We sought to determine a role for NPY overexpression in the dorsomedial hypothalamus (DMH) in obesity etiology using the rat model of adeno-associated virus (AAV)-mediated expression of NPY (AAVNPY) in the DMH.
Rats received bilateral DMH injections of AAVNPY or control vector and were fed on regular chow. Five-week postviral injection, half the rats from each group were switched to access to a high-fat diet for another 11 weeks. We examined variables including body weight, food intake, energy efficiency, meal patterns, glucose tolerance, fat mass, plasma insulin, plasma leptin, and hypothalamic gene expression.
Rats with DMH NPY overexpression had increased food intake and body weight and lowered metabolic efficiency. The hyperphagia was mediated through increased meal size during the dark. Although these rats had normal blood glucose, their plasma insulin levels were increased in both basal and glucose challenge conditions. While high-fat diet induced hyperphagia, obesity, and hyperinsulinemia, these effects were amplified in rats with DMH NPY overexpression. Arcuate Npy, agouti-related protein and proopiomelanocortin expression was appropriately regulated in response to positive energy balance. These results indicate that DMH NPY overexpression can cause hyperphagia and obesity and DMH NPY may have actions in glucose homeostasis.
The dorsomedial hypothalamus (DMH) plays an important role in maintaining energy homeostasis. Lesions of the DMH result in hypophagia, reduced body weight and linear growth . Disinhibition of neurons in the DMH provokes nonshivering thermogenesis and elevates core body temperature . Despite these observations, the neural mechanisms underlying these effects remain incompletely understood. Within the DMH, a number of neurotransmitters and/or receptors have been found, and their roles in controlling energy balance have been investigated [1, 3-5]. We have recently examined the role of the orexigenic peptide neuropeptide Y (NPY) in these actions [6, 7].
Within the hypothalamus, NPY-containing neurons are primarily identified in the arcuate nucleus (ARC) and the DMH [8, 9]. In contrast to the well-characterized actions of ARC NPY in energy balance control [10-12], the importance of DMH NPY in maintaining energy balance is just being unraveled. DMH Npy overexpression or induction has been found in certain rodent models with increased energy demands [8, 13, 14] and obesity [15-19]. Knockdown of NPY in the DMH via adeno-associated virus (AAV)-mediated RNAi ameliorates the hyperphagia, obesity, and diabetes of Otsuka Long-Evans Tokushima Fatty (OLETF) rats . NPY knockdown in the DMH of normally growing rats affects a number of aspects of energy balance control including food intake, energy expenditure, thermogenesis, adiposity, and physical activity . Overall, these findings suggest that DMH NPY acts as an important neuromodulator to modulate energy balance and dysregulation of DMH NPY causes disordered energy balance, leading to obesity and diabetes.
To ascertain a potential causal role for DMH NPY overexpression in these disorders, we have previously examined the effects of DMH NPY overexpression on food intake and body weight using the rat model of AAV-mediated expression of NPY in the DMH. We found that NPY overexpression in the DMH, particularly within and around the compact subregion, causes increased food intake and body weight and enhances high-fat diet-induced obesity . In this study, we sought to more completely characterize the effects of DMH NPY overexpression on food intake, body weight, adiposity and blood glucose using the same model.
Methods and Procedures
Male Sprague-Dawley rats were purchased from Charles River Laboratories and individually housed on a 12:12 h light-dark cycle (lights on at 0600h) in a temperature-controlled colony room (22–24°C) with ad libitum access to tap water and standard rodent chow, except where noted. All procedures were approved by the Institutional Animal Care and Use Committee at the Johns Hopkins University.
AAV-mediated expression of NPY in the DMH
As described previously , we generated a recombinant vector of AAV-mediated NPY expression (AAVNPY). The vector AAVGFP served as a control. We first determined the amount of viral particles injected for AAV-mediated NPY expression in the entire DMH including both compact and noncompact subregions via increasing the amount of vectors from the previous dose of 0.3 μl/site (∼1 × 109 particles/site) (6) to 0.5 μl/site (∼1.7 × 109 particles/site). After verification of AAV-mediated NPY overexpression in the DMH using standard in situ hybridization histochemistry , 24 male rats weighing 100-110 g were randomly assigned to bilateral DMH injections of AAVNPY or AAVGFP (n = 12 rats/group). As described previously , 0.5 μl/site of AAV vectors were bilaterally injected into the DMH with coordinates: 2.3 mm caudal to bregma, 0.4 mm lateral to midline and 7.6 mm ventral to skull surface. After injection, rats continued to have ad libitum access to a regular chow diet (RC: 15.8% fat, 65.6% carbohydrate, and 18.6% protein in kcal%, 3.37 kcal/g, Prolab RMH 1000, PMI Nutrition International, LLC, Brentwood, MO), designated as AAVNPY-RC and AAVGFP-RC. Five weeks postviral injection, half the rats from each group were switched to ad libitum access to a high-fat diet (HF: 60% fat, 20% carbohydrate, and 20% protein in kcal%; 5.2 kcal g−1; Research Diets; New Brunswick, NJ), designated as AAVNPY-HF and AAVGFP-HF, for 11 weeks. Body weights were measured daily and food intake was recorded weekly. Sixteen weeks postviral injection, rats were sacrificed in a 2-h fasted state during the light period. Blood glucose was determined with a FreeStyle glucometer (Abbott Laboratories, Abbott Park, IL). Trunk blood was taken for evaluation of leptin and insulin using a rat insulin and leptin radioimmunoassay kit, respectively (Millipore Corporation; Billerica, MA). Interscapular brown adipose tissue (BAT), epididymal white adipose tissue (WAT), and subcutaneous inguinal WAT were collected and weighed. Brains were saved for subsequent examination of hypothalamic gene expression using quantitative real time RT-PCR.
Analysis of meal patterns
An additional cohort of 12 male rats was used for this study. Rats received DMH injections of AAVNPY or AAVGFP (n = 6 rats/group) as described above. Two-weeks postviral injection, rats were transferred to individual test cages containing computerized feeding devices (MED Associates, Georgia, VT), which delivered 45-mg chow pellets as previously described . Rats had ad libitum access to pellets and water. After 7-days adaptation, data for 24-h food intake were collected and meal patterns were analyzed using a customized program. A meal was defined as the acquisition of at least five pellets preceded and followed by at least 15 min of no feeding . Meal size was defined as the number of pellets delivered during a meal.
Oral glucose tolerance test (OGTT)
An additional cohort of 10 male rats received DMH injections of AAVNPY or AAVGFP (n = 5 rats/group) as described above. Four-weeks postviral injection, OGTT was conducted as previously described . Following an overnight fast, rats were administered oral glucose (2 g kg−1) by gavage. Tail blood was sampled before and 15, 30, 45, 60, and 120 min after giving glucose. Blood glucose and plasma insulin concentrations were determined as described above.
Quantitative real-time RT-PCR
Brains at the levels of the DMH and the ARC were first sliced via a cryostat and then individual nuclei of the DMH and the ARC were punched out. Total RNA was extracted from each sample (punched hypothalamic nuclei, inguinal WAT, or interscapular BAT) by using Trizol reagent (Life Technologies, Grand Island, NY). Two-step quantitative real time RT-PCR was performed for gene expression determination as described previously . β-actin was used as an internal control for quantification of individual mRNA. A list of primer sets is shown in Table 1.
|Forward primer||Reverse primer|
|Uncoupling protein 1||5′-cgttccaggatccgagtcgcaga-3′||5′-tcagctcttgtcgccgggttttg-3′|
All values are presented as means ± SEM. Data were analyzed using the commercial software Statistica 7 (StatSoft, Tulsa, OK). Data for body weight and food intake were analyzed using two-way repeated measures analysis of variance (ANOVA) over the first 5-weeks postviral injection and three-way ANOVA with one repeated factor over the next 11 weeks. Data for meal patterns were analyzed using Student's t test (two-tailed). Data for cumulative intake, fat mass, blood glucose, plasma insulin and leptin, and mRNA expression levels for hypothalamic Npy, agouti-related protein (Agrp) and proopiomelanocortin (Pomc), inguinal WAT leptin and interscapular BAT uncoupling protein 1 (Ucp1) were analyzed using two-way ANOVA. All ANOVAs were followed by pairwise multiple Fisher's LSD comparisons. P < 0.05 was considered as a statistically significant difference.
AAV-mediated expression of NPY in the DMH
We verified that the vector AAVNPY infected neurons within the DMH including both compact and noncompact subregions, and produced robust Npy overexpression in the DMH area (Figure 1a). Consistent with the previous reports showing the long-lasting effect of AAV vectors on gene expression [6, 7], 16 weeks postviral injection, Npy mRNA levels in the DMH of AAVNPY rats remained significantly increased by 2.5 fold compared to control rats (AAVNPY-RC vs. AAVGFP-RC, Figure 5).
Effects of DMH NPY overexpression on body weight
Rats with DMH NPY overexpression gained significantly more body weight than control rats when maintained on regular chow over the first 5 weeks (P < 0.001, Figure 1b). The weight gain of AAVNPY-RC rats was increased by 10, 13, and 17% at 3, 4, and 5 weeks postviral injection respectively relative to AAVGFP-RC rats. Over the next 11 weeks, there were significant main effects of NPY overexpression (P < 0.001) and HF (P < 0.001) as well as a significant interaction between NPY overexpression and HF (P = 0.008, Figure 1b), indicating that access to HF resulted in greater body weight gain in AAVNPY rats than in AAVGFP rats. At sacrifice, AAVNPY-RC rats had sustained increases in body weight and gained 16% more weight than AAVGFP-RC rats (P = 0.037, Figure 1b). While HF caused a 15% increase in weight gain in control animals, HF resulted in a 33% increase in AAVNPY rats (P < 0.05, Figure 1b). Thus, AAVNPY rats on HF gained 34% more weight than control rats on HF or 54% more weight compared to control rats on RC (P < 0.001, Figure 1b).
Effects of DMH NPY overexpression on food intake
DMH NPY overexpression resulted in increased food intake in rats on both regular chow and HF diets. The chow intake of AAVNPY-RC rats was 15% more than that of AAVGFP-RC rats over the 16 weeks (P < 0.001, Figure 1c). When rats were challenged with HF for 11 weeks, there were significant main effects of NPY overexpression (P = 0.004) and HF (P < 0.001), but no significant interaction between NPY overexpression and HF (P = 0.272, Figure 1c). HF induced 29 and 39% increases in 11-week cumulative intake in AAVGFP (P = 0.008) and AAVNPY rats (P < 0.001), respectively. Although the increases in AAVNPY and AAVGFP rats did not significantly differ, AAVNPY-HF rats actually ate 23% more food than AAVGFP-HF rats (P = 0.007, Figure 1c).
Metabolic efficiency (i.e., energy intake divided by body weight gain) has been used for evaluation of metabolic rate . Analysis of metabolic efficiency revealed that DMH NPY overexpression resulted in decreased metabolic efficiency (from 47.5 ± 2.1 in AAVGFP-RC to 34.9 ± 2.7 in AAVNPY-RC, P = 0.001), indicating that DMH NPY overexpression lowered metabolic rate. Access to HF caused significant reductions of metabolic efficiency in both groups (from 47.5 ± 2.1 to 26.4 ± 1.7 in control rats, P < 0.001, and from 34.9 ± 2.7 to 27.3 ± 2.7 in AAVNPY rats, P = 0.025), but HF did not produce a further effect in AAVNPY rats (P = 0.774).
We next examined the effect of DMH NPY overexpression on meal patterns. AAVNPY rats consumed significantly more pelleted chow than AAVGFP rats during the dark period (P = 0.045), but the chow intake of AAVNPY and AAVGFP rats did not differ during the light period (P = 0.587, Figure 2a). Overall, AAVNPY rats had a significant increase in 24-h chow intake (P = 0.039, Figure 2a). Meal pattern analysis revealed that DMH NPY overexpression caused increased meal sizes in the dark (P = 0.043) and over the total 24 h (P = 0.047), but did not affect meal size during the light period (P = 0.492, Figure 2b). Meal numbers were not significantly altered in AAVNPY rats in the total, dark, or light period (Figure 2c).
Effects of DMH NPY overexpression on fat mass and plasma leptin levels
At sacrifice, we found a significant main effect of DMH NPY overexpression on inguinal WAT mass (P = 0.007), but not on epididymal WAT (P = 0.190) and interscapular BAT mass (P = 0.208, Figure 3a), indicating that DMH NPY overexpression produced an inguinal WAT-specific effect. Inguinal WAT mass was increased 42% in AAVNPY-RC rats compared to AAVGFP-RC rats (Figure 3a). HF induced significant increases in fat weights in all three fat depots (P < 0.001 in inguinal WAT, P = 0.001 in epididymal WAT, P = 0.005 in interscapular BAT, Figure 3a), but there were no significant interactions between NPY overexpression and HF in these fat depots (P > 0.05, Figure 3a). As compared to AAVGFP-HF rats, AAVNPY-HF rats accumulated significantly more fat mass in inguinal WAT, but not other two fat depots (Figure 3a).
Plasma leptin levels were not altered in rats with DMH NPY overexpression (p=0.191, Figure 3b) although the rats had increased body weight and fat mass. Consistent with HF-induced obesity, HF resulted in significant increases in plasma leptin levels in both AAVNPY and AAVGFP rats (P < 0.001), but the increases did not differ between the two groups (P > 0.05, Figure 3b).
Effects of DMH NPY overexpression on Ucp1 and leptin gene expression
We examined Ucp1 gene expression in interscapular BAT and leptin gene expression in inguinal WAT as DMH NPY has specific effects on these two fat pads . We found significantly decreased expression of Ucp1 in interscapular BAT of AAVNPY-RC rats relative to AAVGFP-RC rats (P < 0.05, Figure 3c). HF diet caused increased expression of Ucp1 in interscapular BAT in both AAVNPY and AAVGFP rats (P < 0.05), but the changes did not differ between the two groups (P > 0.05, Figure 3c).
DMH NPY overexpression resulted in downregulation of leptin expression in inguinal WAT (P < 0.05, Figure 3d) although this overexpression caused a significant increase in inguinal WAT mass (Figure 3a). While HF resulted in increased expression of leptin in inguinal WAT of AAVGFP rats (P < 0.05), NPY overexpression significantly reduced this increase, leading to the absence of a significant difference between AAVNPY-HF and AAVGFP-RC rats (P = 0.059, Figure 3d).
Effects of DMH NPY overexpression on glucose homeostasis
At sacrifice, blood glucose levels were normal in AAVNPY-RC rats (Figure 4a), but their plasma insulin levels were increased threefold compared to AAVGFP-RC rats (P < 0.05, Figure 4b), indicating that AAVNPY rats required more insulin to maintain normal blood glucose. Consistent with HF-induced hyperglycemia and hyperinsulinemia, access to HF caused significant increases in blood glucose and plasma insulin levels in both groups (P < 0.001 in glucose levels, P = 0.01 in insulin levels). Although HF-induced hyperglycemia did not differ between AAVNPY-HF and AAVGFP-HF rats (P > 0.05, Figure 4a), plasma insulin levels of AAVNPY-HF rats were significantly higher than those of AAVGFP-HF rats (P < 0.05, Figure 4b).
To further examine the effect of DMH NPY overexpression on glucose homeostasis, we conducted an OGTT in an additional cohort of AAVNPY rats with body weight matched with control rats (438 ± 16 g in AAVNPY vs. 410 ± 18 g in AAVGFP, P = 0.280) 4-weeks postviral injection. While fasting glucose levels were normal in AAVNPY rats (Figure 4c), their basal insulin levels were significantly elevated (Figure 4d). In response to oral glucose administration, blood glucose levels did not differ between AAVNPY and AAVGFP rats (Figure 4c), but plasma insulin levels were significantly elevated at 15, 30, and 45 min in AAVNPY rats compared to control rats (Figure 4d). These data indicate that AAVNPY rats required more insulin secretion to clear glucose, suggesting that DMH NPY overexpression causes insulin insensitivity.
Effects of DMH NPY overexpression on hypothalamic gene expression
As mentioned above, the vector AAVNPY-mediated expression of NPY in the DMH was long lasting. At sacrifice, quantitative real-time RT-PCR confirmed that Npy mRNA levels in the DMH of AAVNPY rats remained significantly increased (P < 0.001, Figure 5). Access to HF resulted in significantly decreased expression of Npy in the DMH in both groups (P < 0.05, Figure 5). Overall, there was a significant interaction between viral-mediated NPY overexpression and HF (P = 0.001), implying that HF caused a greater reduction of Npy expression in the DMH of AAVNPY rats. Even with this, the levels of Npy mRNA in the DMH of AAVNPY-HF rats remained significantly higher than those of AAVGFP-HF rats (P = 0.004, Figure 5).
Within the ARC, we found significant main effects of both viral-mediated NPY expression in the DMH and HF on Npy or Agrp mRNA levels (P < 0.05, Figure 5). Both Npy and Agrp mRNA levels in the ARC of AAVNPY-RC, AAVNPY-HF, or AAVGFP-HF rats were significantly lower than those of AAVGFP-RC rats (P < 0.05, Figure 5), but there were no significant interactions between viral-mediated NPY expression and HF on these two genes (P > 0.05). In contrast, Pomc mRNA levels in the ARC were not significantly altered by viral-mediated expression of NPY in the DMH (P = 0.633, Figure 5). Access to HF resulted in significantly increased expression of ARC Pomc in AAVGFP rats (P = 0.012), but not AAVNPY rats (P > 0.05, Figure 5).
We examined the effects of DMH NPY overexpression on food intake, body weight, adiposity, and blood glucose using the rat model of AAV-mediated NPY expression in the DMH. Rats with DMH NPY overexpression had increased food intake and body weight with decreased metabolic efficiency. These animals had increased inguinal WAT, decreased leptin expression in inguinal WAT, and lowered Ucp1expression in interscapular BAT. Although blood glucose was normal, plasma insulin levels were significantly elevated in both basal and glucose challenge conditions in rats with NPY overexpression. While HF induced hyperphagia, obesity, and hyperinsulinemia in control rats, access to HF amplified these effects in rats with NPY overexpression. Together, these results demonstrate that DMH NPY overexpression can produce hyperphagia and obesity and also provide additional evidence suggesting that DMH NPY plays a role in maintaining glucose homeostasis.
Previous reports have shown that alterations in DMH NPY result in a nocturnal and meal size-specific feeding effect. OLETF rats have disordered feeding behavior characterized by increased meal size that was proposed to contribute to their hyperphagia and obesity . Analysis of hypothalamic gene expression revealed elevated expression of Npy in the DMH of OLETF rats [19, 22], whereas DMH NPY knockdown completely normalizes meal size during the dark period in OLETF rats and significantly ameliorates their hyperphagia and obesity . Consistent with this view, the present study identified that DMH NPY overexpression caused increased meal size during the dark period. Thus, the results from both previous knockdown and present overexpression of NPY in the DMH of rats clearly establish a specific role for DMH NPY in the control of meal size during the dark period, and in this way, modulating overall food intake.
A role for DMH NPY in the regulation of adiposity, thermogenesis and energy expenditure has been implicated. DMH NPY knockdown promotes brown adipocyte development in inguinal WAT through sympathetic nervous system and causes elevated expression of thermogenic peptide UCP1 in both inguinal fat and interscapular BAT . This knockdown causes increased temperature in inguinal fat and interscapular BAT  and elevated energy expenditure . In support of this view, DMH NPY overexpression resulted in lowered metabolic efficiency (or metabolic rate) and decreased Ucp1expression in interscapular BAT. These data suggest that both feeding and metabolic effects may contribute to DMH NPY overexpression-induced increases in body weight and fat mass. Although HF-induced elevation of UCP1 in interscapular BAT is consistent with a proposed role for UCP1 in interscapular BAT in diet-induced thermogenesis , how HF interacts with DMH NPY to affect BAT thermogenesis or energy expenditure remains to be determined.
We found that DMH NPY overexpression caused increased inguinal WAT, but lowered leptin expression in this fat, i.e., increased fat mass did not lead to increased leptin expression or production. Although AAVNPY rats were heavier or had more fat than control rats, plasma leptin levels were not increased in AAVNPY rats. These data imply that DMH NPY overexpression may limit leptin products and the resulting reduction may also contribute to NPY overexpression-induced disorders. Nevertheless, the functional significance of altered leptin in inguinal WAT merits further investigation.
Previous reports have suggested a role for DMH NPY in glucose homeostasis. DMH NPY knockdown improves glucose intolerance and ameliorates hyperglycemia and hyperinsulinemia in OLETF rats , an animal model of obesity and diabetes , and diet-induced obese rats . Consistent with these reports, DMH NPY overexpression resulted in increased insulin levels in rats on regular chow and caused more severe diet-induced hyperinsulinemia. OGTT confirmed that DMH NPY overexpression caused impaired glucose tolerance and decreased insulin sensitivity independently of body weight effect. Together, these results suggest that DMH NPY has additional actions in glycemic control.
Although NPY-containing neurons have been identified in the ARC and the DMH, the neural circuits underlying their actions appear to differ. ARC NPY serves as one of downstream mediators of leptin's actions in maintaining energy homeostasis [10, 11], but DMH NPY is not under the control of leptin . DMH NPY signaling is affected by brain cholecystokinin  and other yet to be identified molecules . The present findings of downregulation of Npy/Agrp expression and upregulation of Pomc expression in the ARC of rats with DMH NPY overexpression and/or access to HF were likely in response to positive energy balance or increases in body weight and circulating leptin/insulin levels [10-12, 28]. Although Npy induction was reported in the DMH of diet-induced obese mice , we did not replicate this result in mice . The reason for the difference is unclear. We actually found a reduction of DMH Npy expression in rats on HF in both previous  and present studies, implying that this reduction is likely in response to increased energy intake. Furthermore, DMH NPY neurons project to the brainstem nucleus of solitary tract (NTS) and produce inhibitory effects on NTS neurons to modulate food intake . Whether this neural pathway also underlies other effects of DMH NPY such as thermogenesis or energy expenditure remains to be determined.
In summary, we provide new evidence demonstrating the specific role for DMH NPY overexpression in the overall control of energy balance. DMH NPY overexpression causes increased food intake and body weight and exacerbates diet-induced hyperphagia and obesity. This overexpression produces a nocturnal meal size-specific effect that contributes to overall increased food intake. DMH NPY overexpression also lowers energy efficiency, affects fat mass and leptin expression in inguinal WAT, and alters Ucp1 expression in interscapular BAT. Finally, DMH NPY overexpression leads to insulin insensitivity and exaggerates diet-induced hyperinsulinemia. Overall, these results indicate that DMH NPY is an important neuromodulator in modulating energy balance and DMH NPY overexpression can cause hyperphagia and obesity.
We thank Dr. T.H. Moran for comments and discussions on the manuscript.