part, via dis-inhibiLng the downstream RVM182. Indeed, the anLnocicepLve effect of Nts in the PAG is nullified by electrolyLc lesions of the RVM, indicaLng the criLcal interplay of these regions for Nts-mediated analgesia183. Nts may act via NtsR1 and NtsR2 expressed within in the PAG168,184 including on cells that project to the RVM and the dorsal raphe nucleus185. CollecLvely these findings support the existence of Nts-mediated signaling in the PAG that could contribute to analgesia but the precise mechanisms governing this require further study. 18 1.5.2 Rostroventral Medulla (RVM) Nts signaling indirectly modulates the RVM via projecLons from the PAG. The RVM contains at least two neuronal subsets with contrasLng responses: “ON” cells that increase acLvity and “OFF” cells that decrease acLvity during noxious thermal sLmuli186. Nts modulates the acLvity of both populaLons via NtsR1- and NtsR2-expressing RVM neurons, and in a dose-dependent manner177,187. Low dose Nts sLmulates the “ON” cells, which has been associated with a hyper- reflexive response suggesLve of increased nocicepLve sensiLvity53,188. Conversely, high-dose intra-RVM Nts applicaLon sLmulates both populaLons, alleviaLng pain177,189,190. These findings imply that the response to noxious sLmuli depends on the intensity of the Nts signal. One possible explanaLon for this dose-dependent effect may involve the differenLal engagement of high- affinity NtsR1 and low-affinity NtsR2 receptors. It is conceivable that a low dose of Nts acLvates only NtsR1, targeLng the "ON" cells, while a high dose acLvates both the "ON" cells and the "OFF" cells expressing NtsR2. In support of this interpretaLon, treatment with the NtsR1 selecLve antagonist SR48692 mildly prolongs tail-flick latency; this suggests that blocking low-level endogenous Nts-NtsR1 signaling decreases nocicepLve responding189. In fact, Nts acLons in the RVM are likely mediated by both NtsR1 and NtsR2, as inhibiLng either individually does not fully eliminate the anLnocicepLve response162,176,177. This could be due to NtsR1 and NtsR2 expressing cells of the RVM mediaLng different neural circuitry. For example, NtsR1-expressing neurons are serotonergic and Nts-NtsR1 acLon in the RVM neurons results in the release of serotonin in the SDH177. In contrast, NtsR2-expressing RVM neurons release norepinephrine via an indirect pathway and anLnocicepLon produced by NtsR2 agonist beta-lactotensin (β-LT) is reduced by intrathecal injecLon of yohimbine, an α2-adrenoceptor antagonist176. Therefore, both NtsR1 and 19 NtsR2 play a role in transmiwng anLnocicepLve signals from the RVM to the SDH. AddiLonally, some RVM neurons may express and release Nts directly the spinal cord to modulate signaling there191. 1.5.3 Spinal Dorsal Horn (SDH) Treatment with Nts or Nts analogs into the SDH (e.g. intrathecal injecLon) alleviates pain192– 194. The SDH contains local Nts-expressing interneurons and receives afferent indirect Nts input from RVM, evidenced by the dense Nts-ergic fibers in this region. The Nts-expressing afferent neurons are glutamatergic and form excitatory synapLc connecLons with NtsR2-expressing GABAergic neurons This inhibitory transmission serves to suppress nocicepLve signals at the level of the SDH 195. Nts is also present in dorsal root ganglion (DRG) neurons196. AddiLonally, a subset of DRG neurons also expresses NtsR1, which is also implicated in pain regulaLon at the SDH level, although the precise underlying mechanism remains unclear197. 1.5.4 Does Neurotensin Facilitate or/and A]enuate NocicepSon? While Nts has been generally characterized as having analgesic effects, the impact of Nts on nocicepLon may vary, conLngent on the intensity of Nts signaling and the Nts receptor isoform involved (NtsR1 vs NtsR2). As discussed above, low-level Nts signaling in the RVM is associated with increased nocicepLon, which can be improved by inhibiLng NtsR1. Similarly, knockout mice lacking NtsR2 (NtsR2-KO) have increased jump latency on the hot plate suggesLve of reduced thermal nocicepLon168,198. Some have interpreted this as evidence that NtsR2 may also play a facilitaLng role in nocicepLon. Yet the jump response to thermal sLmuli is ohen associated with a flight response to escape danger or acute pain, ensuring safety. Beyond the reduced 20 nocicepLon, the increased jump latency in NtsR2-KO mice may suggest that NtsR2 signaling is involved in promoLng behavior aimed at escaping danger during acute pain tesLng. Yet, if Nts were to enhance nocicepLon, one would expect that NtsR1 knockout mice and NtsR2 knockout mice would display a decreased paw lick response during the hot plate test for thermal pain responding compared to wildtype mice (indicaLve of reduced pain). However, no differences in paw lick response were observed between the knockout groups and wildtype mice in this regard. When considering both the paw lick response and jump latency data together, it suggests that Nts might not potenLate nocicepLon but rather act as an alarm system to detect and facilitate the escape from acute pain. This interpretaLon is further supported by another study indicaLng that mice lacking Nts (Nts-KO) exhibit significantly less visceral nocicepLon than wildtype mice199, emphasizing the importance of Nts in pain percepLon. A caveat is that knockout mouse studies should be interpreted with cauLon. ConsLtuLve loss of Nts or NtsRs may cause developmental alteraLons such that they no longer model normal physiology, including possibly exacerbaLng the funcLon of the remaining Nts receptor isoforms. More work is required to untangle how the endogenous Nts system modulates pain processing to understand if the system can be pharmacologically harnessed for analgesia. 1.5.5 Neurotensin Receptors in Analgesia NtsR1 has received greater focus for modulaLon of body weight, but NtsR2 has been proposed as the major receptor isoform mediaLng analgesia, largely based on experiments using first-generaLon pharmacological agonists and antagonists of the Nts system. IniLal antagonist studies reported that SR142849, the non-selecLve antagonist that acts at NtsR1 and NtsR2, decreased Nts-mediated analgesia189. Moreover, levocabasLne, a purported NtsR2-selecLve 21 ligand, reduced Nts-mediated analgesia200, whereas the NtsR1-specific antagonist SR48692 did not201,202. Likewise, anLsense oligonucleoLde inhibiLon of NtsR2, but not NtsR1, aAenuated Nts- induced analgesia in aceLc acid-mediated pain aher repeated brain microinjecLons184,202. This led to the conclusion that Nts promotes anLnocicepLon primarily through signaling via NtsR2. Recent advances in pharmacological understanding have challenged this conclusion. SR142849, SR48692, and levocabasLne were revealed to funcLon as parLal agonists for NtsR2 under certain condiLons, such that they cannot truly evaluate NtsR2 signaling, nor discriminate NtsR1 and NtsR2 signaling64,203 For instance, levocabasLne treatment in the brain reduces visceral pain200, but if it is co-administered with Nts it diminishes the overall Nts-mediated anLnocicepLon. This reducLon in analgesic response may be aAributed to the fact that, as a parLal agonist, levocabasLne competes with Nts (a full agonist), resulLng in a compeLLve agonism that overall diminishes the analgesic effect. Recent geneLc studies also quesLon the primacy of NtsR2 over NtsR1 for anLnocicepLon. In a model of thermal pain, anLsense oligonucleoLdes against NtsR1 that were administered prior to Nts treatment notably diminished Nts-induced analgesia88. Remarkably, as liAle as 35-45 % reducLon in NtsR1 expression in the PAG and hypothalamus was adequate to abolish Nts-induced analgesia. In alignment with the geneLc inhibiLon studies, mice lacking NtsR1 exhibited heightened sensiLvity (hyperalgesia) to thermal pain198. Furthermore, a newer NtsR1 agonist (PD149163) alleviates thermal pain and formalin-induced inflammatory pain in a dose-dependent manner. Importantly, this NtsR1 agonist-mediated anLnocicepLon is hindered by the NtsR1 selecLve antagonist SR48692177,193. Taken together, these invesLgaLons support roles for NtsR1 and NtsR2 in modulaLng nocicepLon, and that both receptors warrant aAenLon as targets to provide analgesia. Although very liAle is known about the role of NtsR3 22 and NtsR4 in nocicepLon, a relaLvely recent study revealed that Nts and NtsR2 are upregulated in the brain of NTSR3/sorLlin deficient mice leading to increased anLnocicepLon in thermal and chemical-induced pain. While NtsR1 levels remained unchanged, these data suggest that NtsR3 interacts with other Nts receptors and perhaps indirectly modulates nocicepLon204. 1.5.6 Pharmacological Efforts to Design a Neurotensin Analog for TreaSng Pain For years, researchers have explored Nts and NtsR agonists as a potenLal alternaLve to opioids for pain relief. Yet, leveraging the Nts system for pain relief presents challenges due to the very short half-life of Nts and its metabolic instability in the circulaLon101. More criLcally, central treatment with Nts or the first-generaLon NtsR agonists can provide analgesia, but also accompanying hypothermia and vasodilaLon. These caveats have fueled an acLve drug discovery effort to develop Nts system modulators that can effecLvely ease pain without causing unwanted side effects. IniLally, studies mainly invesLgated the analgesic effects of Nts and its receptor agonists in acute pain scenarios. However, paLents commonly endure chronic pain in clinical sewngs, including visceral pain, neuropathic pain and inflammatory pain. Therefore, recent research has expanded to explore the potenLal nocicepLve role of Nts using various models, including inflammatory and neuropathic condiLons that are linked with chronic pain. In this secLon, we will explore various Nts agonists that have been developed thus far and assess their capacity to relieve diverse physiological and chronic pain condiLons. 1.5.6.1 AceSc Acid-Induced Visceral Pain Model Visceral pain can be modeled in rodents by injecLng aceLc acid and measuring their resulLng writhing behavior. Preclinical NtsR2 agonists have been developed that show promise in providing analgesia in this model of chronic visceral pain without producing hypothermia and 23 hypotension205. The beneficial effect of NT79 may be aAributed, at least in part, to the modulaLon of serotonin expression, which tends to increase in visceral pain206. AddiLonally, the analgesic effect of NT79 against aceLc acid-induced writhing was also observed in NtsR1 homozygous knockout mice, which suggests that NtsR2 signaling alone is sufficient to regulate nocicepLon. InteresLngly, the NtsR1-specific agonist NT72 and the non-selecLve agonist for NtsR1/2, NT69L, also provide analgesia a mouse model of chronic visceral pain, even in mice that lack NtsR2. Taken together, these data support that both NtsR1 and NtsR2 are individually sufficient, even in the absence of the other, to relief visceral pain207. 1.5.6.2 Neuropathic Pain Model Surgical sciaLc nerve injury (SNI) is commonly used to induce and model neuropathic pain in rodents. In this procedure, the sciaLc nerve is exposed, and the Lbial and common peroneal branches are transected while the sural branch is leh intact208,209. Mice undergoing SNI surgery typically develop chronic pain characterized by heightened thermal sensiLvity and mechanical allodynia in laboratory sewngs. SciaLc rodent injury models also exhibit mechanical allodynia measured with the von Frey filament test. Intriguingly, spinal cord expression of Nts was normal 3 days aher sciaLc nerve injury, but is elevated with sustained injury210,211. This implies that the gradual accumulaLon of endogenous Nts over several days might be involved in influencing chronic pain. Pharmacologically, intrathecal administraLon of the NtsR1-selecLve agonist PD 149163 aAenuates nerve-injury-induced thermal and mechanical hyperalgesia192. A more recent analogue of Nts, contulakin-G (CGX), derived from snail venom, has demonstrated safety in clinical trials. Administering CGX intrathecally has been shown to alleviate thermal and mechanical hypersensiLvity in both neuropathic and inflammatory pain models of both sexes. 24 Despite CGX's ability to bind to both NtsR1 and NtsR2, its anLnocicepLve effects are enLrely blocked in the absence of NtsR2212. This suggests that CGX could avoid adverse physiological side effects that have been aAributed to NtsR1 signaling, though more work is needed to assess its full potenLal in modulaLng pain and other behaviors. 1.5.6.3 Inflammatory Pain Model To induce inflammatory pain, mice receive trans-plantar injecLon of either formalin or complete Freund Adjuvant (CFA). Similar to the SNI model, inflammatory pain is associated with increased Nts expression213. Consistent with anatomical data, pharmacological studies confirmed that Nts acLons in the spinal cord are effecLve in alleviaLng inflammatory pain. For instance, the intrathecal administraLon of the NtsR1-selecLve agonist PD 149163, the non-selecLve NtsR1/2 agonist NT69L193, and the selecLve NtsR2 agonist effecLvely alleviate formalin-induced hypersensiLvity in dose-dependent manner214. AddiLonally, the NtsR2 agonist NT79 is a promising candidate for regulaLng persistent pain, parLcularly since it synergized with morphine- induced anLnocicepLon in a formalin-induced inflammatory pain model. Consequently, a combinaLon therapy involving NT79 could potenLally reduce the amount of morphine required for pain management, and reduce the risk of developing life-threatening morphine-related side effects206. With this goal in mind, there is an acLve effort to design hybrid compounds that target both MOR and NtsR2 while avoiding NtsR1-associated side effects215,216 but metabolic stability and brain accessibility remain challenging217. At least one preclinical hybrid compound has been reported that alleviated acute and persistent inflammatory pain in opioid-independent manner without causing hypothermia, but also caused a mild hypotensive response218. ExciLngly, this 25 hybrid compound also synergized morphine-induced analgesia, indicaLng promise for a combined NtsR/MOR targeted approach to maximize analgesic response. 1.5.6.4 Thermal Pain Model Thermal pain is ohen modeled by placing rodents on a hot surface and assessing the Lme it takes for them to withdraw their paw or jump away from the surface. Newer Nts analogs such as NT69L and JMV2012 have improved metabolic stability compared to naLve Nts, can bind to both NtsR1 and NtsR2 with high affinity, and effecLvely diminish thermal pain responses 90 minutes post-injecLon. The analgesic effect persists for up to 5 hours163,214,219. However, despite their promise in alleviaLng thermal pain, these Nts agonists sLll induce hypothermia, rendering them less than ideal as potenLal analgesics 1.6 FUTURE CHALLENGES: IDENTIFICATION OF COMMON SITES TO TREAT OBESITY AND PAIN The existence of peripheral vs. centrally produced pools of Nts raises the quesLon of where and how Nts might simultaneously support weight loss and analgesia, as needed to address comorbid obesity-pain. Both pools of Nts have promise in miLgaLng pain. Enhancing Nts signaling, whether in the peripheral or central nervous system, has been shown to provide pain relief across various modaliLes89,173,220, including the miLgaLon of neuropathic pain resulLng from peripheral nerve injury192,221. However, contrasLng effects are observed when comparing the peripheral vs. central Nts systems on body weight. While Nts promotes weight gain in the periphery, most data suggest that it acts centrally to facilitate weight loss by suppressing appeLte and encouraging physical acLvity114,161,222. Moreover, disrupLons in central endogenous Nts signaling may contribute to the onset or perpetuaLon of obesity, as both geneLcally and diet- induced obese rodent models exhibit reduced Nts expression in the brain154,223. Taken together, 26 these data suggest that modulaLng central Nts signaling may be the beAer target to achieve simultaneous weight loss and pain relief, as needed to effecLvely treat comorbid obesity-pain. Yet, there are many unknowns that must be addressed to understand if and how the central Nts system may be used to address obesity-pain. While there has been ample research directed toward pharmacologically leveraging the Nts system to reduce feeding or pain, much less is known about how the endogenous Nts system is altered in linked obesity-pain. This informaLon could have implicaLons for understanding the origins of obesity-pain and, hopefully in the future, on how to pharmacologically leverage the Nts system to treat it. For instance, is there is a common central Nts system to modulate body weight and pain processing, or are these mediated by separate populaLons of Nts cells, signaling, and circuitry? Given the diverse physiology mediated by various Nts and NtsR-expressing populaLons throughout the brain, it is very possible that there are separate, dedicated central Nts systems devoted to regulaLon of feeding vs. nocicepLon. This seems, on its face, to be supported by the fact that no published studies to date have documented brain sites that simultaneously modulate feeding and pain processing. Yet it is possible that there may be common nodes of central Nts signaling that they have been overlooked because Nts mediated effects have only been studied in separate physiological contexts. A goal of future Nts system studies should be to assess pain and ingesLve behaviors together, so as not to miss potenLal connecLons. However, synthesizing the published findings to date (Table 1.1) indicates at least two brain sites linked with Nts-modulated energy balance and pain processing: the amygdala and the LHA. Intriguingly, while both regions are implicated in Nts-mediated analgesia they may oppositely modulate feeding and body weight. AcLvaLng the Nts-expessing IPAC neurons of the extended amygdala promotes mice to consume 27 highly palatable, obesogenic food. In contrast, acLvaLng Nts-expressing neurons in the LHA suppresses food intake, including blunLng operant responding for palatable food. Further study of each area is necessary to determine whether they can simultaneously modulate body weight and pain and may reveal the reason for their opposing effects on food intake. Indeed, the variability of amygdala and LHA acLons on feeding aligns with the documented pleiotropism of the central Nts system, where disLnct neural pathways produce and release Nts to target cells expressing NtsR, resulLng in diverse physiological and behavioral effects. Going forward, it will be important to characterize the contribuLons of all specific populaLons of Nts-expressing cells, their projecLon targets, and the complement of Nts receptors expressed there in both energy balance and pain processing. AddiLonally, there is ongoing debate over whether NtsR1 or NtsR2 mediate the anorecLc and analgesic effects. This has proven challenging to resolve due to the limited specificity of previously available pharmacological tools or tools to idenLfy precisely where in the brain these receptors are expressed. Resolving the receptor and signaling systems underlying Nts-modulaLon of energy balance and analgesia is criLcal to inform the drug- discovery pipeline and hone in on the most effecLve targets. By uLlizing the array of new pharmacological and geneLc tools available to the filed, there is hope to reveal the intricacies of the Nts signaling system and its potenLal to support weight loss and pain. 1.7 GOALS OF THE DISSERTATION Chronic pain and obesity are prevalent, costly health problems that frequently occur together12,13. There are few intervenLons for obesity and a pikall of current non-opioid pain medicaLons is that they induce weight gain29. Obesity also contributes to use of opioids that alleviate pain but increase risk for dependence and fuel the opioid crisis36, creaLng need for pain 28 therapeuLcs that do not cause dependence. Our limited understanding of the neurobiology underlying obesity and pain comorbidity is a major knowledge gap hindering the design of effecLve treatment strategies needed to improve and save lives. Brain injecLons of the Nts or Nts receptor agonists promotes weight loss114,151,158,161,222,224, and analgesia171,225,226, suggesLng that augmenLng central Nts has potenLal to treat obesity and pain. Yet, the source(s) of endogenous Nts that could mediate these effects has remained elusive. Our lab has previously shown that a large populaLon of Lateral Hypothalamic Area (LHA) neurons express Nts (referred to as LHANts neurons) that support weight loss147. ExciLngly, acLvaLng all LHANts neurons restrains feeding and increases physical acLvity even in hungry mice that reduced their body weight, indicaLng their importance for supporLng weight loss. However, the role of LHANts neurons in pain processing, and whether they might also have potenLal to relieve pain had yet to be explored. If LHANts neurons could simultaneously support weight loss behaviors and provide analgesia they could be promising targets to address comorbid obesity-pain. I examined this via my dissertaLon research, tesLng the role of LHANts neurons in pain processing of normal weight mice (Chapter 2) and in diet-induced obese mice (Chapter 3). Hence, the goals of this dissertaLon are: 1. Detemine whether specifically acSvaSng LHANts neurons alleviate thermal, neuropathic, and inflammatory pain via neurotensin receptor signaling in normal weight mice (Chapter 2) Hypothesis: LHANts neurons are an endogenous source of Nts that can provide anLnocicepLon, and hence, that acLvaLng LHANts neurons would alleviate pain dependent on Nts signaling via NtsRs. 29 Method: I combined NtsCre mice and Cre-mediated excitatory Designer Receptors Exclusively AcLvated by Designer Drugs (DREADDs) to selecLvely acLvate LHANts neurons in thermal, neuropathic and inflammatory pain models, so as to decipher their role in pain regulaLon. AddiLonally, I pretreated these mice with the brain permeable Nts receptor pan-antagonist SR142948 before CNO-mediated acLvaLon of LHANts to assess the role of Nts signaling in pain regulaLon. These data revealed whether acLvaLng LHANts neurons in normal weight alleviates pain and if analgesia mediated by LHANts neurons requires Nts-NtsR signaling. 2. Establish whether acSvaSng LHANts neurons alleviates obesity-induced pain and its dependency on Nts-NtsR signaling (Chapter 3) Hypothesis: AcLvaLng LHANts neurons alleviates obesity-induced pain in an Nts-NtsR dependent manner, such that they are an endogenous source of Nts to mediate analgesia. Method: I injected Cre-dependent excitatory DREADDs AAVs in the LHA of diet-induced obese NtsCre mice to selecLvely acLvate their LHANts neurons. I then performed pain tesLng to study the role of LHANts neurons in pain processing during obesity. To determine if Nts is a key player in LHANts neurons-induced analgesia I injected newly generated Ntsflox/flox mice with AAV-Cre directly into the LHA to enable Cre-mediated, site-specific deleLon of Nts. Together, these data revealed if acLvaLng LHANts neurons in obese mice alleviates obesity-induced sensiLvity and inflammatory pain and whether specific LHA Nts deleLon is needed for pain relief in normal weight and obese mice. Studying these mice will establish the specific role of LHANts neurons in comorbid obesity and pain. By compleLng this research, I revealed that the potenLal of LHANts neurons in pain reducing pain and how endogenous sLmuli modulate LHANts neurons. Specifically, acLvaLng 30 LHANts neurons alleviates inflammatory and neuropathic pain and this analgesic effect diminishes when mice are pretreated with NtsR1/2 pan-antagonist suggesLng that Nts signaling is required for anLnocicepLon. Our work also showed that LHANts neurons do not regulate acute thermal pain in naïve healthy mice. Moreover, we also established that LHANts neurons regulate obesity- induced pain that is dependent on Nts signaling. Furthermore, deleLng Nts specifically from LHA neurons in adulthood revealed that Nts signaling is important for pain relief in hypersensiLve obese mice but not in normal weight naïve mice. These collecLve findings support the premise that augmenLng acLvaLon of all LHANts neurons holds promise in modulaLng body weight and pain. This work advances understanding of LHANts neurons and endogenous roles of Nts that, in the future, might be leveraged to address comorbid obesity-pain. 31 Table 1.1: Central Nervous System Sites of Nts System Effects on Pain Processing and Energy Balance. NR = not reported. IPAC = IntersLLal Nucleus of the Posterior Limb of the anterior commissure, a nucleus of the central extended amygdala. Shading indicates sites in which the Nts system has been documented to modulate pain and energy balance, though via separate studies. Ref Brain Receptor Nts, and/or Type of Pain Analgesi Food Intake Water Locomo