Tyrosine Hydroxylase Knockdown at the Hypothalamic Supramammillary Nucleus Area Induces Obesity and Glucose Intolerance

Yahong Zhang; Tsung-Huang Tsai; Michael Ezrokhi; Carl Stoelzel; Anthony H. Cincotta

doi:10.1159/000535944

Tyrosine Hydroxylase Knockdown at the Hypothalamic Supramammillary Nucleus Area Induces Obesity and Glucose Intolerance

Zhang, Yahong; Tsai, Tsung-Huang; Ezrokhi, Michael; Stoelzel, Carl; Cincotta, Anthony H. 2023-01-01 00:00:00 IntroductionThe supramammillary nucleus (SuMN), localized in the posterior hypothalamus (PH) overlying the mammillary body, consists of a heterogeneous population of neurons (e.g., glutamatergic, GABAergic, neuropeptidergic, and dopaminergic) and exerts marked influences on a broad spectrum of brain areas that modulate cognition, ingestive behavior, reward and energy metabolism including the prefrontal cortex, hippocampus, mesolimbic system, hypothalamus, and raphe nuclei [1, 2]. In addition to its established role in integration and modulation of cognition and emotional aspects of goal-oriented behaviors [1, 3‒6], recent studies have revealed its role in the regulation of feeding [7‒9] and feeding-independent [10] fuel metabolism. We have previously reported that the neuronal activation state of the SuMN regulates peripheral fuel metabolism and body weight [10]. Circadian-timed activation of SuMN neuronal activity among high-fat diet (HFD)-induced insulin-resistant/glucose-intolerant rats via daily circadian-timed direct SuMN administration of glutamate receptor agonists so as to coincide with the natural circadian peak of general and dopaminergic SuMN neuronal activity in healthy rats reversed the HFD-induced insulin resistance (IR) and glucose intolerance [5, 10, 11]. On the other hand, chronic inhibition of SuMN neuronal activity with SuMN administration of GABAa receptor agonist/glutamate receptor antagonists induced IR, glucose intolerance, and obesity in rats held on regular chow diet [10]. It is presently not known which SuMN neuronal type(s) is/are involved in the induction of such metabolic responses, and the SuMN consists of a heterogenous population of neurons having extensive connections with multiple brain metabolic control centers [1, 2]. Of particular interest in this regard however is a subpopulation of dopaminergic neurons located at the medial SuMN that project directly to the suprachiasmatic nucleus (SCN) area as well as to other central nervous system (CNS) metabolic centers such as the lateral septum (LS), medial preoptic area (MPOA), and dorsomedial hypothalamus (DMH) [2, 11‒14].Circadian dopaminergic input activity at the clock SCN area that largely emanates from the medial SuMN (and coincides with the circadian peak of overall SuMN neuronal activity) [11], but also including from other brain centers, is a strong regulator of peripheral metabolism [11, 15]. A chronic reduction in circadian peak dopaminergic activity at the SCN area facilitates IR syndrome in healthy rodents [16], while circadian-timed administration of dopamine (DA) directly to the SCN at the time of day that it peaks in healthy animals reverses the IR syndrome [11, 17] (similar to the circumstance with overall SuMN activity regulation of metabolism described above). A composite of neurophysiological studies has demonstrated that circadian DA input activity at the SCN regulates its output control of metabolism via the neuroendocrine axis. Loss of the circadian peak of dopaminergic input activity at the SCN area via HFD feeding induces SCN neuronal output activity alterations that potentiate an elevation in sympathetic nervous system activity, hypothalamic-pituitary-adrenal axis corticosteroid hyperactivity, hyperinsulinemia, and leptin resistance [11, 16]. In turn, this SCN-driven neuroendocrine shift leads to a wide range of pathophysiological changes including systemic oxidative stress, low-grade inflammation, IR, and glucose intolerance [18] all of which collude to play pivotal roles in the development of obesity and type 2 diabetes mellitus [19‒24]. In addition to SuMN to SCN dopaminergic circuitry, SuMN dopaminergic input to the LS, MPOA, and DMH metabolic control centers may modulate (their) control of metabolism (via their communication with the SCN or otherwise).Therefore, the possibility exists that the previously identified metabolic effects of overall SuMN neuronal activity level on peripheral fuel metabolism may derive in large part from activity of medial SuMN dopaminergic neuronal circuits to the SCN and other relay metabolic centers (e.g., LS, MPOA, DMH) that in turn communicate with the SCN and influence peripheral metabolism. The present study was therefore undertaken to determine whether chronic tyrosine hydroxylase (TH, rate-limiting enzyme in DA synthesis [25]) knockdown (THx) at the medial SuMN area alters peripheral fuel metabolism in rodents. The SuMN-THx was accomplished via site-specific virus-mediated introduction of a short hairpin RNA (shRNA) which specifically targets TH directly to medial SuMN neurons that allows for stable transcription of the shRNA and long-term knockdown of TH expression. Knockdown of TH within CNS neurons is well established to produce a near-equivalent reduction level in neuronal DA content and dopaminergic function [26‒29]. To determine if SuMN-THx would induce progression to metabolic syndrome (MS), this intervention was introduced into a selected cohort of rats highly resistant to naturally developing MS over time based on model strain (female Sprague-Dawley [SD] rat), innate insensitivity to the obesogenic effect of HFD-resistant (HFDr), and maintained on a low-fat regular chow (RC) diet – female SD rat HFDr fed RC [f-HFDr-RC]. Additionally, subsequent experiments investigated metabolic effects of such SuMN-THx in (a) female HFDr rats fed HFD [f-HFDr-HFD], (b) male HFDr rats fed RC [m-HFDr-RC], and (c) female HFD-sensitive rats fed RC [f-HFDs-RC] over 1 year’s time to evaluate the influence of HFD, gender, and long-term response, respectively – upon body weight change, feeding, glucose tolerance, and insulin sensitivity. We hypothesized that such SuMN-THx treatment of MS-resistant animals would result in manifestation of glucose intolerance and obesity.Materials and MethodsAnimalsSD rats (female 10–12 weeks of age; male 6 weeks of age) (Taconic Biosciences, Hudson, NY, USA) were housed individually in a temperature and humidity-controlled room under a 14-h light daily photoperiod (14-h light and 10-h dark [14:10 LD]) with food and water ad libitum. Rats were allowed to adapt to the animal care facility for 1–2 weeks before the initiation of experiments. To best assess the potential MS-inducing impact of SuMN-THx (TH knockdown primarily confined to the medial SuMN plus minor involvement of immediate adjacent PH), studies were conducted in animals naturally resistant to development of MS. To avoid the confounder of age-induced IR upon the background metabolic status of the study animals during the study period, female SD rats were used in the primary studies (except in a separate study with males for gender comparison) inasmuch as they maintain a steady state of insulin sensitivity for a long period of their lifetime versus male rats of this strain that develop IR progressively from an early age [30, 31]. For the similar reason, animals were held on a long 14-h light daily photoperiod to avoid the confounder of a short ≤12-h light photoperiod-induced IR upon the background metabolic status of the study animals during the study period [32]. Finally, such female rats were selected to be resistant to the obesogenic effects of HFD (HFDr). See Experimental Design section of Methods for the protocol methodology to identify HFDr versus HFD-sensitive (HFDs) rats. Thus, the female SD rat resistant to the obesogenic effect of HFD and fed low-fat regular chow (RC) [f-HFDr-RC] was chosen as the primary model to evaluate the impact of SuMN-THx on peripheral fuel metabolism and feeding behavior. In subsequent experiments to assess the influence of HFD, gender, and longer term response, the same SuMN-THx intervention was administered to (a) female SD rat HFDr fed HFD [f-HFDr-HFD], (b) male SD rat HFDr fed RC [m-HFDr-RC], and (c) female SD rat HFDs fed RC [f-HFDs-RC] for over a year, respectively. Depending on the individual experimental design, rats were fed either regular chow (RC; 18% of calories/gram of food from fat, Envigo-Teklad rodent pellet 2018) or an HFD (60% of calories/gram of food from fat, Research Diets Inc, d12492). All animal experiments were conducted in accordance with the National Institutes of Health Guide for the Care and Use of Laboratory Animals [33] and with the protocols approved by the Institutional Animal Care and Use Committee of VeroScience, LLC.Viral ConstructsAdeno-associated viral plasmid (AAVp) containing a shRNA designed to target a distinct nucleotide stretch within the coding region of the tyrosine hydroxylase gene (Th) (AAVp-Th-shRNA-green fluorescent protein [GFP]) and control AAVp containing a shRNA targeting a scrambled nucleotide sequence (AAVp-Scr-shRNA-GFP) were kindly provided by Dr. RJ DiLeon of Yale University, CT, USA (Hommel JD 2003). The viral constructs were packed, purified, and prepared at ready-to-use titers expressed as GC/mL (genome copies per mL) (AAV5-Th-shRNA-GFP, 1.65 × 1013 GC/mL; AAV5-Scr-shRNA-GFP, 1.06 × 1013 GC/mL) by Vigene Biosciences (Rockville, MA, USA). Upon receiving, the viral solutions were aliquoted and stored at −80 C until use.Stereotaxic Surgery/Virus InoculationRats were anesthetized with ketamine/xylazine (80/12 mg/kg, intraperitoneally [ip]) and secured on a stereotaxic apparatus. A small craniotomy (∼1 mm in diameter) was made at 4.55 mm posterior from bregma and 0.2 mm from midline to allow access to the medial SuMN. A microinjector filled with a viral construct was mounted to the stereotaxic apparatus and slowly lowered to the medial SuMN according to the coordinates: 4.55 mm posterior from bregma, 0.2 mm from midline, 8.4 mm below bregma. A volume of 0.2 μL of the viral construct (AAV5-Th-shRNA-GFP, 1.65 × 1013 GC/mL or AAV5-Scr-shRNA-GFP [control], 1.06 × 1013 GC/mL) was continuously microinfused over 2 min. At the end of infusion, the injector was left in place for additional 5 min before slow withdrawal. The incision was sutured, and the rat was returned to the colony and allowed to rest for a minimum of 2 weeks to ensure sufficient viral expansion before experimentation.Tissue Processing/ImmunohistochemistryThe efficiency of a virus-mediated knockdown of Th gene expression was verified by immunohistochemistry. Rats with medial SuMN inoculation of AAV5-Th-shRNA-GFP or control AAV5-Scr-shRNA-GFP were anesthetized with ketamine/xylazine (80/12 mg/kg, ip) and sacrificed by transcardiac perfusion with 4% paraformaldehyde. The brains were collected, post-fixed in 4% paraformaldehyde overnight at 4°C, cryoprotected in 30% sucrose until brains were fully fixed. The brains were then frozen in 2-methylbutane chilled on dry ice and kept at −80°C till being further processed. Sequential coronal brain sections (30 μm in thickness) were cut using a cryostat at −20°C and collected in PBS buffer. The free-floating sections were transferred to a freezing solution (30% ethylene glycol, 25% glycerol prepared in 0.05 m phosphate buffer) and kept at −20°C. For TH-3,3′-diaminobenzidine (DAB)-immunohistochemical staining, the free-floating sections were washed in PBS and treated with 3% hydrogen peroxide for 10 min to quench the endogenous peroxidase activity. The brain sections were then rinsed in PBS and incubated with Animal-Free Blocker (SP-5030, Vector Lab) for 1 h at room temperature to block background non-specific binding. The brain sections were subsequently incubated with monoclonal mouse anti-TH antibody (1:2,000) (MAB318, Millipore) overnight at 4°C. After thorough washing, the brain sections were incubated with biotinylated anti-mouse IgG for 1 h at room temperature followed by ABC complex amplification. The TH immunoreactivity was revealed by a peroxidase reaction using DAB as chromogen. Finally, the stained brain sections were sequentially mounted on gelatin-coated micro-slides, and dehydrated and covered with Permount for light microscopy. For TH-immunofluorescent staining, the brain sections were processed in sequential steps as follows with PBS washing between each step: blocked with 10% normal goat serum (Vector S-1000) for 1 h at room temperature, incubated in monoclonal rabbit antibody against TH (1:250) (Thermo Fisher Scientific, Cat#701949) overnight at 4 C, followed by incubation in Rhodamine Red X (RRX)-goat anti-rabbit IgG (1:200) (Jackson ImmunoResearch, Cat# 111-296-144) for 1 h at room temperature. The antibodies were prepared in PBS containing 5% normal donkey serum and 0.2% Triton-X. At the end of staining, the brain sections were wet mounted on charged micro-slides and covered with VectorShield medium with DAPI (Vector Lab, H1200). Fluorescent images were acquired using BioTek Lionheart FX automated microscope and imaging software (BioTek Instruments, Inc., Winooski, VT, USA) under Texas Red, GFP, and DAPI filters. The acquired images were subsequently analyzed using BioTek image analysis software (BioTek Instruments, Inc.) to obtain cell count and mean fluorescent density with background correction.Experimental DesignStudy 1: Verification of the Extent of THx at the SuMN Area (SuMN-THx)To verify the extent of TH knockdown, 12 female SD rats were separated into 2 groups and surgically inoculated at the medial SuMN area with the viral construct AAV5-Th-shRNA-GFP to knockdown Th gene expression (THx, N = 7) or control AAV5-Scr-shRNA-GFP (Ctr, N = 5). The rats were sacrificed at 3 months post-virus inoculation by transcardiac perfusion with 4% paraformaldehyde, and the brains were collected for immunohistochemistry. Coronal brain sections (30 μm in thickness) were cut through the SuMN and sequentially collected at 90 μm intervals into 3 sets in PBS buffer. One set of brain sections was processed with TH immunostaining using DAB as a chromogen. Another set was processed with TH(+)-RRX-immunofluorescence red staining. TH(+)-DAB-stained brown cells at the SuMN area were counted on 6 representative sections through the SuMN (bregma −4.4, −4.5, −4.6, −4.7, −4.8, and −4.9 mm). Localization of TH (+)- red fluorescence inside virus-transfected GFP-expressing green cells at the SuMN area was examined in THx and Ctr brains. To verify that the inoculated virus did not spill over to the adjacent VTA DA neurons, the spread of virus-transfected GFP-expressing cells at adjacent VTA was evaluated. Moreover, the TH(+)-RRX-immunofluorescent images at VTA of THx and Ctr brains were acquired and subsequently quantified by measuring mean fluorescence density and fluorescent red cell counts (described in detail in Tissue Processing/Immunohistochemistry) on the brain sections cut through the rostral (bregma, −4.5 mm) and caudal (bregma, −4.8 mm) SuMN, respectively.Study 2: Evaluation of the Impact of SuMN-THx on Body Fuel Metabolism and Feeding Behavior in a Rat Model Least Prone to Develop Metabesity – f-HFDr-RC RatsStudy 2Ai–iii: The Effect of SuMN-THx on Body Weight, Food Consumption, Feed Efficiency, Glucose Metabolism, Plasma Metabolic Parameters, Liver Metabolic Gene Expression, and Adipose Fatty Acid Oxidation (FAO) Rate in f-HFDr-RC Rats. Female SD rats were separated into HFDr and HFD-sensitive cohorts using the median % body weight gain over 1 week of high fat feeding (=8%) as separation reference. Half of the rats with % body weight gain less than the median % body weight gain (≤8%) were considered as HFDr rats [34, 35] and employed in the present study. Such selected HFDr rats (BW = 259 ± 4.0 g, 15 weeks old) were fed RC and randomly divided into two groups for local virus inoculation at the medial SuMN area of AAV5-Th-shRNA-GFP to knockdown Th gene expression (THx, N = 12) or control AAV5-Scr-shRNA-GFP (Ctr, N = 10). Throughout the study period, body weight and food consumption were monitored. A glucose tolerance test (GTT) was performed at 11 weeks post-virus inoculation. The rats were sacrificed by decapitation at 17 weeks post-virus inoculation during the fasting period of the day (zeitgeber time, 6–8 [ZT 6–8 h after light onset]). The trunk blood was collected for plasma assay of metabolic parameters. Parametrial white adipose tissue (pWAT) and retroperitoneal white adipose tissue (rWAT) were weighed. Adipocyte oxygen consumption rate (OCR), a measure of cellular metabolic activity, and FAO were each measured in freshly collected pWAT from 7 pairs of rats (7 THx and 7 Ctr). Hepatic tissues were quickly frozen in liquid nitrogen for RT-quantitative PCR (qPCR) analysis of the mRNA expression of metabolic relevant genes and triglyceride content measurement. The brains were collected for immunohistological verification of TH knockdown at the SuMN.Study 2Bi–v: The Effect of SuMN-THx on Feeding, Food Preference (HFD or RC), Hunger/Satiety Balance, and Feeding Response to Systemic Ghrelin or Glucagon-Like Polypeptide −1 (GLP-1) Agonist Administration in f-HFDr-RC Rats. Female HFDr rats were selected as described in detail above and randomly assigned into one of 2 groups receiving an inoculation of either AAV5-Th-shRNA-GFP (THx, N = 15) or AAV5-Scr-shRNA-GFP (Ctr, N = 10) at the medial SuMN and utilized in a series of sequential feeding paradigm studies as follows. Rats were allowed to rest for 1–2 weeks between experiments. Study 2Bi: The Impact of SuMN-THx on circadian food intake rhythm: Regular chow consumption in THx and Ctr rats was measured over a 24-h period for two consecutive days between ZT (hours after light onset) 0, 8, 14, 19. Food consumption during the light and dark periods of the day and total daily food consumption were calculated. Study 2Bii: The impact of SuMN-THx on HFD preference versus RC: Rats accustomed to RC for 3 weeks were given additional access to HFD, and preferential feeding was monitored daily during the subsequent 4 days (cumulative food consumption of each HFD and RC). On day 5 of the study, the RC was removed leaving HFD for 1 week. Next, RC was re-introduced in addition to the HFD and daily food consumption (HFD and RC) was subsequently monitored daily for the following 7 days. Study 2Biii: The impact of SuMN-THx on fasting-refeeding response: Rats maintained on RC were fasted overnight (16 h) during the normal feeding period of the day by removing RC 2 h before light offset (at ZT12). RC was re-introduced the next morning at ZT4, and cumulative food consumption was subsequently measured at 0.5 h, 1 h, 2 h, and 4 h after this initial refeeding time. Study 2Biv: The impact of SuMN-THx on ghrelin-induced food intake: Rats were injected ip with saline or ghrelin (Tocris Bioscience, Cat# 1,465) during the natural fasting (light) period of the day in a treatment cross-over design separated by a washout day. Rats (THx and Ctr) fed RC were injected with saline vehicle (0.2 mL/rat, ip) at ZT5.5–6, and cumulative baseline food consumption was measured at 1 h and 6 h post-vehicle injection. After a 1-day washout period, the THx and Ctr rats received a ghrelin injection (13 μg/kg, ∼0.2 mL/rat, ip) at ZT5.5–6. Cumulative RC consumption was measured at 1 h and 6 h post-ghrelin injection. Study 2Bv: The impact of SuMN-THx on exendin-4 (Ex-4)-induced hypophagia: THx and Ctr rats were injected with saline or the GLP-1 agonist, Ex-4 (Tocris Bioscience, Cat# 1933) during the natural feeding period (dark period of the day) in a treatment cross-over design separated by a washout day. On the day of study initiation, RC was removed from the THx and Ctr animals’ cages at 2 h before light offset (ZT12) and then re-introduced at the time of saline injection (ZT14–14.5). Cumulative RC consumption was measured under a red light at 1 h, 3 h, and 16 h post-saline injection to obtain baseline food consumption. After a 1-day washout period, following removal of the RC at ZT12, the THx and Ctr rats were then injected with Ex-4 (5 μg/kg, ip, 0.2 mL/rat) at ZT14–14.5 followed by an immediate re-introduction of RC, and cumulative RC consumption was measured under a red light at 1 h, 3 h, and 16 h post Ex-4 injection.Study 3: The Effect of SuMN-THx on Body Weight and Glucose Metabolism in f-HFDr-HFD RatsHFDr rats (BW = 265 ± 2.2 g, 15 weeks old) were selected after 1 week of HFD feeding based on % body weight gain less than the median % body weight gain (≤8%). Such selected HFDr rats were randomly divided into 2 groups receiving inoculation of AAV5-Th-shRNA-GFP (THx, N = 24) and AAV5-Scr-shRNA-GFP (Ctr, N = 18), respectively, at the medial SuMN, and fed HFD instead of regular chow during the study period. Body weight and food consumption were monitored. A GTT was performed at 7 weeks post-virus inoculation. At 8 weeks post-virus inoculation, rats were sacrificed by decapitation at the middle of the light period (ZT 6–8). The pWAT from 6 pairs of rats (6 THx and 6 Ctr) was collected for measurement of adipocyte OCR. The trunk blood was collected for plasma triglyceride assay.Study 4: The Effect of SuMN-THx on Body Weight and Glucose Metabolism in m-HFDr-RC ratsHFDr male rats (BW = 321 ± 3.8 g, 9 weeks old) were selected after 1 week of high fat feeding based on % body weight gain less than the median % body weight gain (≤17%) [30] and divided into 2 groups, receiving inoculation of AAV5-Th-shRNA-GFP (THx, N = 12) and AAV5-Scr-shRNA-GFP (Ctr, N = 12), respectively, at the medial SuMN. The rats were fed regular chow during the study period. Body weight and food consumption were monitored throughout the study period. A GTT was performed at 20 weeks post-virus inoculation.Study 5: Long-Term (1 Year) Metabolic Effect of SuMN-THx in f-HFDs-RC RatsHFD-sensitive rats (BW = 269 ± 2.2 g, 13 weeks old) with % body weight gain greater than the medium % body weight gain (>8%) after 1 week of high fat feeding were randomly divided into 2 groups and subsequently inoculated at the medial SuMN area with AAV5-Th-shRNA-GFP (THx, N = 10) or AAV5-Scr-shRNA-GFP (Ctr, N = 10) while maintained on regular chow during the study period. Body weight was monitored at the beginning, at 6 months and 12 months post-AAV inoculation. The rats were sacrificed by decapitation at 1-year post-virus inoculation during the fasting period of the day (ZT 6–8). The trunk blood was collected for the measurement of fasting glucose and insulin. The brains were collected for immunohistochemical verification of TH knockdown at the SuMN area.Blood/Plasma Assays of Metabolic ParametersBlood glucose was measured by a blood glucose monitor (OneTouch Ultra; LifeScan, Inc., Milpitas, CA, USA). Plasma insulin, leptin, and corticosterone concentrations were determined by EIA using commercially available assay kits (ALPCO Diagnostics, Salem, NH): insulin ELISA (80-INSMR), leptin ELISA (22-LEPMS), and corticosterone ELISA (55-CORMS). Belfiore insulin sensitivity index (ISI) (pmol/L*mmol/L*h) was calculated based on the formula: 2/((insulin GTT AUC * glucose GTT AUC) + 1). HOMA-IR was calculated based on the formula: fasting glucose (mmol/L) × fasting insulin (mU/L)/22.5. Hepatic triglyceride was extracted from liver tissue that was homogenized in 5% NP-40, heated to 90°C for 3 min, centrifuged at 10,000 g for 10 min to collect the supernatant. The triglyceride content in the supernatant was subsequently determined by TG Colorimetric Assay Kit (Item No. 10010303; Cayman Chemical, Ann Arbor, MI, USA). Plasma triglycerides were measured with Infinity™ Triglycerides Kit (TR22421; Thermo Fisher Scientific). Plasma free fatty acids (FFAs) were measured using a Free Fatty Acid Quantification Kit (MAK044; Sigma).Glucose Tolerance TestA GTT was performed during the fasting period of the day (ZT 6–8) by injecting rats with 50% dextrose solution (3 g/kg BW, ip, in all female rat studies; 2.5 g/kg BW, ip, in the male rat study 4). Blood samples were taken from the tail before and 30, 60, 90, 120 min after dextrose injection for blood glucose and plasma insulin analyses.qPCR Analysis of Hepatic mRNATotal RNA was isolated from frozen hepatic tissue samples with TRIzol Reagent (Cat # 15596026; Thermo Fisher Scientific, Waltham, MA, USA). Total RNA quantity and purity were determined by UV spectroscopy, and sample RNA concentrations were normalized prior to reverse transcription reaction. cDNA was synthesized using SuperScript IV VILO Master Mix with ezDNAse (Cat#11766050; Thermo Fisher Scientific, Waltham, MA, USA) from 2 μg of total RNA per sample. Real-time qPCR was performed with TaqMan fast advance master mix (Thermo Fisher Scientific, Waltham, MA, USA Cat# 4444964) on a AriaMx (Agilent, Santa Clara, CA, USA) qPCR equipment with the primer/probe sets as listed below. The relative expression of genes was calculated using the 2−∆∆Cq method. Ribosomal protein S17 was quantified with TaqMan assay (Cat# Rn00820807_g1; Thermo Fisher Scientific, Waltham, MA, USA) and used as the internal reference, and the relative gene expression was calculated as a fold change from control samples. Primers for respective mRNAs were purchased from Thermo Fisher Scientific (Waltham, MA, USA) as follows: phosphoenolpyruvate carboxy kinase 1 (Pck1) (Cat# Rn01529014_m1); glucose-6-phosphatase, catalytic subunit (G6pc) (Cat# Rn00689876_m1); sterol regulatory element binding transcription factor 1 (Srebf1) (Cat# Rn01495769_m1); fatty acid synthase (Fasn) (Cat# Rn00569117_m1); mechanistic target of rapamycin kinase (Mtor) (Cat# Rn00693900_m1); PPARG coactivator 1 beta (Ppargc1β) (Cat# Rn00598552_m1); peroxisome proliferator-activated receptor alpha (Pparα) (Cat# Rn00566193_m1); peroxisome proliferator-activated receptor gamma (Pparϒ) (Cat# Rn00440945_m1); PPARG coactivator 1 alpha (Ppargc1α) (Cat# Rn00580241_m1); nuclear factor kappa B subunit 1 (Rela) (Cat# Rn01502266_m1); suppressor of cytokine signaling 3 (Socs3) (Cat# Rn01470502_g1); mitogen-activated protein kinase 8 (Mapk8) (Cat# Rn01218952_m1); C-C motif chemokine ligand 2 (Ccl2) (Cat# Rn00580555_m1); nuclear receptor subfamily 1, group D, member 1 (Nr1d1) (Cat# Rn01460662_m1); period circadian regulator 1 (Per1) (Cat# Rn01496757_m1); Clock circadian regulator (Clock) (Cat# Rn00573120_ml).Analysis of OCRA 24-well format cell-based assay that measures fatty acid-driven oxygen consumption of adipocytes was conducted using the XFe24 Extracellular Flux Analyzer (Seahorse Bioscience, Agilent Technologies, Lexington, MA, USA). pWAT was freshly dissected from a TH knockdown rat and a control rat and placed in one XF24 Islet Capture Microplate (8 mg adipose tissue/per well) submerged in FAO assay medium. OCR (pmol/min/mg) was assessed at 37°C at baseline (basal) over 32 min (total oxygen consumption representing total tissue fuel oxidation) and in the subsequent presence of either 100 μm palmitate-bovine serum albumin or 400 μm etomoxir, a carnitine palmitoyltransferase-1 inhibitor over 64 min. Exogenous FAO (Extra-FAO) was calculated as palmitate-bovine serum albumin OCR above baseline, and endogenous FAO was calculated as baseline OCR minus etomoxir OCR (which represents non-FAO). Each of total oxidation, exogenous FFA oxidation, and endogenous FFA oxidation OCR values from control rats were normalized as 100%, and the relative % changes of THx versus control OCRs were calculated. The data were collected from 7 pairs of rats (7 THx vs. 7 Ctr) in Study 2A and 8 pairs of rats (8 THx vs. 8 Ctr) in Study 3, analyzed using the software “Wave” (Agilent Technologies).Data AnalysisAll data are expressed as mean ± standard error of the mean. Parametric tests were used when groups passed normality (using the Shapiro-Wilk test). Means between two groups were compared by two-tailed Student’s t test. In instances wherein the direction of change was anticipated, a one-tailed Student’s t test was employed. An observed p value <0.05 (p < 0.05) was accepted as statistically significant indicated by * (two-tail) or ƚ (one-tail). In multi-group design (>2 groups), data was analyzed by a two- or three-way repeated measures ANOVA followed by post hoc analysis of pair-wise comparisons using a Bonferroni-corrected α level. β indicates an observed p value less than a Bonferroni-corrected α level as significant. Only rats with the cannula tip localized inside, on the boundary of, or within 150 μm dorsal from the medial SuMN were included in statistical analyses.ResultsStudy 1: Verification of the Extent of THx at the SuMN Area (SuMN-THx)Among brains inoculated with AAV5-Th-shRNA-GFP or AAV5-Scr-shRNA-GFP, a comparable localized distribution of virus-transfected GFP-expressing neurons was observed at the SuMN area (bregma, from −4.4 mm to −4.9 mm) encompassing the medial SuMN and adjacent medial PH (within ∼400 μm dorsal to the medial SuMN). The extent of the viral-mediated TH knockdown was evaluated by TH immunohistochemistry in Th-shRNA versus Scr-shRNA brains. In the Scr-shRNA brains, distinct TH-immunopositive cell bodies and processes were observed mostly clustering within the medial SuMN and to a much lesser extent scattering in the adjacent PH dorsal to the medial SuMN (shown in Fig. 1A. a–f). Dense TH immunoreactivities were visible bilaterally at the VTA at the level of SuMN (shown in Fig. 1A. a, c, e). In contrast, the TH-immunopositive cell bodies in Th-shRNA brains were almost completely eliminated at such defined SuMN area while leaving the TH-immunopositive processes moderately detectable (shown in Fig. 1A. g–l). At the adjacent bilateral VTA, robust TH-immunopositive bodies were detected in the Th-shRNA brains with a density comparable to that in Scr-shRNA brains (shown in Fig. 1A. g, i, k). Immunofluorescence labeling revealed that the virus-transfected GFP-expressing cells in the medial SuMN express TH(+)-RRX red fluorescence in the control Scr-shRNA brain (shown in Fig. 1B. b–d), while in Th-shRNA brains TH(+)-RRX red cell bodies were barely detectable at the medial SuMN (shown in Fig. 1B. g–i). Cell counts of TH(+)-DAB cells on 6 brain sections across the SuMN area (bregma −4.4, −4.5, −4.6, −4.7, −4.8, −4.9 mm) revealed 83% reduction in TH(+)-DAB cells in Th-shRNA brains compared to Scr-shRNA brains (p < 0.0001) (shown in Fig. 1A. m, n). At the adjacent VTA, no detectable reduction in the TH(+)-RRX immunofluorescence in Th-shRNA brains versus Scr-shRNA brains was found via measurement of the mean fluorescence density (bregma, −4.5 mm) or TH(+)-red cell count (bregma −4.8 mm) (shown in Fig. 1C). Thus, the TH knockdown was largely restricted to the medial SuMN and to the sparsely scattered TH-positive neurons in the adjacent PH representing a near complete knockdown of the total TH dopaminergic neurons in such defined SuMN area. Whereas no significant difference in the counts of DAPI nuclear staining at the SuMN area was detected in Th-shRNA brains versus Scr-shRNA brains (data not shown), the TH knockdown at the SuMN area (SuMN-THx) remains relatively stable over the experimental periods as verified by TH immunohistochemistry in the Th-shRNA brains collected at 17 weeks (from Study 2A) and 1 year (from Study 5) after surgical inoculation of the viral construct AAV5-Th-shRNA-GFP (data not shown).Fig. 1.A Verification of Th gene expression knockdown at the SuMN area with TH-immunopositive DAB staining. The extent of Th gene expression knockdown via a virus-mediated shRNA was evaluated by TH immunohistochemistry using DAB as a chromogen. The images represent the TH-immunopositive DAB staining [TH(+)-DAB] of Scr-shRNA versus Th-shRNA brain sections at the rostral (bregma −4.52), middle (bregma −4.70), and caudal (bregma −4.80) levels of the SuMN. In Scr-shRNA brain sections (low magnification (a, c, e) and high magnification (b, d, f)), the DAB staining revealed TH(+)- cell bodies and processes clustered at the medial SuMN and scattered around the adjacent medial posterior hypothalamic area dorsal to the medial SuMN. Dense TH(+)- immunoreactivity at the VTA was detected bilaterally. In Th-shRNA brain sections (low magnification (g, i, k) and high magnification (h, j, l)), TH(+)- cell bodies were no longer detectable in the medial SuMN and dramatically reduced in the adjacent medial PH (within ∼400 μm from dorsal SuMN) leaving TH(+)- processes readily visible. Strong and comparable TH(+)- immunoreactivity was bilaterally detected at the VTA on Th-shRNA and Scr-shRNA brain sections. Cell counts of TH(+)-DAB cells on 6 brain sections across the SuMN area (bregma −4.4, −4.5, −4.6, −4.7, −4.8, −4.9 mm) revealed an 83% reduction in TH(+)-DAB cells in Th-shRNA brains compared to Scr-shRNA brains (p < 0.001) (m, n). The red dotted lines demarcate the contour of the medial SuMN. SuMN m, supramammillary nucleus, medial; mp, mammillary peduncle; VTAr, ventral tegmental area, rostral; SN, substantia nigra; 3V, 3rd ventricle. *p < 0.05, two-tailed Student’s t test. B TH-immunofluorescent staining at the SuMN area. The TH-immunopositive cells were labeled with RRX (red) fluorescence in Scr-shRNA brain (a–e) and Th-shRNA brain (f–j) samples. A cluster of TH(+)- red fluorescent cells was detected in the medial SuMN in Scr-shRNA brain (a, b), while no distinct red fluorescent cell bodies were detected in Th-shRNA brain (f, g). Virus-transfected neuronal cells expressed GFP in Scr-shRNA brain (c) and Th-shRNA brain (h). The zoom-in images showed red fluorescence inside the virus-transfected GFP-expressing cells in Scr-shRNA brain as indicated by arrows (b, c, d). No red immunofluorescence was observed inside the GFP-expressing cells in Th-shRNA brain (g, h, i). DAPI nuclear staining showed apparent normal histological morphology in Scr-shRNA (e) and Th-shRNA (j) brains. C TH-immunofluorescent staining at the VTA TH (+)- red immunofluorescent staining was quantified by mean fluorescence density and cell counts at the VTA. a, b Representative images of TH(+)- red immunofluorescence at the VTA adjacent to rostral SuMN (bregma −4.2 mm). There was no reduction in the mean density of red fluorescence in the Th-shRNA brain versus Scr-shRNA brain (p = 0.333) (c). d, e Representative images of TH(+)- red immunofluorescence at the VTA adjacent to caudal SuMN (bregma −4.8 mm). There was no reduction in the TH(+)- immunofluorescent red cell counts in shRNA brain versus Scr-shRNA brain (p = 0.804) (f).Study 2A: SuMN-THx Induces Obese MS (Metabesity) in f-HFDr-RC RatsStudy 2Ai: SuMN-THx Induces Obesity and Glucose Intolerance and Reduces Adipose FAO RateCompared to Ctr rats, measured at 11 weeks post-virus inoculation, THx rats exhibited a 42% increase in body weight gain (50 ± 3.5 g gain vs. 35 ± 2.3 g gain in Ctr rats, p = 0.003) (shown in Fig. 2a, b). The body weight measured over time showed a significant main effect of time (two-way repeated measures ANOVA, F(12, 240) = 114.47, p < 0.001), but no significant main effect of THx (F(1, 20) = 3.02, p = 0.098). The interaction between time and THx was significant (F(12, 240) = 4.48, p < 0.001). The THx treatment relative to Ctr also increased food consumption by 15% (p = 0.0001) (shown in Fig. 2c, d) with a 25% increase in feed efficiency (p = 0.048) (shown in Fig. 2e). The food consumption over time showed a significant main effect of time (F(11, 220) = 2,899.36, p < 0.001) and of THx (F(1, 20) = 15.89, p = 0.001) as well as a significant interaction between time and THx (F(11, 220) = 13.31, p < 0.001). By the time of sacrifice at 17 weeks post-virus inoculation, THx rats had gained 83 ± 4.8 g in body weight over the 17-week period which was 48% more than Ctr rats (body weight gain = 56 ± 5.5 g) (p = 0.001) (shown in Fig. 2a). The WAT mass was significantly increased in THx rats compared to Ctr rats: parametrial (p)WAT increased by 45% (p = 0.008); retroperitoneal (r) WAT increased by 37% (p = 0.041); and total (tl) WAT increased by 42% (p = 0.012) (shown in Fig. 2f). Finally, FAO of pWAT was reduced in THx versus Ctr rats (total basal oxidation by 51%, p < 0.0001; endogenous FAO by 58%, p < 0.0001; Extra-FAO by 54%, p = 0.024) (shown in Fig. 2g). Thus, such THx treatment increased body weight/fat and feed efficiency while reducing free FAO of adipocytes. Compared with Ctr, THx treatment induced glucose intolerance as indicated by an elevated GTT blood glucose level over time (at 0, 30, 60, 90, and 120 min) resulting in an increase in GTT glucose AUC by 27% relative to Ctr (p = 0.009) (shown in Fig. 3b). The basal blood glucose level was elevated in THx rats compared to Ctr rats (p = 0.024) (shown in Fig. 3a, insert). The results of the GTT demonstrated a main effect of time (two-way repeated measures ANOVA, F(4, 60) = 85.26, p < 0.001) and a main effect of THx (F(1, 15) = 0.007) with no significant interaction between time and THx treatment (F(4, 60) = 1.73, p = 0.154) in GTT glucose. However, only a significant main effect of time (F(4, 60) = 29.28, p < 0.001) was detected in GTT insulin with no significant main effect of THx (F(1, 15) = 0.001, p = 0.978) nor significant interaction between time and THx (F(4, 60) = 0.142, p = 0.966). Thus, GTT insulin AUC was not altered by THx treatment (shown in Fig. 3d), and as a result, there was no significant reduction in Belfiore ISI as compared with Ctr (17%, p = 0.258) (shown in Fig. 3e).Fig. 2.THx at the SuMN area induces obesity, increased feeding and feeding efficiency while decreasing FAO in f-HFDr-RC rats (Study 2Ai). f-HFDr-RC rats were locally inoculated at the medial SuMN area with AAV5-Th-shRNA-GFP to knockdown Th gene expression (THx) or control AAV5-Scr-shRNA-GFP (Ctr). Compared with Ctr, THx treatment increased body weight gain by 42% (p = 0.003) (a, b) and food consumption by 15% (p = 0.0001) (c, d) with a concurrent increase in feed efficiency (25%, p = 0.048) over 72 days (e) post-virus inoculation. When sacrificed at 17 weeks post-virus inoculation, THx rats relative to Ctr rats had an increase in body weight gain (48%, p = 0.001) (a) and WAT weight (parametrial pWAT by 45%, p = 0.008; retroperitoneal rWAT by 37%, p = 0.041; total tl WAT by 42%, p = 0.012) (f). Fatty acid oxidation (FAO) of pWAT (8 mg) was reduced by THx treatment versus Ctr (basal FAO, 51%, p < 0.0001; endogenous FAO [Endo-FAO], 58%, p < 0.0001; exogenous FAO [Exo-FAO], 54%, p = 0.024) (g). *p < 0.05, two-tailed Student’s t test.Fig. 3.THx at the SuMN area induces glucose intolerance in f-HFDr-RC rats (Study 2Ai). f-HFDr-RC rats were locally inoculated at the medial SuMN area with AAV5-Th-shRNA-GFP to knockdown Th gene expression (THx) or control AAV5-Scr-shRNA-GFP (Ctr). At 11 weeks post-virus inoculation, compared with Ctr, THx treatment increased basal blood glucose (p = 0.024) and GTT blood glucose level at 30, 60, 90, and 120 min following glucose challenge (a) with an increase in GTT glucose AUC (27%, p = 0.009) (b). GTT insulin AUC was not increased by THx treatment (c, d). The reduction in Belfiore ISI did not reach significance in THx rats versus Ctr rats (17% reduction, p = 0.258) (e). *p < 0.05, two-tailed Student’s t test.Study 2Aii: SuMN-THx Alters the mRNA Expression Levels of Metabolic, Inflammatory, and Clock Genes in the LiverThe relative mRNA expression levels of key metabolic, inflammatory, and circadian genes were analyzed using RT-qPCR in the livers collected from THx and Ctr rats during the fasting period of the day (ZT 6–8) at 17 weeks following viral inoculation. Compared with Ctr, THx increased relative mRNA expression of key metabolic genes involved in gluconeogenesis, Pck1 and G6pc, by 2.3-fold (p = 0.0013) and 1.6-fold (p = 0.046), respectively (shown in Fig. 4a, gluconeogenic genes). Compared with Ctr, THx increased the relative mRNA expression of the key lipogenic transcriptional coregulator Ppargc1β by 1.4-fold (p = 0.037 Ɨ) with no significant increase in Fasn gene (p = 0.6) or the lipogenic transcriptional regulators (Srebf1, p = 0.9; Mtor, p = 0.9) (shown in Fig. 4b, lipogenic genes). The mRNA expression levels of the key transcriptional factors/coactivators regulating FAO were not significantly altered in THx relative to Ctr rats: Pparα (p = 0.105), Pparγ (p = 0.314), or Ppargc1α (p = 0.203) (shown in Fig. 4c, lipid oxidation genes). Compared with Ctr, THx increased the relative mRNA expression levels of Rela (1.2-fold, p = 0.021) and Ccl2 (1.4-fold, p = 0.037 Ɨ) with a non-significant increase in Socs3 (p = 0.148) and Mapk8 (p = 0.106) (shown in Fig. 4d, pro-inflammatory genes). Compared with Ctr, THx treatment increased the relative mRNA expression levels of Clock (1.4-fold, p = 0.0002) and Per1 (1.8-fold, p = 0.036 Ɨ) with no increase in Nr1d1 (p = 0.33) (shown in Fig. 4e, circadian genes).Fig. 4.THx at the SuMN area alters mRNA expression of metabolic relevant genes in the liver of f-HFDr-RC rats (Study 2Aii). f-HFDr-RC rats were locally inoculated at the medial SuMN area with AAV5-Th-shRNA-GFP to knockdown Th gene expression (THx) or control AAV5-Scr-shRNA-GFP (Ctr). The relative mRNA expression levels of metabolic relevant genes were analyzed by RT-qPCR in the livers of THx and Ctr rats collected during the fasting period of the day (ZT 6–8). a Gluconeogenic genes: compared with Ctr, THx treatment increased relative mRNA expression of phosphoenolpyruvate carboxylase 1 (Pck1) by 2.3-fold (p = 0.0013) and glucose-6-phosphatase, catalytic subunit (G6pc) by 1.6-fold (p = 0.046). b Lipogenic genes: compared with Ctr, THx increased the relative mRNA expression of lipogenic transcriptional coregulator Ppargc1β by 1.4-fold (p = 0.037 Ɨ) with no detectable increase in fatty acid synthase (Fasn) (p = 0.6) and lipogenic transcriptional regulators (Srebf1, p = 0.9; Mtor, p = 0.9). c Lipid oxidation genes: no significant increase was detected in the mRNA expression levels of the key transcriptional factors/coactivators regulating FAO in the livers of THx rats versus Ctr rats, Pparα (p = 0.105), Pparγ (p = 0.314), or Ppargc1α (p = 0.203). d Pro-inflammatory genes: compared with Ctr, THx treatment increased the relative mRNA expression levels of Rela (1.2-fold, p = 0.021) and Ccl2 (1.4-fold, p = 0.037 Ɨ) with no significant increase in suppressor of cytokine signaling 3 (Socs3) (p = 0.148) and mitogen-activated protein kinase 8 (Mapk8) (p = 0.106). e Circadian genes: compared with Ctr, THx treatment increased the relative mRNA expression levels of Clock (1.4-fold, p = 0.0002) and Per1 (1.8-fold, p = 0.036 ƚ) with no effect on Nr1d1 (p = 0.33). *p < 0.05, two-tailed Student’s t test. Ɨp < 0.05, one-tailed Student’s t test.Study 2Aiii: SuMN-THx Induces Hyperleptinemia and Hypertriglyceridemia, and Alters the Plasma Level of CorticosteroneA prototypical profile of metabolic parameters associated with MS [16, 36] was induced by THx at the SuMN. Compared with Ctr, THx treatment increased plasma leptin (46%, p = 0.0382 Ɨ) (shown in Fig. 5a) and plasma triglyceride (86%, p = 0.008) levels (shown in Fig. 5b); however, there was no significant increase in liver triglyceride (p = 0.235) (shown in Fig. 5c) or plasma FFAs (p = 0.153) (shown in Fig. 5d). The plasma corticosterone level assessed during the fasting period of the day (ZT 6–8) was reduced by 68% (p = 0.007) in THx rats versus control rats (shown in Fig. 5e) suggesting a change/shift in the plasma corticosterone circadian rhythm.Fig. 5.THx at the SuMN area induces hyperleptinemia, hypertriglycemia and alters plasma corticosterone in f-HFDr-RC rats (Study 2Aiii). f-HFDr-RC rats were locally inoculated at the medial SuMN area with AAV5-Th-shRNA-GFP to knockdown Th gene expression (THx) or control AAV5-Scr-shRNA-GFP (Ctr). Compared with Ctr, THx treatment increased plasma leptin by 46% (p = 0.0382 Ɨ) (a) and plasma triglyceride by 86% (p = 0.008) (b) with no increase in liver triglyceride content (p = 0.235) (c) or plasma level of FFAs (p = 0.153) (d). Plasma corticosterone measured during fasting period of the day (ZT 6–8) was reduced by 68% (p = 0.007) (e) in THx rats versus control rats. *p < 0.05, two-tailed Student’s t test. ƚp < 0.05, one-tailed Student’s t test.Study 2B: SuMN-THx Alters the Circadian Feeding Rhythm, Increases HFD Consumption in Food Preference Paradigms, Alters Hunger/Satiety Balance, and Attenuates the Feeding Response to Systemic Ghrelin Administration (Ghrelin Resistance) in f-HFDr-RC RatsStudy 2Bi: SuMN-THx Increases Food Consumption during the Natural Fasting Period of the DayRegular chow consumption was monitored throughout light (resting/fasting) and dark (active/feeding) periods of the day in THx and Ctr rats over 2 days (shown in Fig. 6a1). There was a significant main effect of light/dark photoperiod of the day (two-way repeated measures ANOVA, F(1, 24) = 363.43, p < 0.001) such that rats ate mostly during the dark and less during the light period (dark: 15.7 ± 0.51g/rat; light: 3.3 ± 0.23g/rat). There was a significant main effect of TH knockdown (THx) (F(1, 24) = 7.07, p = 0.014) on total daily food consumption such that THx rats ate 14% more daily food than Ctr rats (10.12 ± 0.3g/THx rat vs. 8.89 ± 0.35g/Ctr rat) (shown in Fig. 6a2). The interaction between photoperiod and SuMN-THx was not significant (F(1, 24) = 0.75, p = 0.394) reflecting that THx rats maintain a circadian feeding rhythm with peak food intake during the dark period similar to Ctr rats. Post hoc analysis of pair-wise comparisons using Bonferroni-corrected α-level further revealed that THx rats as compared to Ctr rats consumed 75% more food (4.2 ± 0.31 g/THx rat vs. 2.4 ± 0.36g/Ctr rat, p < 0.001) during the light period, while no significant increase in food intake was detected in THx rats versus Ctr rats (16.1 ± 0.67g/THx rat vs. 15.4 ± 0.78 g/Ctr rat, p = 0.523) during the dark period (shown in Fig. 6a3). Of their total daily food consumption, THx rats consumed 21% during the light and 79% during the dark periods, whereas Ctr rats consumed 14% during the light and 86% during the dark photoperiod (p = 0.001). Thus, compared with controls, SuMN-THx treatment increased the daily food consumption specifically during the light period (normal fasting period) (by 75%) while maintaining the peak food intake during the dark period of the day.Fig. 6.Effect of THx at the SuMN area on circadian feeding rhythm, food preference, hunger/satiety balance, and responses to systemic ghrelin and GLP-1 agonist administration in f-HFDr-RC rats (Study 2Bi-v). f-HFDr-RC rats were locally inoculated at the medial SuMN area with AAV5-Th-shRNA-GFP to knockdown Th gene expression (THx) or control AAV5-Scr-shRNA-GFP (Ctr). SuMN-THx increased food consumption during the light (fasting/resting) period of the day (a1–3). Regular chow consumption was monitored throughout light and dark periods of the day in THx and Ctr rats over two consecutive days. THx rats, like Ctr rats, maintained a circadian feeding rhythm with peak food intake during the dark photoperiod (a1). Nonetheless, THx rats ate 14% more food per day than Ctr rats (p = 0.014) (a2) which was consumed particularly during the light period. THx rats as compared to Ctr rats consumed 75% more food (p < 0.001) during the light period with no difference in food consumption during the dark period (p = 0.523) (a3). SuMN-THx did not alter food preference of HFD over RC, but increased HFD consumption (b1–2). THx rats and Ctr rats were exposed to RC and HFD simultaneously, and food consumption was measured daily. In both scenarios where the order of HFD and RD given was switched, THx and Ctr rats showed the same strong food preference for HFD over RC. While no difference in RC consumption was observed between THx rats and Ctr rats, THx rats consumed more HFD than Ctr rats by 17% (p = 0.002) over 4 days when HFD was given as novel add-on diet (b1) and by 16% (p = 0.0226) over 7 days when RC was given as novel add-on diet (b2). SuMN-THx reduced fasting-primed refeeding (c). Following overnight fasting (16 h), THx and Ctr rats were refed RC (starting at ZT4) during the light period, and cumulative RC consumption was measured at 0.5 h, 1 h, 2 h, and 4 h post-refeeding time. After an initial peak in food intake at 0–0.5 h by both THx rats and Ctr rats, there was a 67% reduction in food consumption in THx rats versus Ctr rats at 0.5–1 h (p < 0.001) and at 1–2 h (p = 0.027). SuMN-THx reduced ghrelin-stimulated feeding (d). THx rats and Ctr rats were injected with ghrelin or vehicle during their natural fasting period of the day (at ZT5.5–6, light period), and cumulative RC consumption was measured at 1 h and 6 h after drug administration. At 0–1 h, there was a non-significant elevation in basal food consumption in Veh/THx versus Veh/Ctr rats (p = 0.16). Ghrelin significantly increased food consumption in Ctr and THx rats by 200% (p < 0.001) and 84% (p < 0.001), respectively, versus their respective vehicle controls. At 1–6 h, ghrelin significantly decreased food consumption by 75% (p = 0.002) in THx rats versus vehicle-treated THx rats, and by 55% (ns, p = 0.141) versus ghrelin-treated Ctr rats. SuMN-THx did not alter Ex-4-induced satiety (e). THx rats and Ctr rats were injected with Ex-4 or vehicle at the beginning of the dark period of the day (ZT14–14.5). Cumulative RC consumption was measured under a red light at 1 h, 3 h, and 16 h after the drug administration during the natural feeding period of the day. Ex-4 inhibited food consumption in THx rats to the same magnitude as in Ctr rats at 0–1 h and 1–3 h, and the Ex-4 effect subsided at 3–16 h in both Ctr and THx rats. At 0–1 h, as compared to vehicle, Ex-4 reduced food consumption by 69% in THx rats (p < 0.001) and by 70% in Ctr rats (p < 0.001), and at 3-1 h by 67% in THx rats (p < 0.001) and by 70% in Ctr rats (p < 0.001). *p < 0.05, two-tailed Student’s t test. β indicates an observed p value is less than Bonferroni-corrected α-level.Study 2Bii: SuMN-THx Increases HFD Consumption with No Change in HFD Preference to RCTHx and Ctr rats were exposed to RC and HFD simultaneously, and food consumption was measured daily. First, rats maintained on RC were subsequently additionally provided HFD as an add-on diet. The consumption of RC and HFD was then monitored the ensuing 4 days. There was a significant main effect of diet (two-way repeated measures ANOVA, F(1, 23) = 1,558.78, p < 0.001) reflecting that both groups of rats preferred HFD versus RC (HFD: 76.31 ± 1.81 g; RC: 1.43 ± 0.49 g). However, there was a significant main effect of THx (F(1, 23) = 9.98, p = 0.004) and a significant interaction between diet and THx (F(1,23) = 10.36, p = 0.004). Post hoc analysis revealed that THx rats consumed 17% more HFD than Ctr rats (82.3 ± 2.4 g/THx rat vs. 70.33 ± 2.71 g/Ctr rat, p = 0.002) over the 4-day study period (shown in Fig. 6b1). Second, rats maintained on HFD were subsequently additionally given RC as an add-on dietary choice and consumption of RC and HFD was then measured over the next 7 days. Once again, there was a significant main effect of diet (F(1, 23) = 509.68, p < 0.001) such that rats consumed predominantly HFD with limited RD consumption due to its novel exposure (HFD: 101.7 ± 3.11 g; RC: 13.6 ± 1.84 g) (shown in Fig. 6b2). No significant main effect of THx (F(1, 23) = 3.21, p = 0.086) was detected, but the interaction between diet and THx was significant (F(1, 23) = 5.67, p = 0.026). Post hoc analysis revealed that THx rats consumed 16% more HFD than Ctr rats (109.3 ± 4.12 g/THx rat vs. 94.1 ± 4.65 g/Ctr rat, p = 0.0226) over the 7-day study period (shown in Fig. 6b2). The RC consumption was largely confined to the first 2 days of novel exposure of RC, and no difference in RC consumption between THx rats and Ctr rats was found (11.86 ± 2.44 g/THx rat vs. 15.25 ± 2.75 g/Ctr rat, p = 0.814). Thus, when given a choice to consume HFD and RC, THx and Ctr rats similarly strongly preferred HFD over RC. However, in both experimental scenarios, THx rats consistently consumed more HFD (by 16∼17%) than Ctr rats.Study 2Biii: SuMN-THx Reduces Overnight Fasting-Enhanced, Morning Refeeding Food ConsumptionFollowing overnight fasting (16 h), THx and Ctr rats were refed RC (starting at ZT4) during the light/fasting period of the day, and cumulative RC consumption was measured at 0.5 h, 1 h, 2 h, and 4 h post-refeeding time. There was a significant main effect of time (F(3, 72) = 53.91, p < 0.001) reflecting an initial peak of food intake (3.3 ± 0.2 g at 0–0.5 h) followed by a gradual reduction over the next 4 h (0.24 ± 0.12 g at 2–4 h). There was a significant main effect of THx (F(1, 24) = 410.15, p < 0.001) and a significant interaction between time and THx (F(3, 72) = 0.041). Post hoc analysis (pair-wise comparisons using Bonferroni-corrected α level) revealed a 67% reduction in food consumption in THx rats versus Ctr rats at 0.5–1 h (0.65 ± 0.21 g/THx rat vs. 1.97 ± 0.24 g/Ctr rat, p < 0.001) and at 1–2 h (0.44 ± 0.25 g/THx rat vs. 1.33 ± 0.29 g/Ctr rat, p = 0.027) (shown in Fig. 6c). Thus, while SuMN-THx increases food consumption (both of RC and HFD) during light/fasting period of the day under free living, ad libitum conditions, when feeding is examined in a fasting-refeeding paradigm SuMN-THx actually reduces the refeeding response, suggestive of ghrelin resistance.Study 2Biv: Ghrelin Reduces Day-Time Food Intake in THx Rats despite Undiminished Ghrelin-Induced Acute HyperphagiaTo further investigate the above-described SuMN-THx decrease of the refeeding response, suggestive of ghrelin resistance, a study was conducted to measure feeding response to systemic ghrelin administration (shown in Fig. 6d). THx and Ctr rats were injected with ghrelin or vehicle during their natural fasting period of the day (at ZT5.5−6 light period), and cumulative RC consumption were measured at 1 h and 6 h after drug administration. There was a significant main effect of ghrelin (F(1, 23) = 7.202, p = 0.013) and a significant main effect of time (F(1, 23) = 35.453, p < 0.001). The interaction between ghrelin and time was also significant (F(1, 23) = 61.027, p < 0.001). Ghrelin induced an increase in food consumption at 0–1 h and the effect was terminated at 1–6 h. There was no significant main effect of THx (F(1, 23) = 0.049, p = 0.826), but the interaction between THx and ghrelin was significant (F(1, 23) = 5.58, p = 0.027). Post hoc analysis (pair-wise comparisons using Bonferroni-corrected α-level) further revealed that at 0–1 h, ghrelin stimulated food intake in Ctr rats by 200% (2.7 ± 0.184 g vs. 0.9 ± 0.231 g, p < 0.001) but by only 84% in THx rats (2.45 ± 0.15 g vs. 1.33 ± 0.189 g, p < 0.001), mainly due to an elevated basal level of food intake in THx rats (Veh/THx rats: 1.33 ± 0.189 g vs. Veh/Ctr rats: 0.9 ± 0.231 g, p = 0.16). The 38% decrease in ghrelin-induced acute hyperphagia level in THx rats versus Ctr rats however did not reach significance (p = 0.12). At later 1–6 h, no difference in food consumption was detected between ghrelin/Ctr and vehicle/Ctr groups, however, SuMN-THx treatment actually reduced the feeding response to ghrelin by 75% compared with vehicle-injected THx rats (0.33 ± 0.169 g vs. 1.35 ± 0.266 g, p = 0.002). Thus, in conjunction with (and possibly due to) a higher baseline level of feeding among SuMN-THx rats over the 6 h test period, such animals were overall less sensitive to ghrelin-induced food intake than Ctr animals.Study 2Bv: SuMN-THx Does Not Alter GLP-1 Agonist (Ex-4)-Induced HypophagiaTHx and Ctr rats were injected with Ex-4 or vehicle at the beginning of the dark period (initiation of normal feeding time) (ZT14-14.5). Cumulative RC consumption was subsequently measured under red light at 1 h, 3 h, and 16 h after drug administration. There was a significant main effect of Ex-4 (F(1, 24) = 12.52, p = 0.002) and a significant main effect of time (F(2, 48) = 242.83, p < 0.001). The interaction between Ex-4 and time was also significant (F(2, 48) = 10.21, p < 0.001). Ex-4 treatment induced hypophagia in rats at early 0–1 h and 1–3 h but not at later 3–16 h versus vehicle. There was no significant main effect of THx (F(1, 24) = 1.65, p = 0.211), nor its interactions with Ex-4 (F(1, 24) = 0.25, p = 0.621) or with time (F(2, 48) = 0.09, p = 0.918). Post hoc analysis revealed that Ex-4 inhibited food consumption in THx rats to the same magnitude as in Ctr rats at 0–1 h by 69% in THx rats (Ex-4/THx rat: 0.79 ± 0.266 g vs. Veh/THx rat: 2.57 ± 0.315 g, p < 0.001) and by 70% in Ctr rats (Ex-4/Ctr rat: 0.99 ± 0.31 g vs. Veh/Ctr rat: 3.29 ± 0.368 g, p < 0.001), as well as at 3-1 h by 67% in THx rats (Ex-4/THx rat: 0.96 ± 0.236 g vs. Veh/THx rat: 2.9 ± 0.319 g, p < 0.001) and by 70% in Ctr rats (Ex-4/Ctr rat: 1.08 ± 0.276 g vs. Veh/Ctr rat: 3.59 ± 0.373 g, p < 0.001) (shown in Fig. 6e).Study 3: Metabolic Effects of SuMN-THx in Female HFDr Rats Are Enhanced in Those Fed HFD (f-HFDr-HFD)f-HFDr-HFD rats were inoculated with AAV5-Th-shRNA-GFP (THx) or AAV-Scr-shRNA-GFP (Ctr). Compared with control, THx increased GTT glucose AUC (16%, p = 0.043) (shown in Fig. 7a, b) and insulin AUC (62%, p = 0.001) (shown in Fig. 7c, d) resulting in a reduction in Belfiore ISI (39%, p < 0.0001) (shown in Fig. 7e). Both the basal glucose (p = 0.006) (shown in Fig. 7a, insert) and basal insulin (p = 0.04) (shown in Fig. 8c) were elevated in THx rats compared with Ctr rats. For both GTT glucose and GTT insulin, there was a main effect of time (Fglucose (4, 160) = 230.24, p < 0.001; Finsulin (4, 160) = 26.83, p < 0.001) and a main effect of THx (Fglucose (1, 40) = 5.8, p = 0.021; Finsulin(1, 40) = 112, p < 0.001) but no significant interaction between time and THx (Fglucose (4, 160) = 2.005, p = 0.133; Finsulin (4, 160) = 0.836, p = 0.504). Such THx treatment increased body weight gain by 23% (67 ± 3.4 g gain vs. 55 ± 2.2 g gain in Ctr rats, p = 0.006) at 37 days after SuMN-THx treatment (shown in Fig. 7f, g) and increased food consumption by 7% (p = 0.024) (shown in Fig. 7h, i) and feed efficiency by 14% (p = 0.008) (shown in Fig. 7j) relative to Ctr. For body weight and food consumption, there was a significant effect of time and no significant effect of THx, and the interaction between time and THx was significant (body weight [Ftime (11, 440) = 372.62, p < 0.001; FTHx (1, 40) = 2.36, p = 0.132; Ftime*THx (11, 440) = 6.7, p < 0.001]; food [Ftime (10, 440) = 3,435.04, p < 0.001; FTHx (1, 40) = 1.94, p = 0.171; Ftime*THx (11, 440) = 6.7, p < 0.001]). A reduction in FAO in pWAT was detected in THx versus Ctr rats (basal FAO by 29%, p < 0.0001; endogenous FAO by 39%, p < 0.0001; exogenous FAO by 53%, p = 0.029) (shown in Fig. 7k). The plasma triglyceride level was increased by 15% (p = 0.033 ƚ) (shown in Fig. 7l).Fig. 7.THx at the SuMN area induces glucose intolerance/IR and increases body weight gain in f-HFDr-HFD rats (Study 3). f-HFDr-HFD rats were locally inoculated at the medial SuMN area with AAV5-Th-shRNA-GFP to knockdown Th gene expression (THx) or control AAV5-Scr-shRNA-GFP (Ctr). Compared with Ctr, THx treatment increased GTT glucose AUC (16%, p = 0.043) (a, b) and insulin AUC (62%, p = 0.001) (c, d) resulting in a reduction in Belfiore ISI (39%, p < 0.0001) (e). THx treatment increased body weight gain by 23% (p = 0.06) (f, g) and increased food consumption by 7% (p = 0.024) (h, i) and feed efficiency by 14% (p = 0.008) (j) relative to Ctr. Compared with Ctr, THx treatment reduced FAO in pWAT (8 mg) (basal FAO by 29%, p < 0.0001; endogenous FAO [Endo-FAO] by 39%, p < 0.0001; exogenous FAO [Exo-FAO] by 53%, p = 0.029) (k). The plasma triglyceride level was increased by 15% in THx rats relative to Ctr rats (p = 0.033 Ɨ) (l). *p < 0.05, two-tailed Student’s t test. Ɨp < 0.05, one-tailed Student’s t test.Fig. 8.THx at the SuMN area induces glucose intolerance and increases body weight gain in m-HFDr-RC rats (Study 4). m-HFDr-RC rats were locally inoculated at the medial SuMN area with AAV5-Th-shRNA-GFP to knockdown Th gene expression (THx) or control AAV5-Scr-shRNA-GFP (Ctr). Compared with Ctr, THx treatment increased GTT glucose AUC by 38% (p = 0.014) (a, b) with no change in insulin AUC (c, d) resulting in 28% reduction in Belfiore ISI (p = 0.015) (e). The THx rats exhibited an increase in body weight gain by 12% (p = 0.02) (f, g) and 7% increase in food consumption (p = 0.04) (h, i) relative to Ctr rats. There was no significant increase in feed efficiency (p = 0.137) (j) in THx rats compared to Ctr rats. *p < 0.05, two-tailed Student’s t test.Study 4: SuMN-THx Induces Glucose Intolerance in m-HFDr-RC RatsTo evaluate the influence of gender in the metabolic effects of SuMN-THx, m-HFDr-RC rats were locally inoculated at the medial SuMN area with AAV5-Th-shRNA-GFP to knockdown Th gene expression (THx) or control AAV5-Scr-shRNA-GFP (Ctr) as described above. Compared with control, THx increased GTT glucose AUC by 38% (p = 0.014) (shown in Fig. 8a, b) with no change in insulin AUC (shown in Fig. 8c, d) resulting in 28% reduction in Belfiore ISI (p = 0.015) indicating an increase in IR (shown in Fig. 8e). The basal blood glucose was elevated in THx rats versus Ctr rats (p = 0.031) (shown in Fig. 8a, insert). The GTT glucose response exhibited significant main effects of time, THx, and their interactions (Ftime (4,88) = 100.57, p < 0.001; FTHx (1, 22) = 288.75, p < 0.001; Ftime*THx (4, 88) = 4.55, p = 0.002), while GTT insulin only showed a significant main effect of time (F(4, 88) = 6.85, p < 0.001) with no significant main effect of THx (F(1, 22) = 0.74, p = 0.788) or THx and time interaction (F(4, 88) = 1.08, p = 0.37). SuMN-THx increased body weight gain by 12% (287 ± 9.1 g gain vs. 257 ± 7.7 g gain in Ctr rats, p = 0.02) (shown in Fig. 8f, g) and food consumption by 7% (p = 0.04) (shown in Fig. 8h, i) relative to control rats. However, such treatment did not result in any significant increase in feed efficiency (p = 0.137) (shown in Fig. 8j) in THx rats compared to Ctr rats. There was a significant main effect of time for body weight (F(10, 220) = 1,515.71, p < 0.001) and for food consumption (F(9, 198) = 4,138.42, p < 0.001). The main effect of THx was not significant for body weight (F(1, 22) = 0.34, p = 0.566) but was significant for food (F(1, 22) = 4.48, p = 0.046). The interaction between time and THx was significant for both body weight (F(10, 220) = 4.11, p < 0.001) and food consumption (F(9, 198) = 5.57, p < 0.001).Study 5: Long-Lasting Effect of SuMN-THx on Body Weight and Glucose Metabolism in f-HFDs-RC RatsTo examine any potential long-lasting metabolic effect of TH knockdown at the SuMN area, f-HFDs-RC rats were inoculated at medial SuMN area with AAV5-Th-shRNA-GFP to knockdown Th gene expression (THx) or control AAV5-Scr-shRNA-GFP (Ctr) and maintained on regular chow for 1 year. Compared with Ctr, SuMN-THx increased % body weight gain by 96% in THx rats versus 65% increase in Ctr rats at 12 months (p = 0.012) post-virus inoculation (shown in Fig. 9a, b). There was a significant main effects of time (F(2, 32) = 196.47, p < 0.001) and THx (F(1, 16) = 7.79, p = 0.013), and the interaction between time and THx was also significant (F(2, 32) = 6.47, p = 0.004). At 12 months post-virus inoculation, compared with Ctr rats, THx rats exhibited elevated levels of fasting blood glucose (p = 0.015) (shown in Fig. 9c) and plasma insulin (p = 0.031) (shown in Fig. 9d) resulting in an increase in HOMA-IR (120%, p = 0.027) indicating IR (shown in Fig. 9e). The extent of TH knockdown at the SuMN area remained stable at 1 year post-virus inoculation as verified by TH immunohistochemistry (data not shown).Fig. 9.Long-lasting effect of THx at the SuMN area on body weight gain and glucose metabolism in f-HFDs-RC rats (Study 5). f-HFDs-RC rats were locally inoculated at the medial SuMN area with AAV5-Th-shRNA to knockdown Th gene expression (THx) or control AAV-Scr-shRNA-GFP (Ctr). THx treatment increased % body weight gain by 1.5-fold (96% vs. 65% increase in Ctr, p = 0.012) after 12 months post-virus inoculation (a, b). Compared with Ctr, THx treatment increased the plasma levels of fasting glucose (18%, p = 0.015) (c) and insulin (78%, p = 0.031) (d) resulting in an increase in HOMA-IR (120%, p = 0.027) (e) indicating IR. *p < 0.05, two-tailed Student’s t test.DiscussionThe present study is the first to demonstrate that TH knockdown of DA neurons at the SuMN area which includes DA neurons densely clustered within the medial SuMN and sparsely scattered DA neurons in the adjacent PH immediately dorsal to the medial SuMN is critically involved in the regulation of whole-body energy balance. Chronic (>50 days) TH knockdown of medial SuMN area dopaminergic neurons leads to development of the obese, glucose-intolerant condition in an animal model extremely resistant to development of such a condition – the lean young female SD rat selected to be refractory to the obesogenic effects of HFD feeding and held on low-fat regular chow diet and long (14 h light) daily photoperiods [30‒32]. Such treatment resulted in a substantial (42%) increase in body weight gain that was similar to the observed increase in body fat store level (42%). This body composition change occurred without any change in fasting or GTT insulin level; however, it was coupled to a concurrent 25% increase in feed efficiency and 15% increase in cumulative food consumption.The increased feed efficiency that coincided with the increased fat store accretion was associated with 1.4-fold increase in circadian fasting time liver mRNA expression of Ppargc1β, a transcription factor for stimulation of lipogenic enzyme transcription during the subsequent feeding period of the day, though liver mRNA levels of the other lipogenic factors Srebf1, Mtor, and Fasn that peak during feeding were unaffected at this fasting period of the day. Moreover, the increased feed efficiency was coupled to a substantial (∼55%) reduction in the capacity of the adipose tissue to oxidize lipid. That is, following treatment, lipid in adipose tissue was less metabolized to energy (ATP) and rather stored than released into the circulation as FFAs (no increase in plasma FFAs in SuMN TH knockdown vs. control rats). The increase in body fat and feed efficiency and decrease in adipose FFA oxidation rate observed with SuMN DA neuronal TH knockdown is consistent with, and importantly may explain in part, previous studies, wherein systemic or intracerebroventricular DA agonist treatment reduced obesity, increased protein turnover (a major sink for calorie consumption), and increased white adipose FFA oxidation rate in MS animals [37‒40]. Furthermore, blockade of the downstream effect of such endogenous CNS DA activity in lean, insulin-sensitive animals that normally exhibit low noradrenergic input activity at the VMH (thereby increasing noradrenergic activity at the VMH) actually not only induced obesity [41] but converted brown fat to white [42], a scenario consistent with low CNS DA action facilitating whole-body lipid storage. This finding in turn is consistent with published observations that hypothalamic bromocriptine (a potent DA D2 receptor agonist) administration activated brown fat thermogenesis [43]. The present findings suggest that SuMN dopaminergic neurons are operative (via connections with the SCN and other centers) in manifesting these multiple observations of CNS DA regulation of body composition and white adipose lipid metabolism. Furthermore, the present findings are consistent with previous studies wherein (a) SuMN circadian efferent DA release contributes to the circadian peak of dopaminergic input to the SCN in healthy animals, an SCN stimulus critical to the maintenance of the lean, insulin-sensitive state, and (b) daily intracranial administration to MS animals of DA directly to the SCN at the time of day of its natural circadian peak reduced MS [10].Concurrent with the SuMN DA neuronal TH knockdown effect to increase body weight and fat in this animal model, such treatment also induced glucose intolerance that was coupled to increases in hepatic mRNA expression of the two key rate-limiting gluconeogenic enzymes, glucose 6 phosphatase and phosphoenolpyruvate carboxy kinase (thus contributing to glucose intolerance) without change in the plasma insulin level suggesting hepatic IR. Inasmuch as the SuMN DA neurons project directly and indirectly to the SCN, this SuMN-THx glucose metabolism response is consistent with previous observations that the SCN is a critical modulator of both glucose tolerance [10, 11, 16, 41] and hepatic gluconeogenesis [44]. The present findings offer insights into dopaminergic input pathways controlling/modifying this SCN output glucose regulatory system. Additionally, chronic SuMN-THx treatment resulted in the increased mRNA expression for two major hepatic pro-inflammatory proteins, Rela, the p65 protein involved in NF-κB activation, a pro-inflammatory transcription factor in liver, and Ccl2/Mcp1, a potent pro-inflammatory protein within liver and the immune system. This is relevant since hepatic inflammation is well known to potentiate hepatic gluconeogenesis, lipogenesis, and IR and can eventually lead to fatty liver [45, 46]. In this context, SuMN dopaminergic neuronal TH knockdown did also increase hepatic Ppargc1β, a lipogenic transcription factor, as well as plasma triglyceride levels (by 86%), coupled with a trend toward increased liver lipid content (without alteration of key liver lipid oxidation transcription factor mRNA levels). This intervention-induced obese-increased hepatic gluconeogenic/lipogenic phenotype is also consistent with leptin resistance and the observed hyperleptinemia (46% increase in plasma leptin) (see Fig. 5a) in these treated animals, especially in the face of increased food consumption over the test period (see Fig. 2c, d). Finally, since SuMN DA neurons project strongly to the SCN area and DA depletion at the SCN area has been reported to induce MS, a preliminary inspection of circadian components regulating metabolism was undertaken and it was observed that SuMN-THx treatment disrupted the diurnal (sleep) phase plasma corticosterone level (68% decrease) and liver mRNA levels for circadian genes, Clock and Per1 (1.4- and 1.8-fold increase, respectively). Although full 24-h profiles for plasma corticosterone level and liver circadian clock genes were not obtained in this study, these initial findings suggest that a shift in the circadian organization of metabolism had occurred potentiating the obese, glucose-intolerant condition that is consistent with previous findings wherein a shift in the plasma corticosteroid rhythm and alterations in liver circadian clock gene expressions potentiated MS [32, 47, 48]. Moreover, such SuMN-THx treatment also altered the circadian feeding rhythm resulting in increased feeding exclusively in the fasting period of the day – a phenomenon repeatedly demonstrated to induce MS when experimentally imposed (see below). In this regard, it has been concluded from a compilation of a multitude of studies that CNS circadian regulation of peripheral circadian metabolic activities manifests the lean, glucose-tolerant or obese, glucose-intolerant condition as a function of the temporal interactions of the phases of multiple circadian CNS output signals (e.g., endocrine and autonomic) that act to organize circadian biology in peripheral tissues (e.g., liver) in a manner to either facilitate or block the obese, glucose-intolerant condition [32, 47, 49]. The present findings along with these previous studies suggest that the SuMN dopaminergic neurons influence this CNS clock circuit regulation of metabolism.In conjunction with the above-described metabolic changes, an increase in daily food consumption was observed following the SuMN-THx treatment. The cumulative increase in food consumption sampled during the test periods (over 35, 72, and 140 days in Study 3, 2A, and 4, respectively) can be expected to work in concert with the increased feed efficiency and lipid accretion biochemical pathways to increase body weight and fat. It appears that decreased SuMN dopaminergic efferent innervation coordinates feeding-independent and feeding-dependent physiological functions that interact to shift metabolism toward the obese, glucose-intolerant state. Low brain DA in the mesolimbic reward system (decreases in DA D2 receptors and/or DA release) is observed in obesity and MS [50‒53] resulting in food overconsumption presumably as a result of the weak (insufficient) DA reward signal to induce meal feeding cessation [54]. While the influence of chronic SuMN-THx to induce increased feeding was consistently observed in the present studies, the influence of such intervention upon striatal DA release and reward-seeking behavior was not assessed. Recent studies by Kesner et al. [55] provide evidence that acute activation of SuMN glutamate, but not SuMN dopaminergic neurons results in, via communication with the septum and septum to VTA circuit, a VTA neuronal DA release at the ventral striatum to influence/stimulate motivation for environmental interaction and reward-seeking behavior. Activation of AMPA glutamate receptors on SuMN neurons, which may be of a variety of types, stimulates DA release and reward effects at the ventral striatum [56]. Importantly, we have previously demonstrated that AMPA stimulation of SuMN neurons causes substantial DA release from such SuMN neurons at the hypothalamic SCN and MPOA [11], and SuMN DA neurons are known to project to the LS and MPOA [2, 12] – all key relay centers to the VTA and as such may influence VTA-striatal DA release [57, 58]. However, whether chronic knockdown of SuMN dopaminergic neurons may result via reprogramming of neuronal communication with the LS, MPOA, and/or SCN to influence the VTA DA release to the ventral striatum to modulate reward-seeking behavior remains to be investigated. Furthermore, this TH knockdown of SuMN DA neurons may act via their direct projections to the SCN, MPOA, and/or LS that in turn communicate with other hypothalamic feeding regulating centers such as the PVN, DMH, LH, and perifornical area [59] to increase feeding as observed herein. The LS that receives abundant dopaminergic projections from the SuMN is a key relay station for hippocampal function [4]. The TH knockdown at the SuMN area may reduce hippocampal function in feeding conditioning and learning processes [60, 61] to thereby increase food consumption. Moreover, as it regards SuMN regulation of feeding, SuMN neurons express both GLP-1 and ghrelin receptors, and thus this nucleus is capable of directly responding to hunger (ghrelin) and satiety (GLP-1) signaling regulation (i.e., ghrelin acutely induces feeding and GLP-1 inhibits feeding) [7, 8]. However, the present study findings suggest that SuMN-THx results in a “ghrelin-resistant” state as it relates to feeding, a circumstance that has previously been associated with the obese, insulin-resistant state [62], without affecting the hypophagia response to GLP-1 agonist. In total, the effect of SuMN-THx on feeding may be summarized as follows. Such treatment induces a general sustained chronic increase in feeding over several weeks whether of RC or HFD, but such increased feeding restricted to the normal fasting period of the day, and such increased basal feeding level linked to a subsequent diminished ghrelin stimulatory feeding response. The effect of SuMN-THx to induce hyperphagia specifically within the natural resting/fasting (light) period of the day in these nocturnal animals is intriguing and worthy of brief separate discussion.Daily feeding time is itself a zeitgeber for entraining peripheral fuel metabolism rhythms and for regulating metabolic status, independent of the SCN output entrainment of such functions [63‒69]. Nonetheless, circadian rhythms of peripheral fuel metabolism, particularly in the liver, are influenced by both timed feeding stimulus and neuroendocrine input signals that derive from the SCN network [70, 71]. Misalignment of the daily feeding and SCN-driven activity rhythms (feeding outside the normal activity cycle) is well established to induce MS in animals [72‒75] and humans [76‒81]; however, what causes an internal drive to eat outside the normal feeding cycle remains a mystery. The present study findings in conjunction with previous study results demonstrating SuMN dopaminergic circadian input activity to the SCN [11] suggest that SuMN dopaminergic activity somehow participates in the regulation and coordination of both SCN circadian organization and output activity as well as CNS centers that regulate the daily feeding rhythm to modulate peripheral fuel metabolism. In support of this postulate is the observation that SuMN-THx simultaneously disrupted normal daily feeding activity (inducing fasting period hyperphagia), plasma corticosterone level, and hepatic clock gene expression while inducing MS in animals extremely resistant to MS. The theoretical dimensions of these findings are worthy of serious further investigation. It should be noted that whole-body metabolic rate was not determined in these studies and as such the possibility remains that SuMN-THx may have reduced overall energy expenditure to facilitate weight gain, in addition to such treatment effects on feeding. Nonetheless, SuMN-THx increased % body fat to a greater extent than % weight gain, increased hepatic lipogenic machinery, and decreased adipose fatty acid metabolism, suggesting a neuroendocrine shift driving fat accrual and storage is a primary modulatory force for the observed weight gain.In addition to the above-described SuMN DA neuron TH knockdown studies in the female SD rat model, the impact of such treatment was also assessed in the male SD rat. The male SD rat, in stark contrast to the MS-resistant female, develops IR at a very early age which worsens rapidly with time concurrent with marked weight gain over the first 6 months of life [30, 31]. Nonetheless, such SuMN-THx treatment in such male SD rats fed on regular chow diet resulted in further impaired glucose tolerance that was also accompanied by worsening IR (highlighting a more exaggerated glucose intolerance response than in the females). This response to SuMN-THx treatment was associated with weight gain (12% over 140 days), albeit less than in the female (42% over 72 days), likely due to the existent underlying obesity at baseline over the test period in males (compare Fig. 2a, b vs. Fig. 8f, g). The composite of these female and male studies suggests that the SuMN-THx response is not gender specific and may be expressed across the continuum of metabolic control status from the lean, insulin-sensitive to obese, insulin-resistant states.To further explore the breadth of baseline fuel metabolism profile upon which the SuMN-THx may function to induce glucose intolerance, another study was conducted wherein the SuMN-THx intervention was imposed on young female SD rats resistant to the obesogenic effects of HFD feeding but maintained on HFD (instead of regular chow as in the initial female SD rat study). In this HFD feeding scenario, the MS-inducing effect of the SuMN-THx treatment was more pronounced relative to such treatment in the same animal model held on regular rodent chow. Under conditions of HFD feeding, SuMN-THx treatment induced glucose intolerance, like the situation under conditions of regular chow feeding; however, such treatment under HFD feeding conditions also further induced an increase in fasting plasma insulin levels, GTT hyperinsulinemia, and frank IR (see Fig. 7a–e vs. Fig. 3 for comparison). Moreover, while food consumption and feed efficiency increase and a decrease in white adipose FFA oxidation rate were evident in SuMN-THx treatment of HFD and regular chow fed animals, body weight gain was much more rapid (time to significant increase at 30 days vs. 50 days) and to a greater extent (23% increase at 35 days vs. no change at 35 days) in SuMN-THx animals held on HFD versus regular chow, respectively. It is well established from multiple studies that chronic HFD feeding inhibits DA function in several brain centers, including the SuMN [10], by reducing DA synthesis and/or DA receptor binding [11, 50, 51, 82‒84]. This HFD effect may be one of several HFD factors contributing to the exaggerated MS response to SuMN DA neuronal TH knockdown in the present study by producing a more global depressed CNS dopaminergic activity background upon which the SuMN DA neuronal TH knockdown occurs. The present study findings in concert with these previous studies of HFD influence on brain dopaminergic activity suggest however that the SuMN itself is an important site for HFD-induced reduction in brain dopaminergic activity that leads to MS. Consequently, in composite these studies highlight a pathophysiological scenario where chronic HFD both triggers and amplifies the MS-inducing effect of decreased SuMN dopaminergic neuronal activity.Finally, a long-term study (1 year) of SuMN-THx metabolic response in young lean female HFD-sensitive SD rats but held on regular chow indicated that such SuMN treatment resulted in a significant increase in % body weight gain (47%), fasting blood glucose (18%), and plasma insulin (78%), and HOMA-IR (120%) levels versus controls. These findings suggest that the MS influences of SuMN-THx treatment are not able to be countered by an endogenous metabolic setpoint control to correct the metabolic disruption even after 1 year (nearly half the lifetime of the SD rat).The SuMN DA neurons project directly and indirectly to multiple areas of the limbic reward system (e.g., hippocampus, VTA, LS) and to hypothalamic metabolic relevant centers (e.g., MPOA, DMH, SCN) [2, 11, 12, 14]. The SuMN DA neurons that project to the clock SCN region are of particular importance with respect to peripheral energy metabolism. Seasonal changes in metabolic status (i.e., summer lean, insulin-sensitive vs. winter obese, insulin-resistant conditions) among animals in the wild are correlated with and driven by the seasonal wax and wane, respectively, in circadian peak DA activity at the SCN region [11, 32]. Reduction of circadian DA activity at the clock SCN region of lean insulin-sensitive animals by DA neuron-specific lesion or HFD feeding induces obesity and IR/glucose intolerance [16, 41]. In contrast, restoration of the diminished circadian DA activities in obese insulin-resistant animals by direct timed daily micro-infusion of DA directly at the SCN region or timed daily administration of DA receptor agonist at the time of day of the circadian peak in SCN dopaminergic input activity improves glucose intolerance and IR [10, 11, 17]. Since the circadian activity peak of TH-positive DA neurons within SuMN provides a significant circadian peak of dopaminergic input to the SCN region, SuMN-THx should reduce the circadian peak DA activity at this region. Moreover, other metabolism control centers are known to receive dopaminergic projections from the SuMN (e.g., the LS, MPOA, DMH) [2, 12, 14], and all communicate strongly with the SCN. Therefore, the SuMN DA neurons could via providing circadian dopaminergic input directly to the SCN region and/or via relay centers to the SCN modulate its function to regulate peripheral energy metabolism. Moreover, the SuMN communicates with the nucleus tractus solitarius, a first-order recipient of primary afferents from the gut [85], wherein dietary constituents from the gut are known to regulate brain DA synthesis [84]. Specifically, gastrointestinal saturated fat has been demonstrated to inhibit brain DA synthesis via parasympathetic connections [84, 86, 87]. Although speculative, available evidence clearly suggests the possibility that gastrointestinal fat from HFD feeding may reduce SuMN DA synthesis that in turn leads to MS potentiation in part via reducing DA input to the hypothalamic SCN and limbic system. Consequently, the SuMN is a unique neuronal integration and coordination center for both behavioral and biochemical regulation of peripheral fuel metabolism.The limitations of these composite studies include the lack assessment of SuMN-THx on whole-body metabolic rate (that may have influenced SuMN-THx effect on body weight) and on post-mealtime lipid synthesis rate. However, now having these study data in hand, future studies of whole-body metabolic rate and post-meal lipogenic activity are warranted.In summary, chronic TH knockdown of SuMN area DA neurons via site-specific inoculation of a virus-mediated shRNA to knockdown TH (the rate-limiting enzyme for DA synthesis) gene expression in animals selected to be extremely resistant to development of MS at baseline and maintained on regular rodent chow induced multiple pathophysiological and phenotypic aspects of MS. Features of this SuMN TH knockdown-induced metabolic shift include increased body weight gain, body fat mass, food consumption, feed efficiency, and glucose intolerance concurrent with increased plasma leptin and triglyceride levels, increased hepatic mRNA expression levels of key gluconeogenic, lipogenic, and pro-inflammatory genes, and a reduction in white adipose FAO rate. Such treatment also simultaneously altered the circadian phase of daily feeding with increased fasting period feeding, plasma corticosterone level, and the expression of circadian genes in the liver that are involved in the regulation of hepatic fuel metabolism, strongly suggestive of an impact on central pacemaker-biological clock regulation of metabolism. This MS-inducing effect of SuMN DA neuronal TH knockdown was further enhanced under conditions of HFD feeding in this animal model system. The dopaminergic neurons at the SuMN area are thus a major integral functional component of a complex CNS system regulating peripheral fuel metabolism.Statement of EthicsThis study protocol was reviewed and approved by the Institutional Animal Care and Use Committee of VeroScience, LLC., approval number 2016-E0197.Conflict of Interest StatementThe authors are current or former employees of VeroScience LLC, a company that owns Cycloset®, a DA D2 receptor agonist, i.e., FDA-approved for the treatment of type 2 diabetes. The authors declare that they have no competing interests.Funding SourcesThis study was funded by VeroScience LLC of Tiverton, RI and S2 Therapeutics of Bristol, TN.Author ContributionsA.H.C. and Y.Z. designed the present studies and wrote the manuscript. Y.Z., M.E., and T.-H.T. performed the experiments. Y.Z. and C.S. analyzed the data. http://www.deepdyve.com/assets/images/DeepDyve-Logo-lg.png Neuroendocrinology Karger http://www.deepdyve.com/lp/karger/tyrosine-hydroxylase-knockdown-at-the-hypothalamic-supramammillary-p1uv0CSbUV

Loading next page...

References (96)

U. Schibler, J. Ripperger, Steven Brown (2003)
Peripheral Circadian Oscillators in Mammals: Time and Food
Journal of Biological Rhythms, 18
E. Maury, K. Ramsey, J. Bass (2010)
Circadian Rhythms and Metabolic Syndrome: From Experimental Genetics to Human Disease
Circulation Research, 106
S. Ikemoto (2010)
Brain reward circuitry beyond the mesolimbic dopamine system: A neurobiological theory
Neuroscience & Biobehavioral Reviews, 35
Direct midbrain dopamine input to the suprachiasmatic nucleus accelerates 508
M. Ezrokhi, Yahong Zhang, S. Luo, A. Cincotta (2021)
Time-of-Day-Dependent Effects of Bromocriptine to Ameliorate Vascular Pathology and Metabolic Syndrome in SHR Rats Held on High Fat Diet
International Journal of Molecular Sciences, 22
M. Tsanov (2018)
Differential and complementary roles of medial and lateral septum in the orchestration of limbic oscillations and signal integration
European Journal of Neuroscience, 48
A. Cincotta, S. Luo, Y. Zhang, Y. Liang, K. Bina, T. Jetton, P. Scislowski (2000)
Chronic infusion of norepinephrine into the VMH of normal rats induces the obese glucose-intolerant state.
American journal of physiology. Regulatory, integrative and comparative physiology, 278 2
Elizabeth Sutton, R. Beyl, K. Early, W. Cefalu, E. Ravussin, C. Peterson (2018)
Early Time-Restricted Feeding Improves Insulin Sensitivity, Blood Pressure, and Oxidative Stress Even without Weight Loss in Men with Prediabetes.
Cell metabolism, 27 6
S. Luo, A. Meier, A. Cincotta (1998)
Bromocriptine Reduces Obesity, Glucose Intolerance and Extracellular Monoamine Metabolite Levels in the Ventromedial Hypothalamus of Syrian Hamsters
Neuroendocrinology, 68
F. Marra, F. Tacke (2014)
Roles for chemokines in liver disease.
Gastroenterology, 147 3
Yahong Zhang, Carl Stoelzel, M. Ezrokhi, Tsung‐Huang Tsai, A. Cincotta (2021)
Activation State of the Supramammillary Nucleus Regulates Body Composition and Peripheral Fuel Metabolism
Neuroscience, 466
(2017)
Direct midbrain dopamine input to the suprachiasmatic nucleus accelerates circadian entrainment
Curr Biol, 27
A. Engin (2017)
Circadian Rhythms in Diet-Induced Obesity.
Advances in experimental medicine and biology, 960
Mei-Chuan Chou, Hsiang-Chun Lee, Yen-Chin Liu, Patrick Yen, Ching-Kuan Liu, Chu‐Huang Chen, Tzu-Han Hsieh, Shiou-Lan Chen (2022)
Long-Term High-Fat Diet Consumption Depletes Glial Cells and Tyrosine Hydroxylase–Containing Neurons in the Brain of Middle-Aged Rats
Cells, 11
M. Goodman, S. Dluz, M. McElaney, E. Belur, N. Ruderman (1983)
Glucose uptake and insulin sensitivity in rat muscle: changes during 3-96 weeks of age.
The American journal of physiology, 244 1
Lauren Pickel, H. Sung (2020)
Feeding Rhythms and the Circadian Regulation of Metabolism
Frontiers in Nutrition, 7
A. Kalsbeek, A. Kalsbeek, E. Bruinstroop, C. Yi, C. Yi, L. Klieverik, S. Fleur, E. Fliers (2010)
Hypothalamic control of energy metabolism via the autonomic nervous system
Annals of the New York Academy of Sciences, 1212
(2013)
Sympathovagal imbalance contributes to prehypertension status and cardiovascular risks attributed by insulin resistance, inflammation, dyslipidemia and oxidative stress in first degree relatives of type 2 diabetics
PLoS One, 8
P. Molinoff, J. Axelrod (1971)
Biochemistry of catecholamines.
Annual review of biochemistry, 40
F. Damiola, Nguyet Minh, Nicolas Preitner, Benoît Kornmann, F. Fleury-Olela, U. Schibler (2000)
Restricted feeding uncouples circadian oscillators in peripheral tissues from the central pacemaker in the suprachiasmatic nucleus.
Genes & development, 14 23
Bulbul Ahmed, R. Sultana, M. Greene (2021)
Adipose tissue and insulin resistance in obese.
Biomedicine & pharmacotherapy = Biomedecine & pharmacotherapie, 137
I. Karatsoreos, S. Bhagat, N. Bowles, Z. Weil, D. Pfaff, B. McEwen (2010)
Endocrine and physiological changes in response to chronic corticosterone: a potential model of the metabolic syndrome in mouse.
Endocrinology, 151 5
E. Burgos-Morón, Z. Abad-Jiménez, Aranzazu Marañón, Francesca Iannantuoni, Irene Escribano-López, S. López-Doménech, Christian Salom, A. Jover, Vicente Mora, I. Roldan, E. Solá, M. Rocha, V. Víctor (2019)
Relationship between Oxidative Stress, ER Stress, and Inflammation in Type 2 Diabetes: The Battle Continues
Journal of Clinical Medicine, 8
P. Scislowski, E. Tozzo, Y. Zhang, S. Phaneuf, R. Prevelige, A. Cincotta (1999)
Biochemical mechanisms responsible for the attenuation of diabetic and obese conditions in ob/ob mice treated with dopaminergic agonists
International Journal of Obesity, 23
(1993)
Circadian neuro-endocrine role in age-related changes in body fat stores and insulin sensitivity of the male Sprague-Dawley rat
S. Kanoski, H. Grill (2017)
Hippocampus Contributions to Food Intake Control: Mnemonic, Neuroanatomical, and Endocrine Mechanisms
Biological Psychiatry, 81
A. Meier, A. Cincotta, W. Lovell (1992)
Timed bromocriptine administration reduces body fat stores in obese subjects and hyperglycemia in type II diabetics
Experientia, 48
C. Bäckman, Yajun Zhang, B. Hoffer, A. Tomac (2003)
Short interfering RNAs (siRNAs) for reducing dopaminergic phenotypic markers
Journal of Neuroscience Methods, 131
Xu-Feng Huang, K. Zavitsanou, Xin Huang, Yinghua Yu, Hongqin Wang, Feng Chen, A. Lawrence, Chao Deng (2006)
Dopamine transporter and D2 receptor binding densities in mice prone or resistant to chronic high fat diet-induced obesity
Behavioural Brain Research, 175
M. Bray, William Ratcliffe, Maximiliano Grenett, Rachel Brewer, K. Gamble, M. Young (2012)
Quantitative Analysis of Light-Phase Restricted Feeding Reveals Metabolic Dyssynchrony in Mice
International journal of obesity (2005), 37
P. Tsiotra, C. Tsigos (2006)
Stress, the Endoplasmic Reticulum, and Insulin Resistance
Annals of the New York Academy of Sciences, 1083
E. Martínez-Martínez, Francisco Souza-Neto, Sara Jiménez-González, V. Cachofeiro (2021)
Oxidative Stress and Vascular Damage in the Context of Obesity: The Hidden Guest
Antioxidants, 10
A. Cincotta (2013)
Hypothalamic Role in the Insulin Resistance Syndrome
A. Cincotta (2023)
Brain Dopamine–Clock Interactions Regulate Cardiometabolic Physiology: Mechanisms of the Observed Cardioprotective Effects of Circadian-Timed Bromocriptine-QR Therapy in Type 2 Diabetes Subjects
International Journal of Molecular Sciences, 24
(2024)
Dampened 85 510
S. Luo, Jing Luo, A. Meier, A. Cincotta (1997)
Dopaminergic neurotoxin administration to the area of the suprachiasmatic nuclei induces insulin resistance
NeuroReport, 8
Fabrice Plaisier, C. Hume, J. Menzies (2020)
Neural connectivity between the hypothalamic supramammillary nucleus and appetite‐ and motivation‐related regions of the rat brain
Journal of Neuroendocrinology, 32
J. Dourmashkin, G. Chang, James Hill, E. Gayles, Susan Fried, S. Leibowitz (2006)
Model for predicting and phenotyping at normal weight the long-term propensity for obesity in Sprague–Dawley rats
Physiology & Behavior, 87
Guillaume Lartigue (2016)
Role of the vagus nerve in the development and treatment of diet‐induced obesity
The Journal of Physiology, 594
(2017)
Supra-mammillary glutamate neurons are a key node of the arousal system
M. Ezrokhi, S. Luo, Yelena Trubitsyna, A. Cincotta (2014)
Neuroendocrine and metabolic components of dopamine agonist amelioration of metabolic syndrome in SHR rats
Diabetology & Metabolic Syndrome, 6
(2017)
accelerates circadian entrainment
S. Miller, J. Lonstein (2009)
Dopaminergic projections to the medial preoptic area of postpartum rats
Neuroscience, 159
S. Luo, Y. Zhang, M. Ezrokhi, Y. Li, Tsung‐Huang Tsai, A. Cincotta (2018)
Circadian peak dopaminergic activity response at the biological clock pacemaker (suprachiasmatic nucleus) area mediates the metabolic responsiveness to a high‐fat diet
Journal of Neuroendocrinology, 30
(2001)
Restricted feeding entrains liver clock without participation of the suprachiasmatic nucleus
Genes Cells, 6
C. Folgueira, Daniel Beiroa, Begoña Porteiro, Manon Duquenne, E. Puighermanal, Marcos Fondevila, S. Barja-Fernández, R. Gallego, R. Hernández-Bautista, Cecilia Castelao, A. Senra, Patricia Seoane-Collazo, Noemí Gómez-Lado, P. Aguiar, D. Guallar, Miguel Fidalgo, Amparo Romero-Picó, R. Adan, C. Blouet, J. Labandeira-García, F. Jeanrenaud, I. Kalló, Zsolt Liposits, J. Salvador, V. Prévot, C. Diéguez, Miguel López, E. Valjent, G. Frühbeck, L. Seoane, R. Nogueiras (2019)
Hypothalamic dopamine signalling regulates brown fat thermogenesis
Nature Metabolism, 1
P. Risold, L. Swanson (1996)
Structural Evidence for Functional Domains in the Rat Hippocampus
Science, 272
K. Stokkan, S. Yamazaki, H. Tei, Y. Sakaki, M. Menaker (2001)
Entrainment of the circadian clock in the liver by feeding.
Science, 291 5503
Andrew Rowan (1996)
Guide for the Care and Use of Laboratory Animals
Journal of Medical Primatology, 8
S. Fulton, P. Pissios, Ramon Manchon, L. Stiles, L. Frank, E. Pothos, E. Maratos-Flier, J. Flier (2006)
Leptin Regulation of the Mesoaccumbens Dopamine Pathway
Neuron, 51
G. Hotamışlıgil (2010)
Endoplasmic Reticulum Stress and the Inflammatory Basis of Metabolic Disease
Cell, 140
(2002)
Hypothalamic role in the insulin resistance syndrome, in insulin Resistance and insulin resistance syndrome
Frontiers in animal diabetes Research series
Andrew Kesner, Rick Shin, Coleman Calva, Reuben Don, Sue Junn, Christian Potter, Leslie Ramsey, Ahmed Abou-Elnaga, Christopher Cover, Dong Wang, Hanbing Lu, Yihong Yang, S. Ikemoto (2021)
Supramammillary neurons projecting to the septum regulate dopamine and motivation for environmental interaction in mice
Nature Communications, 12
S. Luo, M. Ezrokhi, Nicholas Cominos, Tsung‐Huang Tsai, Carl Stoelzel, Yelena Trubitsyna, A. Cincotta (2021)
Experimental dopaminergic neuron lesion at the area of the biological clock pacemaker, suprachiasmatic nuclei (SCN) induces metabolic syndrome in rats
Diabetology & Metabolic Syndrome, 13
D. Jakubowicz, M. Barnea, J. Wainstein, O. Froy (2013)
High Caloric intake at breakfast vs. dinner differentially influences weight loss of overweight and obese women
Obesity, 21
Changjiu Zhao, L. Riters (2023)
The medial preoptic area and its projections to the ventral tegmental area and the periaqueductal gray are activated in response to social play behavior in juvenile rats.
Behavioral neuroscience
Lorena López-Ferreras, K. Eerola, D. Mishra, Olesya Shevchouk, Jennifer Richard, Fredrik Nilsson, M. Hayes, K. Skibicka (2018)
GLP-1 modulates the supramammillary nucleus-lateral hypothalamic neurocircuit to control ingestive and motivated behavior in a sex divergent manner
Molecular Metabolism, 20
S. Paredes, L. Ribeiro (2014)
Cortisol: the villain in metabolic syndrome?
Revista da Associacao Medica Brasileira, 60 1
M. Garaulet, P. Gómez‐Abellán, J. Alburquerque-Bejar, Yu-Chi Lee, J. Ordovás, F. Scheer (2013)
Timing of food intake predicts weight loss effectiveness
International Journal of Obesity, 37
W. Pan, N. McNaughton (2004)
The supramammillary area: its organization, functions and relationship to the hippocampus
Progress in Neurobiology, 74
Sam Chang, B. Graham, F. Yakubu, D. Lin, J. Peters, J. Hill (1990)
Metabolic differences between obesity-prone and obesity-resistant rats.
The American journal of physiology, 259 6 Pt 2
A. Meier, A. Cincotta (1996)
Circadian rhythms regulate the- expression of the thrifty genotypelphenotype
H. Cui, Miguel López, K. Rahmouni (2017)
The cellular and molecular bases of leptin and ghrelin resistance in obesity
Nature Reviews Endocrinology, 13
A. Gallant, J. Lundgren, V. Drapeau (2012)
The night‐eating syndrome and obesity
Obesity Reviews, 13
Jonathan Hommel, Robert Sears, D. Georgescu, Diana Simmons, R. DiLeone (2003)
Local gene knockdown in the brain using viral-mediated RNA interference
Nature Medicine, 9
Marie May, C. Hume, N. Sabatier, E. Schéle, T. Bake, Ulrika Bergström, J. Menzies, Suzanne Dickson (2019)
Activation of the rat hypothalamic supramammillary nucleus by food anticipation, food restriction or ghrelin administration
Journal of Neuroendocrinology, 31
Stefanos Stagkourakis, G. Spigolon, P. Williams, J. Protzmann, G. Fisone, C. Broberger (2018)
A neural network for intermale aggression to establish social hierarchy
Nature Neuroscience, 21
T. Davidson, S. Hargrave, S. Swithers, C. Sample, Xue Fu, K. Kinzig, Wei Zheng (2013)
Inter-relationships among diet, obesity and hippocampal-dependent cognitive function
Neuroscience, 253
Cécile Hryhorczuk, Marcu Florea, D. Rodaros, Isabelle Poirier, C. Daneault, C. Rosiers, A. Arvanitogiannis, T. Alquier, S. Fulton (2016)
Dampened Mesolimbic Dopamine Function and Signaling by Saturated but not Monounsaturated Dietary Lipids
Neuropsychopharmacology, 41
O. Gorbatyuk, Shoudong Li, K. Nash, M. Gorbatyuk, A. Lewin, Layla Sullivan, R. Mandel, Weijun Chen, C. Meyers, F. Manfredsson, N. Muzyczka (2010)
In Vivo RNAi-Mediated α-Synuclein Silencing Induces Nigrostriatal Degeneration.
Molecular therapy : the journal of the American Society of Gene Therapy, 18 8
Gene-Jack Wang, N. Volkow, J. Logan, N. Pappas, Christopher Wong, Wel Zhu, Noelwah Netusll, J. Fowler (2001)
Brain dopamine and obesity
The Lancet, 357
(1994)
basis for thalamic and visceral re ﬂ ex integration
(2021)
Dopamine D2 receptor agonist, bromocriptine, remodels adipose tissue dopaminergic signalling and upregulates catabolic pathways, improving metabolic pro ﬁ le in type 2 diabetes
P. Crosby, R. Hamnett, M. Putker, Nathaniel Hoyle, Martin Reed, Carolyn Karam, E. Maywood, Alessandra Stangherlin, J. Chesham, Edward Hayter, L. Rosenbrier-Ribeiro, P. Newham, H. Clevers, D. Bechtold, J. O’Neill (2019)
Insulin/IGF-1 Drives PERIOD Synthesis to Entrain Circadian Rhythms with Feeding Time
Cell, 177
E. Challet (2019)
The circadian regulation of food intake
Nature Reviews Endocrinology, 15
B. Geiger, M. Haburcak, N. Avena, N. Avena, M. Moyer, B. Hoebel, E. Pothos (2009)
Deficits of mesolimbic dopamine neurotransmission in rat dietary obesity
Neuroscience, 159
S. Bo, G. Musso, G. Beccuti, M. Fadda, Debora Fedele, R. Gambino, L. Gentile, M. Durazzo, E. Ghigo, M. Cassader (2014)
Consuming More of Daily Caloric Intake at Dinner Predisposes to Obesity. A 6-Year Population-Based Prospective Cohort Study
PLoS ONE, 9
Michelle Wang, H. Yin, R. Landick, J. Gelles, L. Churchman, Z. Okten, R. Rock, J. Dawson, J. Spudich, C. Baumann, S. Smith, V. Bloomfield, C. Bustamante, A. Dunn, D. Halpin, D. Herschlag, W. Martin, B. Robinson, J. Schurr, A. Smith, R. Weisinger, S. Wrenn, Y. Bai, S. Doniach, K. Ito, J. Lipfert, I. Millett, S. Seifert, P. Thiyagarajan, H. Tsuruta (2008)
Relation Between Obesity and Blunted Striatal Response to Food Is Moderated by TaqIA A1 Allele
P. Zimmet, K. Alberti, N. Stern, C. Bilu, A. El-Osta, A. El-Osta, A. El-Osta, H. Einat, N. Kronfeld-Schor (2019)
The Circadian Syndrome: is the Metabolic Syndrome and much more!
Journal of Internal Medicine, 286
G. Lartigue, C. Serre, H. Raybould (2011)
Vagal afferent neurons in high fat diet-induced obesity; intestinal microflora, gut inflammation and cholecystokinin
Physiology & Behavior, 105
E. Ravussin, R. Beyl, E. Poggiogalle, D. Hsia, C. Peterson (2019)
Early Time-Restricted Feeding Reduces Appetite and Increases Fat Oxidation but Does Not Affect Energy Expenditure in Humans
Obesity (Silver Spring, Md.), 27
N. Minh, F. Damiola, F. Tronche, G. Schütz, U. Schibler (2001)
Glucocorticoid hormones inhibit food‐induced phase‐shifting of peripheral circadian oscillators
The EMBO Journal, 20
Restricted feeding entrains liver clock without participation Neuroendocrinology
I. Vega-Quiroga, H. Yarur, K. Gysling (2018)
Lateral septum stimulation disinhibits dopaminergic neurons in the antero-ventral region of the ventral tegmental area: Role of GABA-A alpha 1 receptors
Neuropharmacology, 128
(2011)

Guide for the care and use of laboratory animals
(2001)
participation of the suprachiasmatic nucleus
R. Chamorro, Céline Jouffe, H. Oster, N. Uhlenhaut, Sebastian Meyhöfer (2023)
When should I eat: A circadian view on food intake and metabolic regulation
Acta Physiologica, 237
D. Ratna, A. Mondal, B. Mallick (2023)
Modulation of dopamine from ventral tegmental area neurons by the LC-REM-OFF and PPT-REM-ON neurons in REMS regulation in freely moving rats
Neuropharmacology, 237
H. Vogel, S. Wolf, Cristina Rabasa, F. Rodríguez‐Pacheco, Carina Babaei, Franziska Stöber, J. Goldschmidt, R. DiMarchi, Brian Finan, M. Tschöp, S. Dickson, A. Schürmann, K. Skibicka (2016)
GLP-1 and estrogen conjugate acts in the supramammillary nucleus to reduce food-reward and body weight
Neuropharmacology, 110
Megumi Hatori, Christopher Vollmers, Amir Zarrinpar, Luciano DiTacchio, E. Bushong, S. Gill, Mathias Leblanc, A. Chaix, Matthew Joens, J. Fitzpatrick, Mark Ellisman, Satchidananda Panda (2012)
Time-restricted feeding without reducing caloric intake prevents metabolic diseases in mice fed a high-fat diet.
Cell metabolism, 15 6
P. Raskin, A. Cincotta (2016)
Bromocriptine-QR therapy for the management of type 2 diabetes mellitus: developmental basis and therapeutic profile summary
Expert Review of Endocrinology & Metabolism, 11
Luis Tellez, Sara Medina, Wenfei Han, J. Ferreira, P. Licona-Limón, Xueying Ren, Tukiet Lam, G. Schwartz, I. Araújo (2013)
A Gut Lipid Messenger Links Excess Dietary Fat to Dopamine Deficiency
Science, 341
S. Ikemoto, Brian Witkin, A. Zangen, R. Wise (2004)
Rewarding Effects of AMPA Administration into the Supramammillary or Posterior Hypothalamic Nuclei But Not the Ventral Tegmental Area
The Journal of Neuroscience, 24
(2010)
In vivo RNAi-mediated alpha-synuclein silencing induces nigrostriatal degeneration
Mol Ther, 18
K. Otake, Donald Reis, David Ruggierolv (1994)
Afferents to the midline thalamus issue collaterals to the nucleus tractus solitarii: an anatomical basis for thalamic and visceral reflex integration
, 14
Ronghua Hu, C. Xia, E. Butfiloski, M. Clare-Salzler (2018)
Effect of high glucose on cytokine production by human peripheral blood immune cells and type I interferon signaling in monocytes: Implications for the role of hyperglycemia in the diabetes inflammatory process and host defense against infection.
Clinical immunology, 195

Publisher: Karger
Copyright: © 2023 The Author(s). Published by S. Karger AG, Basel
ISSN: 0028-3835
eISSN: 1423-0194
DOI: 10.1159/000535944
Publisher site: See Article on Publisher Site

Abstract

IntroductionThe supramammillary nucleus (SuMN), localized in the posterior hypothalamus (PH) overlying the mammillary body, consists of a heterogeneous population of neurons (e.g., glutamatergic, GABAergic, neuropeptidergic, and dopaminergic) and exerts marked influences on a broad spectrum of brain areas that modulate cognition, ingestive behavior, reward and energy metabolism including the prefrontal cortex, hippocampus, mesolimbic system, hypothalamus, and raphe nuclei [1, 2]. In addition to its established role in integration and modulation of cognition and emotional aspects of goal-oriented behaviors [1, 3‒6], recent studies have revealed its role in the regulation of feeding [7‒9] and feeding-independent [10] fuel metabolism. We have previously reported that the neuronal activation state of the SuMN regulates peripheral fuel metabolism and body weight [10]. Circadian-timed activation of SuMN neuronal activity among high-fat diet (HFD)-induced insulin-resistant/glucose-intolerant rats via daily circadian-timed direct SuMN administration of glutamate receptor agonists so as to coincide with the natural circadian peak of general and dopaminergic SuMN neuronal activity in healthy rats reversed the HFD-induced insulin resistance (IR) and glucose intolerance [5, 10, 11]. On the other hand, chronic inhibition of SuMN neuronal activity with SuMN administration of GABAa receptor agonist/glutamate receptor antagonists induced IR, glucose intolerance, and obesity in rats held on regular chow diet [10]. It is presently not known which SuMN neuronal type(s) is/are involved in the induction of such metabolic responses, and the SuMN consists of a heterogenous population of neurons having extensive connections with multiple brain metabolic control centers [1, 2]. Of particular interest in this regard however is a subpopulation of dopaminergic neurons located at the medial SuMN that project directly to the suprachiasmatic nucleus (SCN) area as well as to other central nervous system (CNS) metabolic centers such as the lateral septum (LS), medial preoptic area (MPOA), and dorsomedial hypothalamus (DMH) [2, 11‒14].Circadian dopaminergic input activity at the clock SCN area that largely emanates from the medial SuMN (and coincides with the circadian peak of overall SuMN neuronal activity) [11], but also including from other brain centers, is a strong regulator of peripheral metabolism [11, 15]. A chronic reduction in circadian peak dopaminergic activity at the SCN area facilitates IR syndrome in healthy rodents [16], while circadian-timed administration of dopamine (DA) directly to the SCN at the time of day that it peaks in healthy animals reverses the IR syndrome [11, 17] (similar to the circumstance with overall SuMN activity regulation of metabolism described above). A composite of neurophysiological studies has demonstrated that circadian DA input activity at the SCN regulates its output control of metabolism via the neuroendocrine axis. Loss of the circadian peak of dopaminergic input activity at the SCN area via HFD feeding induces SCN neuronal output activity alterations that potentiate an elevation in sympathetic nervous system activity, hypothalamic-pituitary-adrenal axis corticosteroid hyperactivity, hyperinsulinemia, and leptin resistance [11, 16]. In turn, this SCN-driven neuroendocrine shift leads to a wide range of pathophysiological changes including systemic oxidative stress, low-grade inflammation, IR, and glucose intolerance [18] all of which collude to play pivotal roles in the development of obesity and type 2 diabetes mellitus [19‒24]. In addition to SuMN to SCN dopaminergic circuitry, SuMN dopaminergic input to the LS, MPOA, and DMH metabolic control centers may modulate (their) control of metabolism (via their communication with the SCN or otherwise).Therefore, the possibility exists that the previously identified metabolic effects of overall SuMN neuronal activity level on peripheral fuel metabolism may derive in large part from activity of medial SuMN dopaminergic neuronal circuits to the SCN and other relay metabolic centers (e.g., LS, MPOA, DMH) that in turn communicate with the SCN and influence peripheral metabolism. The present study was therefore undertaken to determine whether chronic tyrosine hydroxylase (TH, rate-limiting enzyme in DA synthesis [25]) knockdown (THx) at the medial SuMN area alters peripheral fuel metabolism in rodents. The SuMN-THx was accomplished via site-specific virus-mediated introduction of a short hairpin RNA (shRNA) which specifically targets TH directly to medial SuMN neurons that allows for stable transcription of the shRNA and long-term knockdown of TH expression. Knockdown of TH within CNS neurons is well established to produce a near-equivalent reduction level in neuronal DA content and dopaminergic function [26‒29]. To determine if SuMN-THx would induce progression to metabolic syndrome (MS), this intervention was introduced into a selected cohort of rats highly resistant to naturally developing MS over time based on model strain (female Sprague-Dawley [SD] rat), innate insensitivity to the obesogenic effect of HFD-resistant (HFDr), and maintained on a low-fat regular chow (RC) diet – female SD rat HFDr fed RC [f-HFDr-RC]. Additionally, subsequent experiments investigated metabolic effects of such SuMN-THx in (a) female HFDr rats fed HFD [f-HFDr-HFD], (b) male HFDr rats fed RC [m-HFDr-RC], and (c) female HFD-sensitive rats fed RC [f-HFDs-RC] over 1 year’s time to evaluate the influence of HFD, gender, and long-term response, respectively – upon body weight change, feeding, glucose tolerance, and insulin sensitivity. We hypothesized that such SuMN-THx treatment of MS-resistant animals would result in manifestation of glucose intolerance and obesity.Materials and MethodsAnimalsSD rats (female 10–12 weeks of age; male 6 weeks of age) (Taconic Biosciences, Hudson, NY, USA) were housed individually in a temperature and humidity-controlled room under a 14-h light daily photoperiod (14-h light and 10-h dark [14:10 LD]) with food and water ad libitum. Rats were allowed to adapt to the animal care facility for 1–2 weeks before the initiation of experiments. To best assess the potential MS-inducing impact of SuMN-THx (TH knockdown primarily confined to the medial SuMN plus minor involvement of immediate adjacent PH), studies were conducted in animals naturally resistant to development of MS. To avoid the confounder of age-induced IR upon the background metabolic status of the study animals during the study period, female SD rats were used in the primary studies (except in a separate study with males for gender comparison) inasmuch as they maintain a steady state of insulin sensitivity for a long period of their lifetime versus male rats of this strain that develop IR progressively from an early age [30, 31]. For the similar reason, animals were held on a long 14-h light daily photoperiod to avoid the confounder of a short ≤12-h light photoperiod-induced IR upon the background metabolic status of the study animals during the study period [32]. Finally, such female rats were selected to be resistant to the obesogenic effects of HFD (HFDr). See Experimental Design section of Methods for the protocol methodology to identify HFDr versus HFD-sensitive (HFDs) rats. Thus, the female SD rat resistant to the obesogenic effect of HFD and fed low-fat regular chow (RC) [f-HFDr-RC] was chosen as the primary model to evaluate the impact of SuMN-THx on peripheral fuel metabolism and feeding behavior. In subsequent experiments to assess the influence of HFD, gender, and longer term response, the same SuMN-THx intervention was administered to (a) female SD rat HFDr fed HFD [f-HFDr-HFD], (b) male SD rat HFDr fed RC [m-HFDr-RC], and (c) female SD rat HFDs fed RC [f-HFDs-RC] for over a year, respectively. Depending on the individual experimental design, rats were fed either regular chow (RC; 18% of calories/gram of food from fat, Envigo-Teklad rodent pellet 2018) or an HFD (60% of calories/gram of food from fat, Research Diets Inc, d12492). All animal experiments were conducted in accordance with the National Institutes of Health Guide for the Care and Use of Laboratory Animals [33] and with the protocols approved by the Institutional Animal Care and Use Committee of VeroScience, LLC.Viral ConstructsAdeno-associated viral plasmid (AAVp) containing a shRNA designed to target a distinct nucleotide stretch within the coding region of the tyrosine hydroxylase gene (Th) (AAVp-Th-shRNA-green fluorescent protein [GFP]) and control AAVp containing a shRNA targeting a scrambled nucleotide sequence (AAVp-Scr-shRNA-GFP) were kindly provided by Dr. RJ DiLeon of Yale University, CT, USA (Hommel JD 2003). The viral constructs were packed, purified, and prepared at ready-to-use titers expressed as GC/mL (genome copies per mL) (AAV5-Th-shRNA-GFP, 1.65 × 1013 GC/mL; AAV5-Scr-shRNA-GFP, 1.06 × 1013 GC/mL) by Vigene Biosciences (Rockville, MA, USA). Upon receiving, the viral solutions were aliquoted and stored at −80 C until use.Stereotaxic Surgery/Virus InoculationRats were anesthetized with ketamine/xylazine (80/12 mg/kg, intraperitoneally [ip]) and secured on a stereotaxic apparatus. A small craniotomy (∼1 mm in diameter) was made at 4.55 mm posterior from bregma and 0.2 mm from midline to allow access to the medial SuMN. A microinjector filled with a viral construct was mounted to the stereotaxic apparatus and slowly lowered to the medial SuMN according to the coordinates: 4.55 mm posterior from bregma, 0.2 mm from midline, 8.4 mm below bregma. A volume of 0.2 μL of the viral construct (AAV5-Th-shRNA-GFP, 1.65 × 1013 GC/mL or AAV5-Scr-shRNA-GFP [control], 1.06 × 1013 GC/mL) was continuously microinfused over 2 min. At the end of infusion, the injector was left in place for additional 5 min before slow withdrawal. The incision was sutured, and the rat was returned to the colony and allowed to rest for a minimum of 2 weeks to ensure sufficient viral expansion before experimentation.Tissue Processing/ImmunohistochemistryThe efficiency of a virus-mediated knockdown of Th gene expression was verified by immunohistochemistry. Rats with medial SuMN inoculation of AAV5-Th-shRNA-GFP or control AAV5-Scr-shRNA-GFP were anesthetized with ketamine/xylazine (80/12 mg/kg, ip) and sacrificed by transcardiac perfusion with 4% paraformaldehyde. The brains were collected, post-fixed in 4% paraformaldehyde overnight at 4°C, cryoprotected in 30% sucrose until brains were fully fixed. The brains were then frozen in 2-methylbutane chilled on dry ice and kept at −80°C till being further processed. Sequential coronal brain sections (30 μm in thickness) were cut using a cryostat at −20°C and collected in PBS buffer. The free-floating sections were transferred to a freezing solution (30% ethylene glycol, 25% glycerol prepared in 0.05 m phosphate buffer) and kept at −20°C. For TH-3,3′-diaminobenzidine (DAB)-immunohistochemical staining, the free-floating sections were washed in PBS and treated with 3% hydrogen peroxide for 10 min to quench the endogenous peroxidase activity. The brain sections were then rinsed in PBS and incubated with Animal-Free Blocker (SP-5030, Vector Lab) for 1 h at room temperature to block background non-specific binding. The brain sections were subsequently incubated with monoclonal mouse anti-TH antibody (1:2,000) (MAB318, Millipore) overnight at 4°C. After thorough washing, the brain sections were incubated with biotinylated anti-mouse IgG for 1 h at room temperature followed by ABC complex amplification. The TH immunoreactivity was revealed by a peroxidase reaction using DAB as chromogen. Finally, the stained brain sections were sequentially mounted on gelatin-coated micro-slides, and dehydrated and covered with Permount for light microscopy. For TH-immunofluorescent staining, the brain sections were processed in sequential steps as follows with PBS washing between each step: blocked with 10% normal goat serum (Vector S-1000) for 1 h at room temperature, incubated in monoclonal rabbit antibody against TH (1:250) (Thermo Fisher Scientific, Cat#701949) overnight at 4 C, followed by incubation in Rhodamine Red X (RRX)-goat anti-rabbit IgG (1:200) (Jackson ImmunoResearch, Cat# 111-296-144) for 1 h at room temperature. The antibodies were prepared in PBS containing 5% normal donkey serum and 0.2% Triton-X. At the end of staining, the brain sections were wet mounted on charged micro-slides and covered with VectorShield medium with DAPI (Vector Lab, H1200). Fluorescent images were acquired using BioTek Lionheart FX automated microscope and imaging software (BioTek Instruments, Inc., Winooski, VT, USA) under Texas Red, GFP, and DAPI filters. The acquired images were subsequently analyzed using BioTek image analysis software (BioTek Instruments, Inc.) to obtain cell count and mean fluorescent density with background correction.Experimental DesignStudy 1: Verification of the Extent of THx at the SuMN Area (SuMN-THx)To verify the extent of TH knockdown, 12 female SD rats were separated into 2 groups and surgically inoculated at the medial SuMN area with the viral construct AAV5-Th-shRNA-GFP to knockdown Th gene expression (THx, N = 7) or control AAV5-Scr-shRNA-GFP (Ctr, N = 5). The rats were sacrificed at 3 months post-virus inoculation by transcardiac perfusion with 4% paraformaldehyde, and the brains were collected for immunohistochemistry. Coronal brain sections (30 μm in thickness) were cut through the SuMN and sequentially collected at 90 μm intervals into 3 sets in PBS buffer. One set of brain sections was processed with TH immunostaining using DAB as a chromogen. Another set was processed with TH(+)-RRX-immunofluorescence red staining. TH(+)-DAB-stained brown cells at the SuMN area were counted on 6 representative sections through the SuMN (bregma −4.4, −4.5, −4.6, −4.7, −4.8, and −4.9 mm). Localization of TH (+)- red fluorescence inside virus-transfected GFP-expressing green cells at the SuMN area was examined in THx and Ctr brains. To verify that the inoculated virus did not spill over to the adjacent VTA DA neurons, the spread of virus-transfected GFP-expressing cells at adjacent VTA was evaluated. Moreover, the TH(+)-RRX-immunofluorescent images at VTA of THx and Ctr brains were acquired and subsequently quantified by measuring mean fluorescence density and fluorescent red cell counts (described in detail in Tissue Processing/Immunohistochemistry) on the brain sections cut through the rostral (bregma, −4.5 mm) and caudal (bregma, −4.8 mm) SuMN, respectively.Study 2: Evaluation of the Impact of SuMN-THx on Body Fuel Metabolism and Feeding Behavior in a Rat Model Least Prone to Develop Metabesity – f-HFDr-RC RatsStudy 2Ai–iii: The Effect of SuMN-THx on Body Weight, Food Consumption, Feed Efficiency, Glucose Metabolism, Plasma Metabolic Parameters, Liver Metabolic Gene Expression, and Adipose Fatty Acid Oxidation (FAO) Rate in f-HFDr-RC Rats. Female SD rats were separated into HFDr and HFD-sensitive cohorts using the median % body weight gain over 1 week of high fat feeding (=8%) as separation reference. Half of the rats with % body weight gain less than the median % body weight gain (≤8%) were considered as HFDr rats [34, 35] and employed in the present study. Such selected HFDr rats (BW = 259 ± 4.0 g, 15 weeks old) were fed RC and randomly divided into two groups for local virus inoculation at the medial SuMN area of AAV5-Th-shRNA-GFP to knockdown Th gene expression (THx, N = 12) or control AAV5-Scr-shRNA-GFP (Ctr, N = 10). Throughout the study period, body weight and food consumption were monitored. A glucose tolerance test (GTT) was performed at 11 weeks post-virus inoculation. The rats were sacrificed by decapitation at 17 weeks post-virus inoculation during the fasting period of the day (zeitgeber time, 6–8 [ZT 6–8 h after light onset]). The trunk blood was collected for plasma assay of metabolic parameters. Parametrial white adipose tissue (pWAT) and retroperitoneal white adipose tissue (rWAT) were weighed. Adipocyte oxygen consumption rate (OCR), a measure of cellular metabolic activity, and FAO were each measured in freshly collected pWAT from 7 pairs of rats (7 THx and 7 Ctr). Hepatic tissues were quickly frozen in liquid nitrogen for RT-quantitative PCR (qPCR) analysis of the mRNA expression of metabolic relevant genes and triglyceride content measurement. The brains were collected for immunohistological verification of TH knockdown at the SuMN.Study 2Bi–v: The Effect of SuMN-THx on Feeding, Food Preference (HFD or RC), Hunger/Satiety Balance, and Feeding Response to Systemic Ghrelin or Glucagon-Like Polypeptide −1 (GLP-1) Agonist Administration in f-HFDr-RC Rats. Female HFDr rats were selected as described in detail above and randomly assigned into one of 2 groups receiving an inoculation of either AAV5-Th-shRNA-GFP (THx, N = 15) or AAV5-Scr-shRNA-GFP (Ctr, N = 10) at the medial SuMN and utilized in a series of sequential feeding paradigm studies as follows. Rats were allowed to rest for 1–2 weeks between experiments. Study 2Bi: The Impact of SuMN-THx on circadian food intake rhythm: Regular chow consumption in THx and Ctr rats was measured over a 24-h period for two consecutive days between ZT (hours after light onset) 0, 8, 14, 19. Food consumption during the light and dark periods of the day and total daily food consumption were calculated. Study 2Bii: The impact of SuMN-THx on HFD preference versus RC: Rats accustomed to RC for 3 weeks were given additional access to HFD, and preferential feeding was monitored daily during the subsequent 4 days (cumulative food consumption of each HFD and RC). On day 5 of the study, the RC was removed leaving HFD for 1 week. Next, RC was re-introduced in addition to the HFD and daily food consumption (HFD and RC) was subsequently monitored daily for the following 7 days. Study 2Biii: The impact of SuMN-THx on fasting-refeeding response: Rats maintained on RC were fasted overnight (16 h) during the normal feeding period of the day by removing RC 2 h before light offset (at ZT12). RC was re-introduced the next morning at ZT4, and cumulative food consumption was subsequently measured at 0.5 h, 1 h, 2 h, and 4 h after this initial refeeding time. Study 2Biv: The impact of SuMN-THx on ghrelin-induced food intake: Rats were injected ip with saline or ghrelin (Tocris Bioscience, Cat# 1,465) during the natural fasting (light) period of the day in a treatment cross-over design separated by a washout day. Rats (THx and Ctr) fed RC were injected with saline vehicle (0.2 mL/rat, ip) at ZT5.5–6, and cumulative baseline food consumption was measured at 1 h and 6 h post-vehicle injection. After a 1-day washout period, the THx and Ctr rats received a ghrelin injection (13 μg/kg, ∼0.2 mL/rat, ip) at ZT5.5–6. Cumulative RC consumption was measured at 1 h and 6 h post-ghrelin injection. Study 2Bv: The impact of SuMN-THx on exendin-4 (Ex-4)-induced hypophagia: THx and Ctr rats were injected with saline or the GLP-1 agonist, Ex-4 (Tocris Bioscience, Cat# 1933) during the natural feeding period (dark period of the day) in a treatment cross-over design separated by a washout day. On the day of study initiation, RC was removed from the THx and Ctr animals’ cages at 2 h before light offset (ZT12) and then re-introduced at the time of saline injection (ZT14–14.5). Cumulative RC consumption was measured under a red light at 1 h, 3 h, and 16 h post-saline injection to obtain baseline food consumption. After a 1-day washout period, following removal of the RC at ZT12, the THx and Ctr rats were then injected with Ex-4 (5 μg/kg, ip, 0.2 mL/rat) at ZT14–14.5 followed by an immediate re-introduction of RC, and cumulative RC consumption was measured under a red light at 1 h, 3 h, and 16 h post Ex-4 injection.Study 3: The Effect of SuMN-THx on Body Weight and Glucose Metabolism in f-HFDr-HFD RatsHFDr rats (BW = 265 ± 2.2 g, 15 weeks old) were selected after 1 week of HFD feeding based on % body weight gain less than the median % body weight gain (≤8%). Such selected HFDr rats were randomly divided into 2 groups receiving inoculation of AAV5-Th-shRNA-GFP (THx, N = 24) and AAV5-Scr-shRNA-GFP (Ctr, N = 18), respectively, at the medial SuMN, and fed HFD instead of regular chow during the study period. Body weight and food consumption were monitored. A GTT was performed at 7 weeks post-virus inoculation. At 8 weeks post-virus inoculation, rats were sacrificed by decapitation at the middle of the light period (ZT 6–8). The pWAT from 6 pairs of rats (6 THx and 6 Ctr) was collected for measurement of adipocyte OCR. The trunk blood was collected for plasma triglyceride assay.Study 4: The Effect of SuMN-THx on Body Weight and Glucose Metabolism in m-HFDr-RC ratsHFDr male rats (BW = 321 ± 3.8 g, 9 weeks old) were selected after 1 week of high fat feeding based on % body weight gain less than the median % body weight gain (≤17%) [30] and divided into 2 groups, receiving inoculation of AAV5-Th-shRNA-GFP (THx, N = 12) and AAV5-Scr-shRNA-GFP (Ctr, N = 12), respectively, at the medial SuMN. The rats were fed regular chow during the study period. Body weight and food consumption were monitored throughout the study period. A GTT was performed at 20 weeks post-virus inoculation.Study 5: Long-Term (1 Year) Metabolic Effect of SuMN-THx in f-HFDs-RC RatsHFD-sensitive rats (BW = 269 ± 2.2 g, 13 weeks old) with % body weight gain greater than the medium % body weight gain (>8%) after 1 week of high fat feeding were randomly divided into 2 groups and subsequently inoculated at the medial SuMN area with AAV5-Th-shRNA-GFP (THx, N = 10) or AAV5-Scr-shRNA-GFP (Ctr, N = 10) while maintained on regular chow during the study period. Body weight was monitored at the beginning, at 6 months and 12 months post-AAV inoculation. The rats were sacrificed by decapitation at 1-year post-virus inoculation during the fasting period of the day (ZT 6–8). The trunk blood was collected for the measurement of fasting glucose and insulin. The brains were collected for immunohistochemical verification of TH knockdown at the SuMN area.Blood/Plasma Assays of Metabolic ParametersBlood glucose was measured by a blood glucose monitor (OneTouch Ultra; LifeScan, Inc., Milpitas, CA, USA). Plasma insulin, leptin, and corticosterone concentrations were determined by EIA using commercially available assay kits (ALPCO Diagnostics, Salem, NH): insulin ELISA (80-INSMR), leptin ELISA (22-LEPMS), and corticosterone ELISA (55-CORMS). Belfiore insulin sensitivity index (ISI) (pmol/L*mmol/L*h) was calculated based on the formula: 2/((insulin GTT AUC * glucose GTT AUC) + 1). HOMA-IR was calculated based on the formula: fasting glucose (mmol/L) × fasting insulin (mU/L)/22.5. Hepatic triglyceride was extracted from liver tissue that was homogenized in 5% NP-40, heated to 90°C for 3 min, centrifuged at 10,000 g for 10 min to collect the supernatant. The triglyceride content in the supernatant was subsequently determined by TG Colorimetric Assay Kit (Item No. 10010303; Cayman Chemical, Ann Arbor, MI, USA). Plasma triglycerides were measured with Infinity™ Triglycerides Kit (TR22421; Thermo Fisher Scientific). Plasma free fatty acids (FFAs) were measured using a Free Fatty Acid Quantification Kit (MAK044; Sigma).Glucose Tolerance TestA GTT was performed during the fasting period of the day (ZT 6–8) by injecting rats with 50% dextrose solution (3 g/kg BW, ip, in all female rat studies; 2.5 g/kg BW, ip, in the male rat study 4). Blood samples were taken from the tail before and 30, 60, 90, 120 min after dextrose injection for blood glucose and plasma insulin analyses.qPCR Analysis of Hepatic mRNATotal RNA was isolated from frozen hepatic tissue samples with TRIzol Reagent (Cat # 15596026; Thermo Fisher Scientific, Waltham, MA, USA). Total RNA quantity and purity were determined by UV spectroscopy, and sample RNA concentrations were normalized prior to reverse transcription reaction. cDNA was synthesized using SuperScript IV VILO Master Mix with ezDNAse (Cat#11766050; Thermo Fisher Scientific, Waltham, MA, USA) from 2 μg of total RNA per sample. Real-time qPCR was performed with TaqMan fast advance master mix (Thermo Fisher Scientific, Waltham, MA, USA Cat# 4444964) on a AriaMx (Agilent, Santa Clara, CA, USA) qPCR equipment with the primer/probe sets as listed below. The relative expression of genes was calculated using the 2−∆∆Cq method. Ribosomal protein S17 was quantified with TaqMan assay (Cat# Rn00820807_g1; Thermo Fisher Scientific, Waltham, MA, USA) and used as the internal reference, and the relative gene expression was calculated as a fold change from control samples. Primers for respective mRNAs were purchased from Thermo Fisher Scientific (Waltham, MA, USA) as follows: phosphoenolpyruvate carboxy kinase 1 (Pck1) (Cat# Rn01529014_m1); glucose-6-phosphatase, catalytic subunit (G6pc) (Cat# Rn00689876_m1); sterol regulatory element binding transcription factor 1 (Srebf1) (Cat# Rn01495769_m1); fatty acid synthase (Fasn) (Cat# Rn00569117_m1); mechanistic target of rapamycin kinase (Mtor) (Cat# Rn00693900_m1); PPARG coactivator 1 beta (Ppargc1β) (Cat# Rn00598552_m1); peroxisome proliferator-activated receptor alpha (Pparα) (Cat# Rn00566193_m1); peroxisome proliferator-activated receptor gamma (Pparϒ) (Cat# Rn00440945_m1); PPARG coactivator 1 alpha (Ppargc1α) (Cat# Rn00580241_m1); nuclear factor kappa B subunit 1 (Rela) (Cat# Rn01502266_m1); suppressor of cytokine signaling 3 (Socs3) (Cat# Rn01470502_g1); mitogen-activated protein kinase 8 (Mapk8) (Cat# Rn01218952_m1); C-C motif chemokine ligand 2 (Ccl2) (Cat# Rn00580555_m1); nuclear receptor subfamily 1, group D, member 1 (Nr1d1) (Cat# Rn01460662_m1); period circadian regulator 1 (Per1) (Cat# Rn01496757_m1); Clock circadian regulator (Clock) (Cat# Rn00573120_ml).Analysis of OCRA 24-well format cell-based assay that measures fatty acid-driven oxygen consumption of adipocytes was conducted using the XFe24 Extracellular Flux Analyzer (Seahorse Bioscience, Agilent Technologies, Lexington, MA, USA). pWAT was freshly dissected from a TH knockdown rat and a control rat and placed in one XF24 Islet Capture Microplate (8 mg adipose tissue/per well) submerged in FAO assay medium. OCR (pmol/min/mg) was assessed at 37°C at baseline (basal) over 32 min (total oxygen consumption representing total tissue fuel oxidation) and in the subsequent presence of either 100 μm palmitate-bovine serum albumin or 400 μm etomoxir, a carnitine palmitoyltransferase-1 inhibitor over 64 min. Exogenous FAO (Extra-FAO) was calculated as palmitate-bovine serum albumin OCR above baseline, and endogenous FAO was calculated as baseline OCR minus etomoxir OCR (which represents non-FAO). Each of total oxidation, exogenous FFA oxidation, and endogenous FFA oxidation OCR values from control rats were normalized as 100%, and the relative % changes of THx versus control OCRs were calculated. The data were collected from 7 pairs of rats (7 THx vs. 7 Ctr) in Study 2A and 8 pairs of rats (8 THx vs. 8 Ctr) in Study 3, analyzed using the software “Wave” (Agilent Technologies).Data AnalysisAll data are expressed as mean ± standard error of the mean. Parametric tests were used when groups passed normality (using the Shapiro-Wilk test). Means between two groups were compared by two-tailed Student’s t test. In instances wherein the direction of change was anticipated, a one-tailed Student’s t test was employed. An observed p value <0.05 (p < 0.05) was accepted as statistically significant indicated by * (two-tail) or ƚ (one-tail). In multi-group design (>2 groups), data was analyzed by a two- or three-way repeated measures ANOVA followed by post hoc analysis of pair-wise comparisons using a Bonferroni-corrected α level. β indicates an observed p value less than a Bonferroni-corrected α level as significant. Only rats with the cannula tip localized inside, on the boundary of, or within 150 μm dorsal from the medial SuMN were included in statistical analyses.ResultsStudy 1: Verification of the Extent of THx at the SuMN Area (SuMN-THx)Among brains inoculated with AAV5-Th-shRNA-GFP or AAV5-Scr-shRNA-GFP, a comparable localized distribution of virus-transfected GFP-expressing neurons was observed at the SuMN area (bregma, from −4.4 mm to −4.9 mm) encompassing the medial SuMN and adjacent medial PH (within ∼400 μm dorsal to the medial SuMN). The extent of the viral-mediated TH knockdown was evaluated by TH immunohistochemistry in Th-shRNA versus Scr-shRNA brains. In the Scr-shRNA brains, distinct TH-immunopositive cell bodies and processes were observed mostly clustering within the medial SuMN and to a much lesser extent scattering in the adjacent PH dorsal to the medial SuMN (shown in Fig. 1A. a–f). Dense TH immunoreactivities were visible bilaterally at the VTA at the level of SuMN (shown in Fig. 1A. a, c, e). In contrast, the TH-immunopositive cell bodies in Th-shRNA brains were almost completely eliminated at such defined SuMN area while leaving the TH-immunopositive processes moderately detectable (shown in Fig. 1A. g–l). At the adjacent bilateral VTA, robust TH-immunopositive bodies were detected in the Th-shRNA brains with a density comparable to that in Scr-shRNA brains (shown in Fig. 1A. g, i, k). Immunofluorescence labeling revealed that the virus-transfected GFP-expressing cells in the medial SuMN express TH(+)-RRX red fluorescence in the control Scr-shRNA brain (shown in Fig. 1B. b–d), while in Th-shRNA brains TH(+)-RRX red cell bodies were barely detectable at the medial SuMN (shown in Fig. 1B. g–i). Cell counts of TH(+)-DAB cells on 6 brain sections across the SuMN area (bregma −4.4, −4.5, −4.6, −4.7, −4.8, −4.9 mm) revealed 83% reduction in TH(+)-DAB cells in Th-shRNA brains compared to Scr-shRNA brains (p < 0.0001) (shown in Fig. 1A. m, n). At the adjacent VTA, no detectable reduction in the TH(+)-RRX immunofluorescence in Th-shRNA brains versus Scr-shRNA brains was found via measurement of the mean fluorescence density (bregma, −4.5 mm) or TH(+)-red cell count (bregma −4.8 mm) (shown in Fig. 1C). Thus, the TH knockdown was largely restricted to the medial SuMN and to the sparsely scattered TH-positive neurons in the adjacent PH representing a near complete knockdown of the total TH dopaminergic neurons in such defined SuMN area. Whereas no significant difference in the counts of DAPI nuclear staining at the SuMN area was detected in Th-shRNA brains versus Scr-shRNA brains (data not shown), the TH knockdown at the SuMN area (SuMN-THx) remains relatively stable over the experimental periods as verified by TH immunohistochemistry in the Th-shRNA brains collected at 17 weeks (from Study 2A) and 1 year (from Study 5) after surgical inoculation of the viral construct AAV5-Th-shRNA-GFP (data not shown).Fig. 1.A Verification of Th gene expression knockdown at the SuMN area with TH-immunopositive DAB staining. The extent of Th gene expression knockdown via a virus-mediated shRNA was evaluated by TH immunohistochemistry using DAB as a chromogen. The images represent the TH-immunopositive DAB staining [TH(+)-DAB] of Scr-shRNA versus Th-shRNA brain sections at the rostral (bregma −4.52), middle (bregma −4.70), and caudal (bregma −4.80) levels of the SuMN. In Scr-shRNA brain sections (low magnification (a, c, e) and high magnification (b, d, f)), the DAB staining revealed TH(+)- cell bodies and processes clustered at the medial SuMN and scattered around the adjacent medial posterior hypothalamic area dorsal to the medial SuMN. Dense TH(+)- immunoreactivity at the VTA was detected bilaterally. In Th-shRNA brain sections (low magnification (g, i, k) and high magnification (h, j, l)), TH(+)- cell bodies were no longer detectable in the medial SuMN and dramatically reduced in the adjacent medial PH (within ∼400 μm from dorsal SuMN) leaving TH(+)- processes readily visible. Strong and comparable TH(+)- immunoreactivity was bilaterally detected at the VTA on Th-shRNA and Scr-shRNA brain sections. Cell counts of TH(+)-DAB cells on 6 brain sections across the SuMN area (bregma −4.4, −4.5, −4.6, −4.7, −4.8, −4.9 mm) revealed an 83% reduction in TH(+)-DAB cells in Th-shRNA brains compared to Scr-shRNA brains (p < 0.001) (m, n). The red dotted lines demarcate the contour of the medial SuMN. SuMN m, supramammillary nucleus, medial; mp, mammillary peduncle; VTAr, ventral tegmental area, rostral; SN, substantia nigra; 3V, 3rd ventricle. *p < 0.05, two-tailed Student’s t test. B TH-immunofluorescent staining at the SuMN area. The TH-immunopositive cells were labeled with RRX (red) fluorescence in Scr-shRNA brain (a–e) and Th-shRNA brain (f–j) samples. A cluster of TH(+)- red fluorescent cells was detected in the medial SuMN in Scr-shRNA brain (a, b), while no distinct red fluorescent cell bodies were detected in Th-shRNA brain (f, g). Virus-transfected neuronal cells expressed GFP in Scr-shRNA brain (c) and Th-shRNA brain (h). The zoom-in images showed red fluorescence inside the virus-transfected GFP-expressing cells in Scr-shRNA brain as indicated by arrows (b, c, d). No red immunofluorescence was observed inside the GFP-expressing cells in Th-shRNA brain (g, h, i). DAPI nuclear staining showed apparent normal histological morphology in Scr-shRNA (e) and Th-shRNA (j) brains. C TH-immunofluorescent staining at the VTA TH (+)- red immunofluorescent staining was quantified by mean fluorescence density and cell counts at the VTA. a, b Representative images of TH(+)- red immunofluorescence at the VTA adjacent to rostral SuMN (bregma −4.2 mm). There was no reduction in the mean density of red fluorescence in the Th-shRNA brain versus Scr-shRNA brain (p = 0.333) (c). d, e Representative images of TH(+)- red immunofluorescence at the VTA adjacent to caudal SuMN (bregma −4.8 mm). There was no reduction in the TH(+)- immunofluorescent red cell counts in shRNA brain versus Scr-shRNA brain (p = 0.804) (f).Study 2A: SuMN-THx Induces Obese MS (Metabesity) in f-HFDr-RC RatsStudy 2Ai: SuMN-THx Induces Obesity and Glucose Intolerance and Reduces Adipose FAO RateCompared to Ctr rats, measured at 11 weeks post-virus inoculation, THx rats exhibited a 42% increase in body weight gain (50 ± 3.5 g gain vs. 35 ± 2.3 g gain in Ctr rats, p = 0.003) (shown in Fig. 2a, b). The body weight measured over time showed a significant main effect of time (two-way repeated measures ANOVA, F(12, 240) = 114.47, p < 0.001), but no significant main effect of THx (F(1, 20) = 3.02, p = 0.098). The interaction between time and THx was significant (F(12, 240) = 4.48, p < 0.001). The THx treatment relative to Ctr also increased food consumption by 15% (p = 0.0001) (shown in Fig. 2c, d) with a 25% increase in feed efficiency (p = 0.048) (shown in Fig. 2e). The food consumption over time showed a significant main effect of time (F(11, 220) = 2,899.36, p < 0.001) and of THx (F(1, 20) = 15.89, p = 0.001) as well as a significant interaction between time and THx (F(11, 220) = 13.31, p < 0.001). By the time of sacrifice at 17 weeks post-virus inoculation, THx rats had gained 83 ± 4.8 g in body weight over the 17-week period which was 48% more than Ctr rats (body weight gain = 56 ± 5.5 g) (p = 0.001) (shown in Fig. 2a). The WAT mass was significantly increased in THx rats compared to Ctr rats: parametrial (p)WAT increased by 45% (p = 0.008); retroperitoneal (r) WAT increased by 37% (p = 0.041); and total (tl) WAT increased by 42% (p = 0.012) (shown in Fig. 2f). Finally, FAO of pWAT was reduced in THx versus Ctr rats (total basal oxidation by 51%, p < 0.0001; endogenous FAO by 58%, p < 0.0001; Extra-FAO by 54%, p = 0.024) (shown in Fig. 2g). Thus, such THx treatment increased body weight/fat and feed efficiency while reducing free FAO of adipocytes. Compared with Ctr, THx treatment induced glucose intolerance as indicated by an elevated GTT blood glucose level over time (at 0, 30, 60, 90, and 120 min) resulting in an increase in GTT glucose AUC by 27% relative to Ctr (p = 0.009) (shown in Fig. 3b). The basal blood glucose level was elevated in THx rats compared to Ctr rats (p = 0.024) (shown in Fig. 3a, insert). The results of the GTT demonstrated a main effect of time (two-way repeated measures ANOVA, F(4, 60) = 85.26, p < 0.001) and a main effect of THx (F(1, 15) = 0.007) with no significant interaction between time and THx treatment (F(4, 60) = 1.73, p = 0.154) in GTT glucose. However, only a significant main effect of time (F(4, 60) = 29.28, p < 0.001) was detected in GTT insulin with no significant main effect of THx (F(1, 15) = 0.001, p = 0.978) nor significant interaction between time and THx (F(4, 60) = 0.142, p = 0.966). Thus, GTT insulin AUC was not altered by THx treatment (shown in Fig. 3d), and as a result, there was no significant reduction in Belfiore ISI as compared with Ctr (17%, p = 0.258) (shown in Fig. 3e).Fig. 2.THx at the SuMN area induces obesity, increased feeding and feeding efficiency while decreasing FAO in f-HFDr-RC rats (Study 2Ai). f-HFDr-RC rats were locally inoculated at the medial SuMN area with AAV5-Th-shRNA-GFP to knockdown Th gene expression (THx) or control AAV5-Scr-shRNA-GFP (Ctr). Compared with Ctr, THx treatment increased body weight gain by 42% (p = 0.003) (a, b) and food consumption by 15% (p = 0.0001) (c, d) with a concurrent increase in feed efficiency (25%, p = 0.048) over 72 days (e) post-virus inoculation. When sacrificed at 17 weeks post-virus inoculation, THx rats relative to Ctr rats had an increase in body weight gain (48%, p = 0.001) (a) and WAT weight (parametrial pWAT by 45%, p = 0.008; retroperitoneal rWAT by 37%, p = 0.041; total tl WAT by 42%, p = 0.012) (f). Fatty acid oxidation (FAO) of pWAT (8 mg) was reduced by THx treatment versus Ctr (basal FAO, 51%, p < 0.0001; endogenous FAO [Endo-FAO], 58%, p < 0.0001; exogenous FAO [Exo-FAO], 54%, p = 0.024) (g). *p < 0.05, two-tailed Student’s t test.Fig. 3.THx at the SuMN area induces glucose intolerance in f-HFDr-RC rats (Study 2Ai). f-HFDr-RC rats were locally inoculated at the medial SuMN area with AAV5-Th-shRNA-GFP to knockdown Th gene expression (THx) or control AAV5-Scr-shRNA-GFP (Ctr). At 11 weeks post-virus inoculation, compared with Ctr, THx treatment increased basal blood glucose (p = 0.024) and GTT blood glucose level at 30, 60, 90, and 120 min following glucose challenge (a) with an increase in GTT glucose AUC (27%, p = 0.009) (b). GTT insulin AUC was not increased by THx treatment (c, d). The reduction in Belfiore ISI did not reach significance in THx rats versus Ctr rats (17% reduction, p = 0.258) (e). *p < 0.05, two-tailed Student’s t test.Study 2Aii: SuMN-THx Alters the mRNA Expression Levels of Metabolic, Inflammatory, and Clock Genes in the LiverThe relative mRNA expression levels of key metabolic, inflammatory, and circadian genes were analyzed using RT-qPCR in the livers collected from THx and Ctr rats during the fasting period of the day (ZT 6–8) at 17 weeks following viral inoculation. Compared with Ctr, THx increased relative mRNA expression of key metabolic genes involved in gluconeogenesis, Pck1 and G6pc, by 2.3-fold (p = 0.0013) and 1.6-fold (p = 0.046), respectively (shown in Fig. 4a, gluconeogenic genes). Compared with Ctr, THx increased the relative mRNA expression of the key lipogenic transcriptional coregulator Ppargc1β by 1.4-fold (p = 0.037 Ɨ) with no significant increase in Fasn gene (p = 0.6) or the lipogenic transcriptional regulators (Srebf1, p = 0.9; Mtor, p = 0.9) (shown in Fig. 4b, lipogenic genes). The mRNA expression levels of the key transcriptional factors/coactivators regulating FAO were not significantly altered in THx relative to Ctr rats: Pparα (p = 0.105), Pparγ (p = 0.314), or Ppargc1α (p = 0.203) (shown in Fig. 4c, lipid oxidation genes). Compared with Ctr, THx increased the relative mRNA expression levels of Rela (1.2-fold, p = 0.021) and Ccl2 (1.4-fold, p = 0.037 Ɨ) with a non-significant increase in Socs3 (p = 0.148) and Mapk8 (p = 0.106) (shown in Fig. 4d, pro-inflammatory genes). Compared with Ctr, THx treatment increased the relative mRNA expression levels of Clock (1.4-fold, p = 0.0002) and Per1 (1.8-fold, p = 0.036 Ɨ) with no increase in Nr1d1 (p = 0.33) (shown in Fig. 4e, circadian genes).Fig. 4.THx at the SuMN area alters mRNA expression of metabolic relevant genes in the liver of f-HFDr-RC rats (Study 2Aii). f-HFDr-RC rats were locally inoculated at the medial SuMN area with AAV5-Th-shRNA-GFP to knockdown Th gene expression (THx) or control AAV5-Scr-shRNA-GFP (Ctr). The relative mRNA expression levels of metabolic relevant genes were analyzed by RT-qPCR in the livers of THx and Ctr rats collected during the fasting period of the day (ZT 6–8). a Gluconeogenic genes: compared with Ctr, THx treatment increased relative mRNA expression of phosphoenolpyruvate carboxylase 1 (Pck1) by 2.3-fold (p = 0.0013) and glucose-6-phosphatase, catalytic subunit (G6pc) by 1.6-fold (p = 0.046). b Lipogenic genes: compared with Ctr, THx increased the relative mRNA expression of lipogenic transcriptional coregulator Ppargc1β by 1.4-fold (p = 0.037 Ɨ) with no detectable increase in fatty acid synthase (Fasn) (p = 0.6) and lipogenic transcriptional regulators (Srebf1, p = 0.9; Mtor, p = 0.9). c Lipid oxidation genes: no significant increase was detected in the mRNA expression levels of the key transcriptional factors/coactivators regulating FAO in the livers of THx rats versus Ctr rats, Pparα (p = 0.105), Pparγ (p = 0.314), or Ppargc1α (p = 0.203). d Pro-inflammatory genes: compared with Ctr, THx treatment increased the relative mRNA expression levels of Rela (1.2-fold, p = 0.021) and Ccl2 (1.4-fold, p = 0.037 Ɨ) with no significant increase in suppressor of cytokine signaling 3 (Socs3) (p = 0.148) and mitogen-activated protein kinase 8 (Mapk8) (p = 0.106). e Circadian genes: compared with Ctr, THx treatment increased the relative mRNA expression levels of Clock (1.4-fold, p = 0.0002) and Per1 (1.8-fold, p = 0.036 ƚ) with no effect on Nr1d1 (p = 0.33). *p < 0.05, two-tailed Student’s t test. Ɨp < 0.05, one-tailed Student’s t test.Study 2Aiii: SuMN-THx Induces Hyperleptinemia and Hypertriglyceridemia, and Alters the Plasma Level of CorticosteroneA prototypical profile of metabolic parameters associated with MS [16, 36] was induced by THx at the SuMN. Compared with Ctr, THx treatment increased plasma leptin (46%, p = 0.0382 Ɨ) (shown in Fig. 5a) and plasma triglyceride (86%, p = 0.008) levels (shown in Fig. 5b); however, there was no significant increase in liver triglyceride (p = 0.235) (shown in Fig. 5c) or plasma FFAs (p = 0.153) (shown in Fig. 5d). The plasma corticosterone level assessed during the fasting period of the day (ZT 6–8) was reduced by 68% (p = 0.007) in THx rats versus control rats (shown in Fig. 5e) suggesting a change/shift in the plasma corticosterone circadian rhythm.Fig. 5.THx at the SuMN area induces hyperleptinemia, hypertriglycemia and alters plasma corticosterone in f-HFDr-RC rats (Study 2Aiii). f-HFDr-RC rats were locally inoculated at the medial SuMN area with AAV5-Th-shRNA-GFP to knockdown Th gene expression (THx) or control AAV5-Scr-shRNA-GFP (Ctr). Compared with Ctr, THx treatment increased plasma leptin by 46% (p = 0.0382 Ɨ) (a) and plasma triglyceride by 86% (p = 0.008) (b) with no increase in liver triglyceride content (p = 0.235) (c) or plasma level of FFAs (p = 0.153) (d). Plasma corticosterone measured during fasting period of the day (ZT 6–8) was reduced by 68% (p = 0.007) (e) in THx rats versus control rats. *p < 0.05, two-tailed Student’s t test. ƚp < 0.05, one-tailed Student’s t test.Study 2B: SuMN-THx Alters the Circadian Feeding Rhythm, Increases HFD Consumption in Food Preference Paradigms, Alters Hunger/Satiety Balance, and Attenuates the Feeding Response to Systemic Ghrelin Administration (Ghrelin Resistance) in f-HFDr-RC RatsStudy 2Bi: SuMN-THx Increases Food Consumption during the Natural Fasting Period of the DayRegular chow consumption was monitored throughout light (resting/fasting) and dark (active/feeding) periods of the day in THx and Ctr rats over 2 days (shown in Fig. 6a1). There was a significant main effect of light/dark photoperiod of the day (two-way repeated measures ANOVA, F(1, 24) = 363.43, p < 0.001) such that rats ate mostly during the dark and less during the light period (dark: 15.7 ± 0.51g/rat; light: 3.3 ± 0.23g/rat). There was a significant main effect of TH knockdown (THx) (F(1, 24) = 7.07, p = 0.014) on total daily food consumption such that THx rats ate 14% more daily food than Ctr rats (10.12 ± 0.3g/THx rat vs. 8.89 ± 0.35g/Ctr rat) (shown in Fig. 6a2). The interaction between photoperiod and SuMN-THx was not significant (F(1, 24) = 0.75, p = 0.394) reflecting that THx rats maintain a circadian feeding rhythm with peak food intake during the dark period similar to Ctr rats. Post hoc analysis of pair-wise comparisons using Bonferroni-corrected α-level further revealed that THx rats as compared to Ctr rats consumed 75% more food (4.2 ± 0.31 g/THx rat vs. 2.4 ± 0.36g/Ctr rat, p < 0.001) during the light period, while no significant increase in food intake was detected in THx rats versus Ctr rats (16.1 ± 0.67g/THx rat vs. 15.4 ± 0.78 g/Ctr rat, p = 0.523) during the dark period (shown in Fig. 6a3). Of their total daily food consumption, THx rats consumed 21% during the light and 79% during the dark periods, whereas Ctr rats consumed 14% during the light and 86% during the dark photoperiod (p = 0.001). Thus, compared with controls, SuMN-THx treatment increased the daily food consumption specifically during the light period (normal fasting period) (by 75%) while maintaining the peak food intake during the dark period of the day.Fig. 6.Effect of THx at the SuMN area on circadian feeding rhythm, food preference, hunger/satiety balance, and responses to systemic ghrelin and GLP-1 agonist administration in f-HFDr-RC rats (Study 2Bi-v). f-HFDr-RC rats were locally inoculated at the medial SuMN area with AAV5-Th-shRNA-GFP to knockdown Th gene expression (THx) or control AAV5-Scr-shRNA-GFP (Ctr). SuMN-THx increased food consumption during the light (fasting/resting) period of the day (a1–3). Regular chow consumption was monitored throughout light and dark periods of the day in THx and Ctr rats over two consecutive days. THx rats, like Ctr rats, maintained a circadian feeding rhythm with peak food intake during the dark photoperiod (a1). Nonetheless, THx rats ate 14% more food per day than Ctr rats (p = 0.014) (a2) which was consumed particularly during the light period. THx rats as compared to Ctr rats consumed 75% more food (p < 0.001) during the light period with no difference in food consumption during the dark period (p = 0.523) (a3). SuMN-THx did not alter food preference of HFD over RC, but increased HFD consumption (b1–2). THx rats and Ctr rats were exposed to RC and HFD simultaneously, and food consumption was measured daily. In both scenarios where the order of HFD and RD given was switched, THx and Ctr rats showed the same strong food preference for HFD over RC. While no difference in RC consumption was observed between THx rats and Ctr rats, THx rats consumed more HFD than Ctr rats by 17% (p = 0.002) over 4 days when HFD was given as novel add-on diet (b1) and by 16% (p = 0.0226) over 7 days when RC was given as novel add-on diet (b2). SuMN-THx reduced fasting-primed refeeding (c). Following overnight fasting (16 h), THx and Ctr rats were refed RC (starting at ZT4) during the light period, and cumulative RC consumption was measured at 0.5 h, 1 h, 2 h, and 4 h post-refeeding time. After an initial peak in food intake at 0–0.5 h by both THx rats and Ctr rats, there was a 67% reduction in food consumption in THx rats versus Ctr rats at 0.5–1 h (p < 0.001) and at 1–2 h (p = 0.027). SuMN-THx reduced ghrelin-stimulated feeding (d). THx rats and Ctr rats were injected with ghrelin or vehicle during their natural fasting period of the day (at ZT5.5–6, light period), and cumulative RC consumption was measured at 1 h and 6 h after drug administration. At 0–1 h, there was a non-significant elevation in basal food consumption in Veh/THx versus Veh/Ctr rats (p = 0.16). Ghrelin significantly increased food consumption in Ctr and THx rats by 200% (p < 0.001) and 84% (p < 0.001), respectively, versus their respective vehicle controls. At 1–6 h, ghrelin significantly decreased food consumption by 75% (p = 0.002) in THx rats versus vehicle-treated THx rats, and by 55% (ns, p = 0.141) versus ghrelin-treated Ctr rats. SuMN-THx did not alter Ex-4-induced satiety (e). THx rats and Ctr rats were injected with Ex-4 or vehicle at the beginning of the dark period of the day (ZT14–14.5). Cumulative RC consumption was measured under a red light at 1 h, 3 h, and 16 h after the drug administration during the natural feeding period of the day. Ex-4 inhibited food consumption in THx rats to the same magnitude as in Ctr rats at 0–1 h and 1–3 h, and the Ex-4 effect subsided at 3–16 h in both Ctr and THx rats. At 0–1 h, as compared to vehicle, Ex-4 reduced food consumption by 69% in THx rats (p < 0.001) and by 70% in Ctr rats (p < 0.001), and at 3-1 h by 67% in THx rats (p < 0.001) and by 70% in Ctr rats (p < 0.001). *p < 0.05, two-tailed Student’s t test. β indicates an observed p value is less than Bonferroni-corrected α-level.Study 2Bii: SuMN-THx Increases HFD Consumption with No Change in HFD Preference to RCTHx and Ctr rats were exposed to RC and HFD simultaneously, and food consumption was measured daily. First, rats maintained on RC were subsequently additionally provided HFD as an add-on diet. The consumption of RC and HFD was then monitored the ensuing 4 days. There was a significant main effect of diet (two-way repeated measures ANOVA, F(1, 23) = 1,558.78, p < 0.001) reflecting that both groups of rats preferred HFD versus RC (HFD: 76.31 ± 1.81 g; RC: 1.43 ± 0.49 g). However, there was a significant main effect of THx (F(1, 23) = 9.98, p = 0.004) and a significant interaction between diet and THx (F(1,23) = 10.36, p = 0.004). Post hoc analysis revealed that THx rats consumed 17% more HFD than Ctr rats (82.3 ± 2.4 g/THx rat vs. 70.33 ± 2.71 g/Ctr rat, p = 0.002) over the 4-day study period (shown in Fig. 6b1). Second, rats maintained on HFD were subsequently additionally given RC as an add-on dietary choice and consumption of RC and HFD was then measured over the next 7 days. Once again, there was a significant main effect of diet (F(1, 23) = 509.68, p < 0.001) such that rats consumed predominantly HFD with limited RD consumption due to its novel exposure (HFD: 101.7 ± 3.11 g; RC: 13.6 ± 1.84 g) (shown in Fig. 6b2). No significant main effect of THx (F(1, 23) = 3.21, p = 0.086) was detected, but the interaction between diet and THx was significant (F(1, 23) = 5.67, p = 0.026). Post hoc analysis revealed that THx rats consumed 16% more HFD than Ctr rats (109.3 ± 4.12 g/THx rat vs. 94.1 ± 4.65 g/Ctr rat, p = 0.0226) over the 7-day study period (shown in Fig. 6b2). The RC consumption was largely confined to the first 2 days of novel exposure of RC, and no difference in RC consumption between THx rats and Ctr rats was found (11.86 ± 2.44 g/THx rat vs. 15.25 ± 2.75 g/Ctr rat, p = 0.814). Thus, when given a choice to consume HFD and RC, THx and Ctr rats similarly strongly preferred HFD over RC. However, in both experimental scenarios, THx rats consistently consumed more HFD (by 16∼17%) than Ctr rats.Study 2Biii: SuMN-THx Reduces Overnight Fasting-Enhanced, Morning Refeeding Food ConsumptionFollowing overnight fasting (16 h), THx and Ctr rats were refed RC (starting at ZT4) during the light/fasting period of the day, and cumulative RC consumption was measured at 0.5 h, 1 h, 2 h, and 4 h post-refeeding time. There was a significant main effect of time (F(3, 72) = 53.91, p < 0.001) reflecting an initial peak of food intake (3.3 ± 0.2 g at 0–0.5 h) followed by a gradual reduction over the next 4 h (0.24 ± 0.12 g at 2–4 h). There was a significant main effect of THx (F(1, 24) = 410.15, p < 0.001) and a significant interaction between time and THx (F(3, 72) = 0.041). Post hoc analysis (pair-wise comparisons using Bonferroni-corrected α level) revealed a 67% reduction in food consumption in THx rats versus Ctr rats at 0.5–1 h (0.65 ± 0.21 g/THx rat vs. 1.97 ± 0.24 g/Ctr rat, p < 0.001) and at 1–2 h (0.44 ± 0.25 g/THx rat vs. 1.33 ± 0.29 g/Ctr rat, p = 0.027) (shown in Fig. 6c). Thus, while SuMN-THx increases food consumption (both of RC and HFD) during light/fasting period of the day under free living, ad libitum conditions, when feeding is examined in a fasting-refeeding paradigm SuMN-THx actually reduces the refeeding response, suggestive of ghrelin resistance.Study 2Biv: Ghrelin Reduces Day-Time Food Intake in THx Rats despite Undiminished Ghrelin-Induced Acute HyperphagiaTo further investigate the above-described SuMN-THx decrease of the refeeding response, suggestive of ghrelin resistance, a study was conducted to measure feeding response to systemic ghrelin administration (shown in Fig. 6d). THx and Ctr rats were injected with ghrelin or vehicle during their natural fasting period of the day (at ZT5.5−6 light period), and cumulative RC consumption were measured at 1 h and 6 h after drug administration. There was a significant main effect of ghrelin (F(1, 23) = 7.202, p = 0.013) and a significant main effect of time (F(1, 23) = 35.453, p < 0.001). The interaction between ghrelin and time was also significant (F(1, 23) = 61.027, p < 0.001). Ghrelin induced an increase in food consumption at 0–1 h and the effect was terminated at 1–6 h. There was no significant main effect of THx (F(1, 23) = 0.049, p = 0.826), but the interaction between THx and ghrelin was significant (F(1, 23) = 5.58, p = 0.027). Post hoc analysis (pair-wise comparisons using Bonferroni-corrected α-level) further revealed that at 0–1 h, ghrelin stimulated food intake in Ctr rats by 200% (2.7 ± 0.184 g vs. 0.9 ± 0.231 g, p < 0.001) but by only 84% in THx rats (2.45 ± 0.15 g vs. 1.33 ± 0.189 g, p < 0.001), mainly due to an elevated basal level of food intake in THx rats (Veh/THx rats: 1.33 ± 0.189 g vs. Veh/Ctr rats: 0.9 ± 0.231 g, p = 0.16). The 38% decrease in ghrelin-induced acute hyperphagia level in THx rats versus Ctr rats however did not reach significance (p = 0.12). At later 1–6 h, no difference in food consumption was detected between ghrelin/Ctr and vehicle/Ctr groups, however, SuMN-THx treatment actually reduced the feeding response to ghrelin by 75% compared with vehicle-injected THx rats (0.33 ± 0.169 g vs. 1.35 ± 0.266 g, p = 0.002). Thus, in conjunction with (and possibly due to) a higher baseline level of feeding among SuMN-THx rats over the 6 h test period, such animals were overall less sensitive to ghrelin-induced food intake than Ctr animals.Study 2Bv: SuMN-THx Does Not Alter GLP-1 Agonist (Ex-4)-Induced HypophagiaTHx and Ctr rats were injected with Ex-4 or vehicle at the beginning of the dark period (initiation of normal feeding time) (ZT14-14.5). Cumulative RC consumption was subsequently measured under red light at 1 h, 3 h, and 16 h after drug administration. There was a significant main effect of Ex-4 (F(1, 24) = 12.52, p = 0.002) and a significant main effect of time (F(2, 48) = 242.83, p < 0.001). The interaction between Ex-4 and time was also significant (F(2, 48) = 10.21, p < 0.001). Ex-4 treatment induced hypophagia in rats at early 0–1 h and 1–3 h but not at later 3–16 h versus vehicle. There was no significant main effect of THx (F(1, 24) = 1.65, p = 0.211), nor its interactions with Ex-4 (F(1, 24) = 0.25, p = 0.621) or with time (F(2, 48) = 0.09, p = 0.918). Post hoc analysis revealed that Ex-4 inhibited food consumption in THx rats to the same magnitude as in Ctr rats at 0–1 h by 69% in THx rats (Ex-4/THx rat: 0.79 ± 0.266 g vs. Veh/THx rat: 2.57 ± 0.315 g, p < 0.001) and by 70% in Ctr rats (Ex-4/Ctr rat: 0.99 ± 0.31 g vs. Veh/Ctr rat: 3.29 ± 0.368 g, p < 0.001), as well as at 3-1 h by 67% in THx rats (Ex-4/THx rat: 0.96 ± 0.236 g vs. Veh/THx rat: 2.9 ± 0.319 g, p < 0.001) and by 70% in Ctr rats (Ex-4/Ctr rat: 1.08 ± 0.276 g vs. Veh/Ctr rat: 3.59 ± 0.373 g, p < 0.001) (shown in Fig. 6e).Study 3: Metabolic Effects of SuMN-THx in Female HFDr Rats Are Enhanced in Those Fed HFD (f-HFDr-HFD)f-HFDr-HFD rats were inoculated with AAV5-Th-shRNA-GFP (THx) or AAV-Scr-shRNA-GFP (Ctr). Compared with control, THx increased GTT glucose AUC (16%, p = 0.043) (shown in Fig. 7a, b) and insulin AUC (62%, p = 0.001) (shown in Fig. 7c, d) resulting in a reduction in Belfiore ISI (39%, p < 0.0001) (shown in Fig. 7e). Both the basal glucose (p = 0.006) (shown in Fig. 7a, insert) and basal insulin (p = 0.04) (shown in Fig. 8c) were elevated in THx rats compared with Ctr rats. For both GTT glucose and GTT insulin, there was a main effect of time (Fglucose (4, 160) = 230.24, p < 0.001; Finsulin (4, 160) = 26.83, p < 0.001) and a main effect of THx (Fglucose (1, 40) = 5.8, p = 0.021; Finsulin(1, 40) = 112, p < 0.001) but no significant interaction between time and THx (Fglucose (4, 160) = 2.005, p = 0.133; Finsulin (4, 160) = 0.836, p = 0.504). Such THx treatment increased body weight gain by 23% (67 ± 3.4 g gain vs. 55 ± 2.2 g gain in Ctr rats, p = 0.006) at 37 days after SuMN-THx treatment (shown in Fig. 7f, g) and increased food consumption by 7% (p = 0.024) (shown in Fig. 7h, i) and feed efficiency by 14% (p = 0.008) (shown in Fig. 7j) relative to Ctr. For body weight and food consumption, there was a significant effect of time and no significant effect of THx, and the interaction between time and THx was significant (body weight [Ftime (11, 440) = 372.62, p < 0.001; FTHx (1, 40) = 2.36, p = 0.132; Ftime*THx (11, 440) = 6.7, p < 0.001]; food [Ftime (10, 440) = 3,435.04, p < 0.001; FTHx (1, 40) = 1.94, p = 0.171; Ftime*THx (11, 440) = 6.7, p < 0.001]). A reduction in FAO in pWAT was detected in THx versus Ctr rats (basal FAO by 29%, p < 0.0001; endogenous FAO by 39%, p < 0.0001; exogenous FAO by 53%, p = 0.029) (shown in Fig. 7k). The plasma triglyceride level was increased by 15% (p = 0.033 ƚ) (shown in Fig. 7l).Fig. 7.THx at the SuMN area induces glucose intolerance/IR and increases body weight gain in f-HFDr-HFD rats (Study 3). f-HFDr-HFD rats were locally inoculated at the medial SuMN area with AAV5-Th-shRNA-GFP to knockdown Th gene expression (THx) or control AAV5-Scr-shRNA-GFP (Ctr). Compared with Ctr, THx treatment increased GTT glucose AUC (16%, p = 0.043) (a, b) and insulin AUC (62%, p = 0.001) (c, d) resulting in a reduction in Belfiore ISI (39%, p < 0.0001) (e). THx treatment increased body weight gain by 23% (p = 0.06) (f, g) and increased food consumption by 7% (p = 0.024) (h, i) and feed efficiency by 14% (p = 0.008) (j) relative to Ctr. Compared with Ctr, THx treatment reduced FAO in pWAT (8 mg) (basal FAO by 29%, p < 0.0001; endogenous FAO [Endo-FAO] by 39%, p < 0.0001; exogenous FAO [Exo-FAO] by 53%, p = 0.029) (k). The plasma triglyceride level was increased by 15% in THx rats relative to Ctr rats (p = 0.033 Ɨ) (l). *p < 0.05, two-tailed Student’s t test. Ɨp < 0.05, one-tailed Student’s t test.Fig. 8.THx at the SuMN area induces glucose intolerance and increases body weight gain in m-HFDr-RC rats (Study 4). m-HFDr-RC rats were locally inoculated at the medial SuMN area with AAV5-Th-shRNA-GFP to knockdown Th gene expression (THx) or control AAV5-Scr-shRNA-GFP (Ctr). Compared with Ctr, THx treatment increased GTT glucose AUC by 38% (p = 0.014) (a, b) with no change in insulin AUC (c, d) resulting in 28% reduction in Belfiore ISI (p = 0.015) (e). The THx rats exhibited an increase in body weight gain by 12% (p = 0.02) (f, g) and 7% increase in food consumption (p = 0.04) (h, i) relative to Ctr rats. There was no significant increase in feed efficiency (p = 0.137) (j) in THx rats compared to Ctr rats. *p < 0.05, two-tailed Student’s t test.Study 4: SuMN-THx Induces Glucose Intolerance in m-HFDr-RC RatsTo evaluate the influence of gender in the metabolic effects of SuMN-THx, m-HFDr-RC rats were locally inoculated at the medial SuMN area with AAV5-Th-shRNA-GFP to knockdown Th gene expression (THx) or control AAV5-Scr-shRNA-GFP (Ctr) as described above. Compared with control, THx increased GTT glucose AUC by 38% (p = 0.014) (shown in Fig. 8a, b) with no change in insulin AUC (shown in Fig. 8c, d) resulting in 28% reduction in Belfiore ISI (p = 0.015) indicating an increase in IR (shown in Fig. 8e). The basal blood glucose was elevated in THx rats versus Ctr rats (p = 0.031) (shown in Fig. 8a, insert). The GTT glucose response exhibited significant main effects of time, THx, and their interactions (Ftime (4,88) = 100.57, p < 0.001; FTHx (1, 22) = 288.75, p < 0.001; Ftime*THx (4, 88) = 4.55, p = 0.002), while GTT insulin only showed a significant main effect of time (F(4, 88) = 6.85, p < 0.001) with no significant main effect of THx (F(1, 22) = 0.74, p = 0.788) or THx and time interaction (F(4, 88) = 1.08, p = 0.37). SuMN-THx increased body weight gain by 12% (287 ± 9.1 g gain vs. 257 ± 7.7 g gain in Ctr rats, p = 0.02) (shown in Fig. 8f, g) and food consumption by 7% (p = 0.04) (shown in Fig. 8h, i) relative to control rats. However, such treatment did not result in any significant increase in feed efficiency (p = 0.137) (shown in Fig. 8j) in THx rats compared to Ctr rats. There was a significant main effect of time for body weight (F(10, 220) = 1,515.71, p < 0.001) and for food consumption (F(9, 198) = 4,138.42, p < 0.001). The main effect of THx was not significant for body weight (F(1, 22) = 0.34, p = 0.566) but was significant for food (F(1, 22) = 4.48, p = 0.046). The interaction between time and THx was significant for both body weight (F(10, 220) = 4.11, p < 0.001) and food consumption (F(9, 198) = 5.57, p < 0.001).Study 5: Long-Lasting Effect of SuMN-THx on Body Weight and Glucose Metabolism in f-HFDs-RC RatsTo examine any potential long-lasting metabolic effect of TH knockdown at the SuMN area, f-HFDs-RC rats were inoculated at medial SuMN area with AAV5-Th-shRNA-GFP to knockdown Th gene expression (THx) or control AAV5-Scr-shRNA-GFP (Ctr) and maintained on regular chow for 1 year. Compared with Ctr, SuMN-THx increased % body weight gain by 96% in THx rats versus 65% increase in Ctr rats at 12 months (p = 0.012) post-virus inoculation (shown in Fig. 9a, b). There was a significant main effects of time (F(2, 32) = 196.47, p < 0.001) and THx (F(1, 16) = 7.79, p = 0.013), and the interaction between time and THx was also significant (F(2, 32) = 6.47, p = 0.004). At 12 months post-virus inoculation, compared with Ctr rats, THx rats exhibited elevated levels of fasting blood glucose (p = 0.015) (shown in Fig. 9c) and plasma insulin (p = 0.031) (shown in Fig. 9d) resulting in an increase in HOMA-IR (120%, p = 0.027) indicating IR (shown in Fig. 9e). The extent of TH knockdown at the SuMN area remained stable at 1 year post-virus inoculation as verified by TH immunohistochemistry (data not shown).Fig. 9.Long-lasting effect of THx at the SuMN area on body weight gain and glucose metabolism in f-HFDs-RC rats (Study 5). f-HFDs-RC rats were locally inoculated at the medial SuMN area with AAV5-Th-shRNA to knockdown Th gene expression (THx) or control AAV-Scr-shRNA-GFP (Ctr). THx treatment increased % body weight gain by 1.5-fold (96% vs. 65% increase in Ctr, p = 0.012) after 12 months post-virus inoculation (a, b). Compared with Ctr, THx treatment increased the plasma levels of fasting glucose (18%, p = 0.015) (c) and insulin (78%, p = 0.031) (d) resulting in an increase in HOMA-IR (120%, p = 0.027) (e) indicating IR. *p < 0.05, two-tailed Student’s t test.DiscussionThe present study is the first to demonstrate that TH knockdown of DA neurons at the SuMN area which includes DA neurons densely clustered within the medial SuMN and sparsely scattered DA neurons in the adjacent PH immediately dorsal to the medial SuMN is critically involved in the regulation of whole-body energy balance. Chronic (>50 days) TH knockdown of medial SuMN area dopaminergic neurons leads to development of the obese, glucose-intolerant condition in an animal model extremely resistant to development of such a condition – the lean young female SD rat selected to be refractory to the obesogenic effects of HFD feeding and held on low-fat regular chow diet and long (14 h light) daily photoperiods [30‒32]. Such treatment resulted in a substantial (42%) increase in body weight gain that was similar to the observed increase in body fat store level (42%). This body composition change occurred without any change in fasting or GTT insulin level; however, it was coupled to a concurrent 25% increase in feed efficiency and 15% increase in cumulative food consumption.The increased feed efficiency that coincided with the increased fat store accretion was associated with 1.4-fold increase in circadian fasting time liver mRNA expression of Ppargc1β, a transcription factor for stimulation of lipogenic enzyme transcription during the subsequent feeding period of the day, though liver mRNA levels of the other lipogenic factors Srebf1, Mtor, and Fasn that peak during feeding were unaffected at this fasting period of the day. Moreover, the increased feed efficiency was coupled to a substantial (∼55%) reduction in the capacity of the adipose tissue to oxidize lipid. That is, following treatment, lipid in adipose tissue was less metabolized to energy (ATP) and rather stored than released into the circulation as FFAs (no increase in plasma FFAs in SuMN TH knockdown vs. control rats). The increase in body fat and feed efficiency and decrease in adipose FFA oxidation rate observed with SuMN DA neuronal TH knockdown is consistent with, and importantly may explain in part, previous studies, wherein systemic or intracerebroventricular DA agonist treatment reduced obesity, increased protein turnover (a major sink for calorie consumption), and increased white adipose FFA oxidation rate in MS animals [37‒40]. Furthermore, blockade of the downstream effect of such endogenous CNS DA activity in lean, insulin-sensitive animals that normally exhibit low noradrenergic input activity at the VMH (thereby increasing noradrenergic activity at the VMH) actually not only induced obesity [41] but converted brown fat to white [42], a scenario consistent with low CNS DA action facilitating whole-body lipid storage. This finding in turn is consistent with published observations that hypothalamic bromocriptine (a potent DA D2 receptor agonist) administration activated brown fat thermogenesis [43]. The present findings suggest that SuMN dopaminergic neurons are operative (via connections with the SCN and other centers) in manifesting these multiple observations of CNS DA regulation of body composition and white adipose lipid metabolism. Furthermore, the present findings are consistent with previous studies wherein (a) SuMN circadian efferent DA release contributes to the circadian peak of dopaminergic input to the SCN in healthy animals, an SCN stimulus critical to the maintenance of the lean, insulin-sensitive state, and (b) daily intracranial administration to MS animals of DA directly to the SCN at the time of day of its natural circadian peak reduced MS [10].Concurrent with the SuMN DA neuronal TH knockdown effect to increase body weight and fat in this animal model, such treatment also induced glucose intolerance that was coupled to increases in hepatic mRNA expression of the two key rate-limiting gluconeogenic enzymes, glucose 6 phosphatase and phosphoenolpyruvate carboxy kinase (thus contributing to glucose intolerance) without change in the plasma insulin level suggesting hepatic IR. Inasmuch as the SuMN DA neurons project directly and indirectly to the SCN, this SuMN-THx glucose metabolism response is consistent with previous observations that the SCN is a critical modulator of both glucose tolerance [10, 11, 16, 41] and hepatic gluconeogenesis [44]. The present findings offer insights into dopaminergic input pathways controlling/modifying this SCN output glucose regulatory system. Additionally, chronic SuMN-THx treatment resulted in the increased mRNA expression for two major hepatic pro-inflammatory proteins, Rela, the p65 protein involved in NF-κB activation, a pro-inflammatory transcription factor in liver, and Ccl2/Mcp1, a potent pro-inflammatory protein within liver and the immune system. This is relevant since hepatic inflammation is well known to potentiate hepatic gluconeogenesis, lipogenesis, and IR and can eventually lead to fatty liver [45, 46]. In this context, SuMN dopaminergic neuronal TH knockdown did also increase hepatic Ppargc1β, a lipogenic transcription factor, as well as plasma triglyceride levels (by 86%), coupled with a trend toward increased liver lipid content (without alteration of key liver lipid oxidation transcription factor mRNA levels). This intervention-induced obese-increased hepatic gluconeogenic/lipogenic phenotype is also consistent with leptin resistance and the observed hyperleptinemia (46% increase in plasma leptin) (see Fig. 5a) in these treated animals, especially in the face of increased food consumption over the test period (see Fig. 2c, d). Finally, since SuMN DA neurons project strongly to the SCN area and DA depletion at the SCN area has been reported to induce MS, a preliminary inspection of circadian components regulating metabolism was undertaken and it was observed that SuMN-THx treatment disrupted the diurnal (sleep) phase plasma corticosterone level (68% decrease) and liver mRNA levels for circadian genes, Clock and Per1 (1.4- and 1.8-fold increase, respectively). Although full 24-h profiles for plasma corticosterone level and liver circadian clock genes were not obtained in this study, these initial findings suggest that a shift in the circadian organization of metabolism had occurred potentiating the obese, glucose-intolerant condition that is consistent with previous findings wherein a shift in the plasma corticosteroid rhythm and alterations in liver circadian clock gene expressions potentiated MS [32, 47, 48]. Moreover, such SuMN-THx treatment also altered the circadian feeding rhythm resulting in increased feeding exclusively in the fasting period of the day – a phenomenon repeatedly demonstrated to induce MS when experimentally imposed (see below). In this regard, it has been concluded from a compilation of a multitude of studies that CNS circadian regulation of peripheral circadian metabolic activities manifests the lean, glucose-tolerant or obese, glucose-intolerant condition as a function of the temporal interactions of the phases of multiple circadian CNS output signals (e.g., endocrine and autonomic) that act to organize circadian biology in peripheral tissues (e.g., liver) in a manner to either facilitate or block the obese, glucose-intolerant condition [32, 47, 49]. The present findings along with these previous studies suggest that the SuMN dopaminergic neurons influence this CNS clock circuit regulation of metabolism.In conjunction with the above-described metabolic changes, an increase in daily food consumption was observed following the SuMN-THx treatment. The cumulative increase in food consumption sampled during the test periods (over 35, 72, and 140 days in Study 3, 2A, and 4, respectively) can be expected to work in concert with the increased feed efficiency and lipid accretion biochemical pathways to increase body weight and fat. It appears that decreased SuMN dopaminergic efferent innervation coordinates feeding-independent and feeding-dependent physiological functions that interact to shift metabolism toward the obese, glucose-intolerant state. Low brain DA in the mesolimbic reward system (decreases in DA D2 receptors and/or DA release) is observed in obesity and MS [50‒53] resulting in food overconsumption presumably as a result of the weak (insufficient) DA reward signal to induce meal feeding cessation [54]. While the influence of chronic SuMN-THx to induce increased feeding was consistently observed in the present studies, the influence of such intervention upon striatal DA release and reward-seeking behavior was not assessed. Recent studies by Kesner et al. [55] provide evidence that acute activation of SuMN glutamate, but not SuMN dopaminergic neurons results in, via communication with the septum and septum to VTA circuit, a VTA neuronal DA release at the ventral striatum to influence/stimulate motivation for environmental interaction and reward-seeking behavior. Activation of AMPA glutamate receptors on SuMN neurons, which may be of a variety of types, stimulates DA release and reward effects at the ventral striatum [56]. Importantly, we have previously demonstrated that AMPA stimulation of SuMN neurons causes substantial DA release from such SuMN neurons at the hypothalamic SCN and MPOA [11], and SuMN DA neurons are known to project to the LS and MPOA [2, 12] – all key relay centers to the VTA and as such may influence VTA-striatal DA release [57, 58]. However, whether chronic knockdown of SuMN dopaminergic neurons may result via reprogramming of neuronal communication with the LS, MPOA, and/or SCN to influence the VTA DA release to the ventral striatum to modulate reward-seeking behavior remains to be investigated. Furthermore, this TH knockdown of SuMN DA neurons may act via their direct projections to the SCN, MPOA, and/or LS that in turn communicate with other hypothalamic feeding regulating centers such as the PVN, DMH, LH, and perifornical area [59] to increase feeding as observed herein. The LS that receives abundant dopaminergic projections from the SuMN is a key relay station for hippocampal function [4]. The TH knockdown at the SuMN area may reduce hippocampal function in feeding conditioning and learning processes [60, 61] to thereby increase food consumption. Moreover, as it regards SuMN regulation of feeding, SuMN neurons express both GLP-1 and ghrelin receptors, and thus this nucleus is capable of directly responding to hunger (ghrelin) and satiety (GLP-1) signaling regulation (i.e., ghrelin acutely induces feeding and GLP-1 inhibits feeding) [7, 8]. However, the present study findings suggest that SuMN-THx results in a “ghrelin-resistant” state as it relates to feeding, a circumstance that has previously been associated with the obese, insulin-resistant state [62], without affecting the hypophagia response to GLP-1 agonist. In total, the effect of SuMN-THx on feeding may be summarized as follows. Such treatment induces a general sustained chronic increase in feeding over several weeks whether of RC or HFD, but such increased feeding restricted to the normal fasting period of the day, and such increased basal feeding level linked to a subsequent diminished ghrelin stimulatory feeding response. The effect of SuMN-THx to induce hyperphagia specifically within the natural resting/fasting (light) period of the day in these nocturnal animals is intriguing and worthy of brief separate discussion.Daily feeding time is itself a zeitgeber for entraining peripheral fuel metabolism rhythms and for regulating metabolic status, independent of the SCN output entrainment of such functions [63‒69]. Nonetheless, circadian rhythms of peripheral fuel metabolism, particularly in the liver, are influenced by both timed feeding stimulus and neuroendocrine input signals that derive from the SCN network [70, 71]. Misalignment of the daily feeding and SCN-driven activity rhythms (feeding outside the normal activity cycle) is well established to induce MS in animals [72‒75] and humans [76‒81]; however, what causes an internal drive to eat outside the normal feeding cycle remains a mystery. The present study findings in conjunction with previous study results demonstrating SuMN dopaminergic circadian input activity to the SCN [11] suggest that SuMN dopaminergic activity somehow participates in the regulation and coordination of both SCN circadian organization and output activity as well as CNS centers that regulate the daily feeding rhythm to modulate peripheral fuel metabolism. In support of this postulate is the observation that SuMN-THx simultaneously disrupted normal daily feeding activity (inducing fasting period hyperphagia), plasma corticosterone level, and hepatic clock gene expression while inducing MS in animals extremely resistant to MS. The theoretical dimensions of these findings are worthy of serious further investigation. It should be noted that whole-body metabolic rate was not determined in these studies and as such the possibility remains that SuMN-THx may have reduced overall energy expenditure to facilitate weight gain, in addition to such treatment effects on feeding. Nonetheless, SuMN-THx increased % body fat to a greater extent than % weight gain, increased hepatic lipogenic machinery, and decreased adipose fatty acid metabolism, suggesting a neuroendocrine shift driving fat accrual and storage is a primary modulatory force for the observed weight gain.In addition to the above-described SuMN DA neuron TH knockdown studies in the female SD rat model, the impact of such treatment was also assessed in the male SD rat. The male SD rat, in stark contrast to the MS-resistant female, develops IR at a very early age which worsens rapidly with time concurrent with marked weight gain over the first 6 months of life [30, 31]. Nonetheless, such SuMN-THx treatment in such male SD rats fed on regular chow diet resulted in further impaired glucose tolerance that was also accompanied by worsening IR (highlighting a more exaggerated glucose intolerance response than in the females). This response to SuMN-THx treatment was associated with weight gain (12% over 140 days), albeit less than in the female (42% over 72 days), likely due to the existent underlying obesity at baseline over the test period in males (compare Fig. 2a, b vs. Fig. 8f, g). The composite of these female and male studies suggests that the SuMN-THx response is not gender specific and may be expressed across the continuum of metabolic control status from the lean, insulin-sensitive to obese, insulin-resistant states.To further explore the breadth of baseline fuel metabolism profile upon which the SuMN-THx may function to induce glucose intolerance, another study was conducted wherein the SuMN-THx intervention was imposed on young female SD rats resistant to the obesogenic effects of HFD feeding but maintained on HFD (instead of regular chow as in the initial female SD rat study). In this HFD feeding scenario, the MS-inducing effect of the SuMN-THx treatment was more pronounced relative to such treatment in the same animal model held on regular rodent chow. Under conditions of HFD feeding, SuMN-THx treatment induced glucose intolerance, like the situation under conditions of regular chow feeding; however, such treatment under HFD feeding conditions also further induced an increase in fasting plasma insulin levels, GTT hyperinsulinemia, and frank IR (see Fig. 7a–e vs. Fig. 3 for comparison). Moreover, while food consumption and feed efficiency increase and a decrease in white adipose FFA oxidation rate were evident in SuMN-THx treatment of HFD and regular chow fed animals, body weight gain was much more rapid (time to significant increase at 30 days vs. 50 days) and to a greater extent (23% increase at 35 days vs. no change at 35 days) in SuMN-THx animals held on HFD versus regular chow, respectively. It is well established from multiple studies that chronic HFD feeding inhibits DA function in several brain centers, including the SuMN [10], by reducing DA synthesis and/or DA receptor binding [11, 50, 51, 82‒84]. This HFD effect may be one of several HFD factors contributing to the exaggerated MS response to SuMN DA neuronal TH knockdown in the present study by producing a more global depressed CNS dopaminergic activity background upon which the SuMN DA neuronal TH knockdown occurs. The present study findings in concert with these previous studies of HFD influence on brain dopaminergic activity suggest however that the SuMN itself is an important site for HFD-induced reduction in brain dopaminergic activity that leads to MS. Consequently, in composite these studies highlight a pathophysiological scenario where chronic HFD both triggers and amplifies the MS-inducing effect of decreased SuMN dopaminergic neuronal activity.Finally, a long-term study (1 year) of SuMN-THx metabolic response in young lean female HFD-sensitive SD rats but held on regular chow indicated that such SuMN treatment resulted in a significant increase in % body weight gain (47%), fasting blood glucose (18%), and plasma insulin (78%), and HOMA-IR (120%) levels versus controls. These findings suggest that the MS influences of SuMN-THx treatment are not able to be countered by an endogenous metabolic setpoint control to correct the metabolic disruption even after 1 year (nearly half the lifetime of the SD rat).The SuMN DA neurons project directly and indirectly to multiple areas of the limbic reward system (e.g., hippocampus, VTA, LS) and to hypothalamic metabolic relevant centers (e.g., MPOA, DMH, SCN) [2, 11, 12, 14]. The SuMN DA neurons that project to the clock SCN region are of particular importance with respect to peripheral energy metabolism. Seasonal changes in metabolic status (i.e., summer lean, insulin-sensitive vs. winter obese, insulin-resistant conditions) among animals in the wild are correlated with and driven by the seasonal wax and wane, respectively, in circadian peak DA activity at the SCN region [11, 32]. Reduction of circadian DA activity at the clock SCN region of lean insulin-sensitive animals by DA neuron-specific lesion or HFD feeding induces obesity and IR/glucose intolerance [16, 41]. In contrast, restoration of the diminished circadian DA activities in obese insulin-resistant animals by direct timed daily micro-infusion of DA directly at the SCN region or timed daily administration of DA receptor agonist at the time of day of the circadian peak in SCN dopaminergic input activity improves glucose intolerance and IR [10, 11, 17]. Since the circadian activity peak of TH-positive DA neurons within SuMN provides a significant circadian peak of dopaminergic input to the SCN region, SuMN-THx should reduce the circadian peak DA activity at this region. Moreover, other metabolism control centers are known to receive dopaminergic projections from the SuMN (e.g., the LS, MPOA, DMH) [2, 12, 14], and all communicate strongly with the SCN. Therefore, the SuMN DA neurons could via providing circadian dopaminergic input directly to the SCN region and/or via relay centers to the SCN modulate its function to regulate peripheral energy metabolism. Moreover, the SuMN communicates with the nucleus tractus solitarius, a first-order recipient of primary afferents from the gut [85], wherein dietary constituents from the gut are known to regulate brain DA synthesis [84]. Specifically, gastrointestinal saturated fat has been demonstrated to inhibit brain DA synthesis via parasympathetic connections [84, 86, 87]. Although speculative, available evidence clearly suggests the possibility that gastrointestinal fat from HFD feeding may reduce SuMN DA synthesis that in turn leads to MS potentiation in part via reducing DA input to the hypothalamic SCN and limbic system. Consequently, the SuMN is a unique neuronal integration and coordination center for both behavioral and biochemical regulation of peripheral fuel metabolism.The limitations of these composite studies include the lack assessment of SuMN-THx on whole-body metabolic rate (that may have influenced SuMN-THx effect on body weight) and on post-mealtime lipid synthesis rate. However, now having these study data in hand, future studies of whole-body metabolic rate and post-meal lipogenic activity are warranted.In summary, chronic TH knockdown of SuMN area DA neurons via site-specific inoculation of a virus-mediated shRNA to knockdown TH (the rate-limiting enzyme for DA synthesis) gene expression in animals selected to be extremely resistant to development of MS at baseline and maintained on regular rodent chow induced multiple pathophysiological and phenotypic aspects of MS. Features of this SuMN TH knockdown-induced metabolic shift include increased body weight gain, body fat mass, food consumption, feed efficiency, and glucose intolerance concurrent with increased plasma leptin and triglyceride levels, increased hepatic mRNA expression levels of key gluconeogenic, lipogenic, and pro-inflammatory genes, and a reduction in white adipose FAO rate. Such treatment also simultaneously altered the circadian phase of daily feeding with increased fasting period feeding, plasma corticosterone level, and the expression of circadian genes in the liver that are involved in the regulation of hepatic fuel metabolism, strongly suggestive of an impact on central pacemaker-biological clock regulation of metabolism. This MS-inducing effect of SuMN DA neuronal TH knockdown was further enhanced under conditions of HFD feeding in this animal model system. The dopaminergic neurons at the SuMN area are thus a major integral functional component of a complex CNS system regulating peripheral fuel metabolism.Statement of EthicsThis study protocol was reviewed and approved by the Institutional Animal Care and Use Committee of VeroScience, LLC., approval number 2016-E0197.Conflict of Interest StatementThe authors are current or former employees of VeroScience LLC, a company that owns Cycloset®, a DA D2 receptor agonist, i.e., FDA-approved for the treatment of type 2 diabetes. The authors declare that they have no competing interests.Funding SourcesThis study was funded by VeroScience LLC of Tiverton, RI and S2 Therapeutics of Bristol, TN.Author ContributionsA.H.C. and Y.Z. designed the present studies and wrote the manuscript. Y.Z., M.E., and T.-H.T. performed the experiments. Y.Z. and C.S. analyzed the data.

Journal

Neuroendocrinology – Karger

Published: Jan 1, 2023

Keywords: Supramammillary nucleus; Dopamine; Glucose intolerance; Obesity; Feeding

Get 20M+ Full-Text Papers For Less Than $1.50/day. Start a 7-Day Trial for You or Your Team.

Learn More →

Tyrosine Hydroxylase Knockdown at the Hypothalamic Supramammillary Nucleus Area Induces Obesity and Glucose Intolerance

Tyrosine Hydroxylase Knockdown at the Hypothalamic Supramammillary Nucleus Area Induces Obesity and Glucose Intolerance

Get 20M+ Full-Text Papers For Less Than $1.50/day. Start a 7-Day Trial for You or Your Team.

Learn More →

Tyrosine Hydroxylase Knockdown at the Hypothalamic Supramammillary Nucleus Area Induces Obesity and Glucose Intolerance

Tyrosine Hydroxylase Knockdown at the Hypothalamic Supramammillary Nucleus Area Induces Obesity and Glucose Intolerance

References (96)

Abstract

Journal

Recommended Articles

There are no references for this article.

Our policy towards the use of cookies