Endogenous VMH amylin signaling is required for full leptin signaling and protection from diet-induced obesity

Abstract

Amylin enhances arcuate (ARC) and ventromedial (VMN) hypothalamic nuclei leptin signaling and synergistically reduces food intake and body weight in selectively bred diet-induced obese (DIO) rats. Since DIO ¹²⁵I-amylin dorsomedial nucleus-dorsomedial VMN binding was reduced, we postulated that this contributed to DIO ventromedial hypothalamus (VMH) leptin resistance, and that impairing VMH (ARC + VMN) calcitonin receptor (CTR)-mediated signaling by injecting adeno-associated virus (AAV) expressing a short hairpin portion of the CTR mRNA would predispose diet-resistant (DR) rats to obesity on high-fat (45%) diet (HFD). Depleting VMH CTR by 80–90% in 4-wk-old male DR rats reduced their ARC and VMN ¹²⁵I-labeled leptin binding by 57 and 51%, respectively, and VMN leptin-induced phospho-signal transducer and activator of transcription 3-positive neurons by 59% vs. AAV control rats. After 6 wk on chow, VMH CTR-depleted DR rats ate and gained the equivalent amount of food and weight but had 18% heavier fat pads (relative to carcass weight), 144% higher leptin levels, and were insulin resistant compared with control AAV DR rats. After 6 wk more on HFD, VMH CTR-depleted DR rats ate the same amount but gained 28% more weight, had 60% more carcass fat, 254% higher leptin levels, and 132% higher insulin areas under the curve during an oral glucose tolerance test than control DR rats. Therefore, impairing endogenous VMH CTR-mediated signaling reduced leptin signaling and caused DR rats to become more obese and insulin resistant, both on chow and HFD. These results suggest that endogenous VMH amylin signaling is required for full leptin signaling and protection from HFD-induced obesity.

amylin is a 37-amino acid peptide pancreatic β-cell hormone coreleased with insulin in response to food intake (32). Amylin acts on a heterodimeric receptor complex composed of one of two G_s/G_q-protein-coupled calcitonin receptors (CTR_A or CTR_B), combined with one of three receptor activity modifying proteins (RAMP1, -2, and -3) (1, 19, 25, 27). CTR_A is the most prevalent isoform and is widely expressed in the periphery and brain (1, 26). CTR_B is found only in brain but at lower levels than CTR_A. RAMP1, -2, and -3 increase the affinity of amylin for CTR_A (3, 52, 76). The mRNA and protein for CTR_A and CTR_B are present in the ventromedial nucleus (VMN) and arcuate (ARC) hypothalamic nuclei where their genes, as are those for RAMP1, -2 and, -3, are expressed in equal proportions (10, 35). There are high levels of ¹²⁵I-labeled amylin binding to its receptor complex in dispersed brain areas. These include the area postrema (AP), the caudal nucleus tractus solitarius (NTS), dorsomedial hypothalamic nucleus (DMN), and the dorsomedial portion of the VMN (VMNdm) (Fig. 1) (35, 60, 74).

Fig. 1.

Representative ¹²⁵I-labeled amylin receptor binding autoradiograms from diet-induced obese (DIO; A–C) and diet-resistant (DR; D–F) rats in sections through the midpoint of the arcuate (ARC), ventromedial (VMN), and dorsomedial hypothalamic nuclei (DMN). Readings for Fig. 2 were taken in the ARC and the combined DMN and dorsomedial portion of the VMN (VMNdm) and ventrolateral portion of the VMN (VMNvl), as delineated in the respective areas outlined by dotted lines. Total binding (A and D), nonspecific binding (B and E), and cresyl violet (C and F) stained sections, from which autoradiograms in A and D were generated, are shown.

Eating elicits an early rise in circulating amylin levels (68) and peripheral amylin administration reduces eating within minutes, most likely by its rapid access to its receptors in the AP (45), a circumventricular organ that lacks a blood-brain barrier (21). Amylin's actions in the AP/NTS contribute to meal termination and reduction of meal size similar to other satiety hormones, such as cholecystokinin (43, 50, 67). However, peripherally administered amylin also amplifies the leptin-induced phosphorylation of signal transducer and activator of transcription 3 (pSTAT3) in the ARC and VMN [VMN hypothalamus (VMH)] (80) and increases the expression of proopiomelanocortin (POMC) mRNA (70). These actions most likely reflect amylin's ability to cross the blood-brain barrier to reach neurons in the ARC and VMN (6, 7). While intrahypothalamic amylin reduces food intake in rats, this requires supraphysiological doses (16, 17), making it unlikely to be a primary site at which peripheral amylin acts on its own as a short-term satiation factor. On the other hand, chronic peripheral (2, 44) and central (73) amylin administration decreases body weight and fat gain. Most importantly, pharmacological doses of amylin have a prominent synergistic or additive effect with leptin on reducing food intake and body weight in obese rats and humans (2, 18, 20, 44, 55, 66, 71, 72, 77, 78).

Amylin's enhancement of leptin's catabolic effects are paralleled by amylin's increase in leptin-induced pSTAT3 (35, 71, 80), ¹²⁵I-labeled leptin receptor binding (80) and mRNA expression of the long (signaling) form of the leptin receptor (Lepr-b) (35). Our laboratory recently demonstrated (35) that this leptin sensitization occurs when amylin stimulates VMH microglia to produce and secrete IL-6, which, through its action on its IL-6/gp130 receptor, interacts downstream of Lepr-b to increase pSTAT3 expression. This coactivation of pSTAT3 expression is the presumptive mechanism by which exogenous amylin increases VMH leptin signaling (35, 71, 80).

We have used the selectively bred diet-induced obese (DIO) and diet-resistant (DR) rat strains to identify the mechanisms that contribute to the obesity that develops during ingestion of a high-fat diet (HFD) (37, 40). One of the prominent characteristics of the DIO rat is its inherent leptin resistance, which is present during the early postnatal period, before it becomes obese (24, 29, 30, 38, 39). Because of the dramatic ability of exogenously administered amylin to increase defective VMH leptin signaling and reduce food intake and body weight in obese DIO rats (35, 71, 80), we hypothesized that DIO rats might have decreased amylin signaling in the VMH as a contributor to their impaired leptin signaling. Our initial results demonstrated a decrease in ¹²⁵I-labeled amylin binding to its receptor complex in the DMN/VMNdm of DIO rats (Fig. 2), which supported this hypothesis. The present studies were undertaken to test the hypothesis that endogenous CTR-mediated signaling is required for fully intact VMH leptin signaling. Although ¹²⁵I-labeled amylin receptor binding was decreased in the DMN and VMNdm areas of DIO vs. DR rats, we chose to focus on the ARC and VMN in the present studies because of previous studies demonstrating a strong interaction between amylin and leptin signaling in these hypothalamic nuclei (35, 70, 80). Toward this aim, we utilized local VMH injection of an adeno-associated virus (AAV) expressing a short hairpin portion of CTR (CTR shRNA) to deplete the CTR component of the amylin receptor complex (3, 49) in the VMH. We carried this out in the obesity-resistant DR rat to assess whether impairing VMH CTR-mediated signaling would similarly reduce their leptin signaling and make them prone to become obese on a HFD similarly to the DIO rats.

Fig. 2.

Specific ¹²⁵I-labeled amylin receptor binding in the ARC, VMNvl, and DMN/VMNdm of adult male DR and DIO rats fed chow from weaning. Specific binding = total − nonspecific binding. Values are means ± SE. *P ≤ 0.05 when binding (fmol/mg protein) was compared between DIO and DR rats in the VMNdm.

METHODS

Animals

Animals were housed at 23–24°C on a reversed 12:12-h light-dark cycle (lights off at 0900). Male rats selectively bred to express the DR or DIO phenotypes (37) were raised in our in-house colony and used for all studies. Litters were culled to 10 pups per dam on postnatal day 2 (P2) and weaned at P21 onto Purina rat chow and water ad libitum. Purina rat chow (no. 5001) contains 13.5% fat, 28.5% protein, and 58% carbohydrate. In experiment 1, rats were also fed a 45% fat diet, which also contains 20% protein and 35% carbohydrate as a percentage of total energy content (Research Diets, New Brunswick, NJ). Throughout the course of all experiments, cumulative food intake and body weight measurements were obtained weekly. Terminally, rats were fasted 2 h before lights off and killed by decapitation with collection of trunk blood for glucose, insulin, and leptin assays after 2 h. Brains were removed and snap frozen on dry ice, followed by the removal and weighing of five representative fat pads (retroperitoneal, mesenteric, perirenal, epididymal, and inguinal), as well as the liver. All work was in compliance with the Institutional Animal Care and Use Committee of the E. Orange Veterans Affairs Medical Center, which reviewed and approved the studies.

Experiment 1

Groups of male DR rats (n = 8/group) were weaned onto chow at 3 wk of age and, at 23–28 days of age, were stereotaxically injected bilaterally in the ARC or VMN with 0.5 μl/site containing either an AAV (1.25 E9 genome copies) expressing green fluorescent protein (control) or an AAV designed to reduce expression of both CTR_A and CTR_B through expression of CTR shRNA (AAV CTR shRNA) together with green fluorescent protein (47, 49). A third group of DR rats (sham; n = 8) underwent anesthesia, but had no stereotaxic injections. All rats were fed chow for an additional 6 wk. After 6 wk on chow, tail blood was sampled for leptin, insulin, and glucose levels. Then all groups were switched to a 45% fat diet for an additional 6 wk. On the 4th wk on 45% fat diet (10 wk after AAV injections), they underwent an oral glucose tolerance test (OGTT) (23). They were assessed terminally for trunk blood glucose, insulin and leptin, brain punches for a variety of mRNA species (see below), and fat pad and liver weights.

Experiment 2

Based on the findings in experiment 1 demonstrating that VMH (ARC + VMN) CTR depletion caused DR rats to become obese on 45% fat diet, but with no substantial differences between sham and control AAV groups, a second set of studies was undertaken. These studies tested the hypothesis that reducing VMH amylin signaling would produce defective leptin signaling comparable to what is seen in chow-fed DIO rats, which are leptin resistant and predisposed to become obese on HFD (24, 36, 38, 39). Thus male DR rats were weaned onto chow at 3 wk of age and injected with control AAV or AAV CTR shRNA (n = 12/group) at 23–28 days of age as in experiment 1. An additional group of sham-operated DIO rats of comparable age (n = 12) was included for comparison. Given the relatively high titer of AAV used and the ∼90% depletion of VMH CTR_A and CTR_B in experiment 1, only 10% (1.25 E8 genome copies) of the original concentration of AAVs was used in this second set of studies. Rats were fed chow for 6 wk, and, during the 4th wk, all rats underwent an OGTT. During the 6th wk, each group was divided such that one-half of each group was perfused and assessed for leptin-induced pSTAT3 expression, and the other one-half was assessed for ¹²⁵I-labeled leptin receptor binding in the ARC and VMN, respectively.

Stereotaxic Surgery

Rats were anesthetized with 80 mg/kg ketamine/7.5 mg/kg xylazine ip, placed in a stereotaxic frame, and injected bilaterally with 0.5 μl AAV into the ARC (9.6 mm) and VMN (9.3 mm from dura) at 5.9 mm anterior to the intra-aural line and 0.3 mm (ARC) or 0.6 mm (VMN) lateral to the midline. They were given buprenorphine 0.02 mg/kg sc postoperatively. Cannula placements were confirmed histologically postmortem.

Quantitative Real-Time PCR Assessment of CTR shRNA-Induced Changes in Gene Expression

Brains were removed, snap frozen on dry ice, and stored at −80°C for mRNA analysis by quantitative real-time PCR. Frozen brains were sectioned at 300 μm through the midpoint of the VMH (61), and slices were placed in RNA Later (Ambion, Grand Island, NY) until the ARC, VMN, or entire VMH was micropunched under microscopic guidance. Expression of mRNA was assayed by quantitative real-time PCR (Applied Biosystems, Grand Island, NY) using primers shown in Table 1. For data calculations, a standard curve was generated for each gene, and then the abundance of that gene in a given sample was expressed relative to this standard curve (i.e., fold change). This value was then normalized against constitutively expressed cyclophilin in that sample, as previously described (34, 59).

Table 1.

Quantitative PCR probe sets

Gene	Genbank ID	Forward	Reverse	Probe
AGRP	NM_033650.1	TCTCCGCGTCGCTGTGTA	CGCAGCAAGGTACCTGTTGTC	AGGACTCGTGCAGCCT
CTRB	NM_053816.2	TGGTTGAGGTTGTGCCCAAT	GTTCCCACTGCATTGTCCACATATA	TCGCACCAGGTCTCC
Cyclophilin	NM_017101.1	AATGGCACTGGTGGCAAGTC	GCCAGGACCTGTATGCTTCAG	TCTACGGAGAGAAATT
GP130	NM_001008725.3	GGATCTGGCTAATGCAAGCTTTG	GACAACTGGAAACTCAGGGTAGAT	CTGAGTCTATAGGTCAACTTG
IL6	NM_012589.2	TTCACAAGTCGGAGGCTTAATTACAT	GCATCATCGCTGTTCATACAATCAG	CACAACTCTTTTCTCATTTCC
IL6rα	NM_017020.3	CACTATTTGTGCTTCCTGGATGATC	GCCACTCACAAAAGGCATTTACAAG	CTGGTGGATGTTCCCC
NPY	NM_012614.2	TCGTGTGTTTGGGCATTCTG	GCGGAGTAGTATCTGGCCATGT	ACAATCCGGGCGAGGA
Leprb	NM_012596.1	AACCTGTGAGGATGAGTGTCAGAGT	CCTTGCTCTTCATCAGTTTCCA	ATGCAACGCTGGTCAG
POMC	NM_139326.2	GGCGTGCGGAGGAAGAG	GCCCTCCCGTGGACTTG	TGGCCGTCCGGAGC
RAMP1	NM_031645.1	GGAAGACTCTGTGGTGTGACT	GCCAGAAACAGCCAATCTTGT	TCACTGCACCAAACTC
RAMP2	NM_031646.1	GTGCAACTGGACTTTGATTAGCA	GCCTCGTACTCCAAGCAATACC	CAGGTTGCTGTAATACC
RAMP3	NM_020100.2	ACGTCTGGAAGTGGTGCAA	TGCAGTTAGTGAAGCTTTCGTAGT	CCTGTCGGAGTTCATC
SOCS3	NM_053565.1	CGCCGTGCGCAAGCT	TCGCCGCCGGTTACG	CTCCAGTAGAATCCG

For most of the genes, TAQman MGB probe sets (Applied Biosystems) were designed by the manufacturer. This table shows the unlabeled forward and reverse PCR primers, along with the FAM-labeled probe sequence. For calcitonin (CTR) receptor A (CTR_A) (GenBank ID NM_001034015.1), a predesigned TaqMan primer set was used (Applied Biosystems assay ID Rn01526770_m1). AGRP, agouti-related protein; CTR_B, CTR receptor B; IL6rα, IL-6 receptor alpha; NPY, neuropeptide Y; Leprb, long (signaling) form of the leptin receptor; POMC, proopiomelanocortin; RAMP, receptor activity modifying protein; SOCS3, suppressors of cytokine signaling-3.

Leptin-Induced Expression of ARC and VMN pSTAT3 Immunohistochemistry

Food was removed at 2 h before lights off (0900) in one set of rats from experiment 2. At lights off, rats were injected intraperitoneally with 0.5 ml PBS containing rat leptin (5 mg/kg; National Hormone and Peptide Program, Torrance, CA). After 45 min, they were anesthetized with ketamine-xylazine and then rapidly perfused with ice-cold saline for 1 min, followed by ice-cold 4% paraformaldehyde for 10 min. Brain sections (30 μm) were cut through the mid-VMH (61) and then assayed for pSTAT3 immunocytochemistry, as previously described (58). Neurons expressing pSTAT3 were counted in three consecutive sections per brain using an image analysis system (Bioquant, Nashville, TN) by an experimentally naive observer. No rats were injected only with PBS to assess constitutive expression of pSTAT3, since our laboratory's previous studies demonstrated extremely low levels of such expression in the ARC and VMN of DIO and DR rats (38).

OGTT

Food was removed 2 h before lights off, and, at lights off, rats were gavaged with 0.5 g/kg, and blood was sampled by tail nip at 15, 30, 60, 90, and 120 min for glucose and insulin.

¹²⁵I-Labeled Amylin Receptor Binding Autoradiography

Brains from groups of nonfasted male, 8 wk old, chow-fed DIO and DR rats (n = 8/group) were removed at lights out and quickly frozen on powdered dry ice. Sections (12 μm) were cut through the midpoint of the ARC, VMN, and DMN, pars compacta (Fig. 1) (61), mounted on gel-coated slides, desiccated, and stored at −80°C. Amylin receptor binding was carried out by methods adapted (35) from Sexton et al. (74). Briefly, sections were thawed and rinsed in incubation buffer (20 mM HEPES containing 100 mM NaCl, 1 mg/ml BSA, and 0.5 mg/ml bacitracin). Sections were then incubated at room temperature for 1 h in incubation buffer containing 70–75 pM ¹²⁵I-labeled amylin (NEX44; Perkin Elmer, Boston, MA) or 70–75 pM radiolabeled amylin plus 1 μM unlabeled rat amylin (nonspecific “binding”; Bachem, Torrance, CA). Slides were rinsed in incubation buffer at 4°C and rinsed two more times at 4°C in modified incubation buffer (20 mM HEPES containing 100 mM NaCl). After a brief dip in dH₂O, sections were dried under forced cold air and desiccated for 24 h. Sections were then exposed to BioMax MR Film (Kodak, Rochester, NY) at −80°C for 7–14 days. Quantification was carried out by an experimentally naive observer, whereby the areas to be read (ARC, DMN/VMN, and dorsolateral VMN) were first defined as shown in Fig. 1 on the same cresyl violet-stained section that had been used to generate that specific autoradiogram. This image was digitally superimposed on the autoradiographic image to assess binding density in the selected brain regions using computer-assisted densitometry (Drexel University, Philadelphia, PA). Film density was equated to the density of known ¹⁴C radioactive standards (calibrated against ¹²⁵I) placed on each film, and binding was calculated directly from density and the specific activities of ¹²⁵I-labeled amylin (29). Specific binding was calculated as the total binding in a specific area minus nonspecific “binding” as defined above. Nonspecific “binding” averaged 24% of total binding in the ARC and VMN.

¹²⁵I-Labeled Leptin Receptor Binding Autoradiography

The procedure for ¹²⁵I-labeled leptin binding was carried out by procedures previously established in our laboratory (29), as adapted from the original method of Baskin et al. (8). Frozen slides were brought to room temperature, dipped in cold 0.5% paraformaldehyde, and then rinsed several times in cold Tris buffer. Binding was carried out by incubating the slides for 1 h at room temperature in a solution containing 0.25 nM murine ¹²⁵I-labeled leptin (PerkinElmer Life Sciences, Boston, MA), Tris·HCl (pH 7.2), 1% BSA (Sigma, St. Louis, MO), 0.05% leupeptin (Sigma), and 0.001% pepstatin (Sigma). Nonspecific binding (23% of total binding) was defined as that seen in the presence of a 1,000-fold excess of unlabeled rat leptin (National Hormone and Peptide Program, Torrance, CA), and this accounted for <30% of total binding. After 1 h incubation in the respective radioactive solutions, total and nonspecific labeled slides were rinsed in cold Tris·HCl and dipped in cold 4% paraformaldehyde, followed by a dip in Tris·HCl. The slides were dipped in cold distilled water, dried overnight, exposed to BioMax MR film (Kodak, Rochester, NY) for 3–5 days, and read by an experimentally naive observer, as described above.

Assays of Insulin, Leptin, Free Fatty Acid, β-Hydroxybutyrate, and Glucose

Plasma insulin and leptin levels were analyzed with radioimmunoassay and ELISA kits (EMD Millipore, Billerica, MA) with sensitivities of 0.8 ng/μl per 100 μl and 10-μl sample size, respectively. Blood glucose levels were measured using a glucose Analox instrument (Lunenburg, MA). Colorimetric assays from Wako Diagnostics (Mountainview, CA) were used to measure fatty acids (NEFA kit) and β-hydroxybutyrate (3-H3 kit).

Statistics

Groups were compared by one-way and repeated-measures ANOVA. Post hoc assessment was by Bonferroni correction for multiple comparisons. For OGTTs, areas under the curve (AUC) for insulin and glucose levels during the testing period were calculated using the trapezoidal rule. Insulin sensitivity index (ISI) was calculated according to the Matsuda and DeFronzo (46) formula: [10,000/square root of (fasting glucose × fasting insulin) × (mean glucose × mean insulin during OGTT)].

RESULTS

Experiment 1

¹²⁵I-labeled amylin binding in DIO vs. DR rats.

To assess potential differences between the ability of amylin binding, receptor binding autoradiography was carried out in sections through the midpoint of the ARC-VMN-DMN area of the hypothalamus (Fig. 1) in adult DIO vs. DR rats. This binding was 30% lower in the VMNdm and DMN in DIO vs. DR rats (Fig. 2). There were no differences in binding between DIO and DR rats in the ARC or ventrolateral portion of the VMN.

Effects of depleting VMH CTR on energy homeostasis in DR rats fed low-fat diets and HFDs.

To test the hypothesis that the reduced amylin binding to its CTR-RAMP receptor complex in the VMH contributes to the propensity of DIO rats to become obese when fed a HFD, DR rats were injected in the VMH with either a control AAV or an AAV expressing CTR shRNA, and these results were compared with those of sham-operated DR rats. When fed chow from the time of VMH injections (4 wk of age) for 6 wk, there were no differences in body weight gain or food intake among the three groups (Fig. 3, Table 2). Given their similar weight gains and intakes, there were no intergroup differences in feed efficiency [body weight gain (g)/food intake (kcal)] after 6 wk on chow. However, blood samples drawn at the end of the 6 wk on chow showed that, despite the lack of differences in intake or weight gain, fasting plasma leptin levels in VMH CTR-depleted DR rats were 207 and 106% higher than sham- and control AAV-injected DR rats, respectively [F(2,17) = 13.7; P = 0.001; Table 2]. In keeping with the likely increase in adiposity of VMH CTR-depleted rats inferred by their increased leptin levels, these rats had 31 and 76% higher fasting insulin levels than sham and AAV control DR rats, respectively [F(2,17) = 21.7; P = 0.001; Table 1]. Thus, although depleting VMH CTR had no apparent effect on body weight gain or food intake, elevated plasma leptin and insulin levels in these rats suggested that they were more obese and possibly insulin resistant compared with sham- and control AAV-injected rats when fed only low-fat chow for 6 wk.

Fig. 3.

Experiment 1: body weight gain (A) and food intake (B) in DR rats that were sham, control adeno-associated virus (AAV), and AAV containing a short hairpin portion of the calcitonin receptor mRNA (AAV CTR shRNA) injected bilaterally in the ARC and VMN at 4 wk of age and fed chow (13.5% fat) for 6 wk, followed by 6 wk on 45% fat diet. Values are means ± SE. *P ≤ 0.05 when food intake and body weight gain data at a specific time point (on x-axis) in AAV CTR shRNA-injected rats were compared with that in sham and control AAV-injected rats by post hoc Bonferroni test after intergroup differences were found by one-way ANOVA for repeated measures.

Table 2.

Experiment 1: Morphometric and biochemical data for male DR rats injected at 4 wk of age with AAV control shRNA or AAV CTR shRNA in the VMH

	Sham	AAV Control	AAV CTR
6-wk chow
Initial body weight, g	87.6 ± 4.2	89.8 ± 3.8	84.4 ± 5.7
Body weight chow, g	306 ± 8	301 ± 12	287 ± 14
Body weight gain on chow, g	218 ± 9	212 ± 9	202 ± 11
Food intake on chow, kcal	2,629 ± 42	2,619 ± 110	2,505 ± 135
Feed efficiency on chow, BWG (g)/FI (kcal) × 1,000	83 ± 3	81 ± 3	81 ± 3
Leptin, ng/ml	5.14 ± 0.9a	7.64 ± 0.89a	15.8 ± 2.43b
Glucose, mg/dl	137.63 ± 1.75	137.94 ± 7.46	138.69 ± 3.25
Insulin, ng/ml	0.80 ± 11a	1.05 ± 0.10a	1.85 ± 0.13b
6-wk 45% fat diet
Body weight gain, g	126 ± 9a	131 ± 18a	168 ± 13b
Food intake, kcal	1,594 ± 39	1,560 ± 72	1,613 ± 87
Feed efficiency, BWG (g)/FI (kcal) × 1,000	33 ± 1	35 ± 3	39 ± 5
Total 6-wk chow, 6-wk 45% fat diet
Final body weight, g	435 ± 14	433 ± 25	459 ± 23
Total body weight gain, g	347 ± 18	343 ± 23	375 ± 23
Total food intake, kcal	6,488 ± 173	6,310 ± 312	6,609 ± 370
Feed efficiency total, BWG (g)/FI (kcal) × 1,000	53 ± 1	54 ± 2	54 ± 3
Retroperitoneal, g	9.28 ± 0.78a	10.52 ± 0.79a	13.87 ± 1.22b
Retroperitoneal, %BW	2.15 ± 0.17a	2.43 ± 0.12a,b	3.00 ± 0.22b
Mesenteric, g	7.79 ± 1.45a	8.39 ± 0.81a	15.0 ± 1.7b
Mesenteric, %BW	1.79 ± 0.16a	1.94 ± 0.17a	3.31 ± 0.31b
Inguinal, g	6.47 ± 0.68a	8.77 ± 1.01a	16.89 ± 1.03b
Inguinal, %BW	1.81 ± 0.34a	2.05 ± 0.25a	3.35 ± 0.28b
Epididymal, g	9.38 ± 0.66	11.0 ± 1.3	16.0 ± 1.7
Epididymal, %BW	2.18 ± 0.15a	2.53 ± 0.20a	3.75 ± 0.34b
Perirenal, g	3.50 ± 0.12	4.05 ± 0.52	5.70 ± 0.68
Perirenal, %BW	0.81 ± 0.04	0.93 ± 0.08	1.14 ± 0.09
Total fat pads, g	38.0 ± 3.3a	42.8 ± 3.8a	70.4 ± 8.5b
Total fat pad/carcass weight, %	8.07 ± 0.45a	9.88 ± 0.68a	15.8 ± 0.9b
Leptin, ng/ml	11.2 ± 0.7a	18.8 ± 2.3a	39.6 ± 2.9b
Glucose, mg/dl	148 ± 4	152 ± 6	143 ± 3
Insulin, ng/ml	1.75 ± 0.28a	2.63 ± 0.46a,b	3.66 ± 0.46b
Glucose AUC, mg·dl−1·120 min−1	4,932 ± 813	4,643 ± 606	5,731 ± 764
Insulin AUC, mg·dl−1·120 min−1	87.9 ± 19.7a	91.1 ± 17.7a	204.1 ± 30.6b
Insulin sensitivity index	34.2 ± 5.2a	30.9 ± 4.2a	17.4 ± 2.5b
Liver, g	13.64 ± 1.08	12.80 ± 0.89	12.66 ± 0.68
Naso-anal length, cm	24.8 ± 0.2	24.4 ± 0.4	23.6 ± 0.3

Values are means ± SE; n = 7 AAV control shRNA rats; n = 8 AAV CTR shRNA rats. Additional diet-resistant (DR) rats were sham-injected (n = 7). Rats were maintained on chow diet for 6 wk, followed by 6 wk on 45% fat diet.

VMH, ventromedial hypothalamus; AAV CTR shRNA, adeno-associated virus containing a short hairpin portion of the CTR mRNA; AAV, adeno-associated virus; BW, body weight; BWG, BW gain; FI, food intake; AUC, area under the curve.

^a,bData for each parameter with differing superscripted letters differ from each other by P = 0.05 by post hoc Bonferroni test after intergroup differences were found by one-way ANOVA.

When all three groups of DR rats were switched to a 45% HFD for 6 wk, there were no intergroup differences in food intake. However, VMH CTR-depleted rats gained 33 and 28% more weight than sham- and control AAV-injected rats, respectively, over the ensuing 6 wk [F(2,19) = 5.59; P = 0.05; Fig. 3, Table 2]. In keeping with their increased body weight gain, CTR-depleted rats had 65% heavier total fat pad weights [F(2,18) = 25.3; P = 0.001; Table 2] and 60% higher fat pad weights as a function of carcass weight compared with AAV controls [F(2,18) = 23.3; P = 0.001; Table 2]. This increased adiposity was distributed fairly equally among both intra-abdominal (retroperitoneal, mesenteric) and subcutaneous (inguinal) depots and was reflected in a 254 and 111% increase in plasma leptin levels compared with sham and control AAV-injected rats, respectively [F(2,17) = 41.9, P = 0.001; Table 2]. Fasting insulin levels in CTR-depleted rats were 109% greater than those in sham rats [F(2,19) = 5.57, P = 0.05], while insulin AUCs during an OGTT were 223 and 244% higher [F(2,17) = 14.8; P = 0.001], and ISIs were 44 and 49% lower than control AAV-injected and sham rats, respectively [F(2,17) = 12.9; P = 0.001; Table 2].

Effects of depleting VMH CTR on amylin-IL-6 signaling and energy homeostasis mRNAs.

Injecting AAV expressing CTR shRNA bilaterally into both the ARC and VMN reduced ARC CTR_A [F(2,18) = 17.2, P = 0.001] and CTR_B mRNA [F(2,18) = 10.3, P = 0.001] expression to 12% of control AAV levels (Table 3) and VMN CTR_A to 14% [F(2,18) = 12.3, P = 0.001] and CTR_B to 5% of control AAV levels [F(2,18) = 15.3, P = 0.001; Table 4]. As a consequence of depleting ARC CTR, RAMP1 mRNA was decreased by 24% [F(2,18) = 12.3, P = 0.001], while RAMP2 was increased by 17% relative to AAV controls [F(2,18) = 8.53, P = 0.01]. Furthermore, IL-6 mRNA was reduced by 67% compared with AAV controls [F(2,18) = 15.6, P = 0.001; Table 4], while there were no effects on the expression of the IL-6 receptor complex (IL6 receptor α, gp130), neuropeptide Y, agouti-related protein, POMC, Lepr-b, or suppressors of cytokine signaling-3 (SOCS3) expression. VMN depletion of CTR had no effects on any other mRNA species assessed (Table 4).

Table 3.

Experiment 1: Effects of VMH CTR depletion from 4 wk of age on arcuate gene expression in DR rats after 6 wk on chow and 6 wk on 45% fat diet

	Sham	AAV Control	AAV CTR shRNA
CTRA	2.92 ± 0.60a	3.23 ± 0.70a	0.37 ± 0.10b
CTRB	2.16 ± 0.34a	1.12 ± 0.21b	0.12 ± 0.04c
RAMP1	1.20 ± 0.17a	1.17 ± 0.15a	0.90 ± 0.05b
RAMP2	1.28 ± 0.14a	1.12 ± 0.07b	1.31 ± 0.13a
RAMP3	0.94 ± 0.10	1.30 ± 0.20	1.11 ± 0.15
NPY	1.38 ± 0.32	1.02 ± 0.27	1.39 ± 0.25
POMC	1.75 ± 0.47	1.28 ± 0.48	1.03 ± 0.23
AgRP	1.51 ± 0.43	0.30 ± 0.45	1.24 ± 0.32
IL-6	0.76 ± 0.09a	2.82 ± 0.84b	0.92 ± 0.19a
IL-6Rα	0.72 ± 0.09	0.75 ± 0.12	1.01 ± 0.12
gp130	0.93 ± 0.12	1.00 ± 0.21	0.95 ± 0.11
Lepr-b	1.09 ± 0.22	1.07 ± 0.20	1.22 ± 0.17
SOCS3	0.86 ± 0.16	1.31 ± 0.34	1.03 ± 0.18

Values are means ± SE of duplicate determinations expressed relative to the amount of the mRNA expression of cyclophilin (n = 7–8/group).

^a,b,cData for each parameter with differing superscripted letters differ from each other by P = 0.05 by post hoc Bonferroni test after intergroup differences were found by one-way ANOVA.

Table 4.

Experiment 1: Effects of VMH CTR depletion from 4 wk of age on ventromedial nucleus gene expression in DR rats after 6 wk on chow and 6 wk on 45% fat diet

	Sham	AAV Control	AAV CTR shRNA
CTRA	1.18 ± 0.25a	1.73 ± 0.37a	0.25 ± 0.13b
CTRB	0.99 ± 0.19a	1.30 ± 0.31a	0.07 ± 0.04b
RAMP1	1.57 ± 0.29	1.75 ± 0.49	1.27 ± 0.23
RAMP2	1.22 ± 0.10	0.90 ± 0.14	0.97 ± 0.12
RAMP3	1.65 ± 0.17	1.55 ± 0.16	1.45 ± 0.16
IL-6	0.77 ± 0.11	0.87 ± 0.13	0.82 ± 0.09
IL-6Rα	0.50 ± 0.16	1.06 ± 0.31	1.30 ± 0.30
gp130	2.01 ± 0.19	2.38 ± 0.24	2.43 ± 0.23
Lepr-b	1.07 ± 0.13	0.95 ± 0.18	0.81 ± 0.12
SOCS3	0.42 ± 0.11	0.74 ± 0.13	0.52 ± 0.03

Values are means ± SE of duplicate determinations expressed relative to the amount of the mRNA expression of cyclophilin (n = 7–8/group).

^a,bData for each parameter with differing superscripted letters differ from each other by P = 0.05 by post hoc Bonferroni test after intergroup differences were found by one-way ANOVA.

Experiment 2

Effects of depleting VMH CTR on energy and glucose homeostasis in DR rats fed low-fat diet for 6 wk.

Inherent leptin resistance is a major factor predisposing selectively bred DIO rats to become obese when exposed to a HFD (24, 36, 38, 39). To test the hypothesis that the reduced DMN/VMNdm amylin receptor binding seen in DIO rats (Fig. 2) might underlie their reduced leptin signaling and propensity to become obese on HFD, DR rats were injected in the VMH at 4 wk of age with control AAV or AAV expressing CTR shRNA and compared with each other and to sham-injected DIO rats fed only low-fat chow diet for 6 wk. Similar to the results in experiment 1, DR rats injected in the VMH with AAV CTR shRNA gained the same amount of body weight and ate the same amount of food as control AAV DR rats (Fig. 4, Table 5). Despite similarities in weight gain and energy intake, VMH CTR-depleted DR rats had 18% more total fat and 25% heavier retroperitoneal fat pads (as a percentage of carcass weight) and 144% higher plasma leptin levels than control AAV-injected DR rats (Table 5). In keeping with their greater adiposity, CTR-depleted DR rats also had 30% higher plasma insulin levels, 25% higher insulin AUC during an OGTT, and 23% lower ISI than DR controls (Table 5). Their increased adiposity and likely insulin resistance was associated with an 81% decrease in total VMH (ARC + VMN) CTR mRNA compared with control AAV-injected DR rats (Table 5). Thus substantial depletion of intrinsic CTR amylin signaling caused DR rats fed only low-fat chow diet to become more obese and probably insulin resistant.

Fig. 4.

Experiment 2: body weight gain (A) and food intake (B) in DIO sham and DR rats that were injected with control AAV or AAV CTR shRNA bilaterally in the ARC and VMN at 4 wk of age and fed chow (13.5% fat) for 6 wk. Values are means ± SE. *P ≤ 0.05 when food intake and body weight gain data at a specific time point (on x-axis) from sham DIO rats were compared with that from control AAV and AAV CTR shRNA-injected DR rats by post hoc Bonferroni test after intergroup differences were found by one-way ANOVA for repeated measures.

Table 5.

Experiment 2: Morphometric and biochemical data for male sham-operated DIO and DR rats injected at 4 wk of age with control AAV shRNA or AAV CTR shRNA in the VMH

	DIO Sham	DR Ctrl AAV	DR AAV CTR shRNA
Initial body weight, g	86 ± 4a	66 ± 2b	65 ± 2b
Final body weight, g	433 ± 11	288 ± 6b	283 ± 8b
Body weight gain, g	347 ± 9a	222 ± 6b	218 ± 8b
Food intake, kcal	4,669 ± 115a	3,221 ± 81b	3,264 ± 112b
Feed efficiency, BWG (g)/FI (kcal) × 1,000	74.3 ± 7.2	68.9 ± 6.9	66.8 ± 7.1
Perirenal, g	1.31 ± 0.14a	0.45 ± 0.08b	0.43 ± 0.04b
Mesenteric, g	5.86 ± 0.30a	2.81 ± 0.13b	2.85 ± 0.16b
Inguinal, g	5.72 ± 0.27a	2.85 ± 0.16b	3.08 ± 0.17b
Epididymal, g	5.92 ± 0.32a	2.62 ± 0.10b	2.60 ± 0.21b
Perirenal, g	1.31 ± 0.14a	0.45 ± 0.08b	0.43 ± 0.04b
Retroperitoneal, g	4.64 ± 0.24a	1.55 ± 0.13b	1.87 ± 0.12c
Retroperitoneal carcass, %	1.10 ± 0.11a	0.52 ± 0.04b	0.65 ± 0.05c
Total fat pad total, g	25.8 ± 2.2a	10.1 ± 0.4b	11.6 ± 0.4c
Total fat pad/carcass weight, %	5.9 ± 0.3a	3.4 ± 0.1b	4.0 ± 0.1c
Leptin, ng/ml	16.2 ± 2.2a	6.1 ± 0.7b	14.9 ± 2.1a
Glucose, mg/dl	154 ± 8a	132 ± 5b	130 ± 6b
Insulin, ng/ml	1.45 ± 0.14a	0.87 ± 0.11b	1.12 ± 0.10c
Glucose AUC, mg·dl−1·120 min−1	3,789 ± 540a	2,398 ± 425b	2,934 ± 469b
Insulin AUC, mg·dl−1·120 min−1	110 ± 10a	72 ± 5b	90 ± 6c
Insulin sensitivity index	19.4 ± 1.3a	33.3 ± 3.5b	25.5 ± 2.3c
CTR mRNA	0.71 ± 0.08a	1.40 ± 0.20b	0.27 ± 0.11c

Values are means ± SE; n = 12/group.

DIO, diet-induced obese. Rats were maintained on chow diet for 6 wk.

^a,b,cData for each parameter with differing superscripted letters differ from each other by P = 0.05 by post hoc Bonferroni test after intergroup differences were found by one-way ANOVA. CTR mRNA levels are given as relative abundance normalized against cyclophilin.

As in previous studies (24, 36–39), DIO rats were heavier than both groups of DR rats at 4 wk of age [F(2,33) = 25.5; P = 0.001] and gained 23% more body weight [F(2,33) = 113; P = 0.001], ate 45% more chow [F(2,33) = 60.9, P = 0.001] (Fig. 4), had 18% more total carcass fat as a percentage of body weight [F(2,33) = 39.0; P = 0.001] and 165% higher plasma leptin levels [F(2,33) = 30.3, P = 0.001] than did chow-fed control DR rats (Table 5). However, although the VMH CTR-depleted DR rats were more obese than their DR controls, they had only 68% as much total fat as a percentage of carcass weight, as did DIO rats when assessing five fat pads. On the other hand, plasma leptin levels were comparable between DIO and DR CTR-depleted rats, suggesting that impairing DR VMH amylin signaling might have caused them to become almost as obese as DIO rats when fed only low-fat diet. Finally, while depleting VMH CTR in DR rats made them more insulin resistant than control DR rats, they still had 23% lower basal insulin levels, 18% lower insulin AUC, and 31% higher ISI than did DIO rats (Table 5). Thus, although depleting VMH CTR expression in DR rats made them more obese and insulin resistant than control DR rats, they did not become as obese or insulin resistant as DIO rats when all groups were fed chow.

Effects of depleting DR VMH CTR on leptin signaling.

Depleting DR VMH CTR by 81% resulted in a 57% decrease in ARC and 51% decrease in VMN ¹²⁵I-labeled leptin binding compared with DR controls, respectively (Fig. 5A). This reduced binding was mirrored by a 59% decrease in the number of leptin-induced pSTAT3-positive neurons in the VMN, but no difference in the ARC compared with control DR rats (Fig. 5B). As our laboratory has shown previously (29), in the present studies, chow-fed DIO rats had 55 and 16% lower leptin receptor binding in the ARC and VMN, respectively, and 27% reduced leptin-induced pSTAT3 only in the VMN compared with control AAV-injected DR rats (Fig. 5B). These data strongly support the hypothesis that significantly impairing endogenous VMH amylin signaling by depleting the CTR component of its receptor complex impairs leptin signaling, increases adiposity, produces insulin resistance, and may be an important factor that predisposes such depleted DR rats to become obese on HFD (experiment 1). However, despite 80% reduction in VMH CTR expression in DR rats, they did not become as obese or insulin resistant as control DIO rats on low-fat chow diets.

Fig. 5.

Experiment 2: ¹²⁵I-labeled leptin binding (A) and leptin-induced phosphorylation of signal transducer and activator of transcription 3 (pSTAT3) immunohistochemistry (B) in the ARC and VMN of DIO sham, DR AAV control, and DR AAV CTR shRNA rats injected at 4 wk of age and fed chow for 6 wk. Values are means ± SE; N = 12/group. ^a,b,c|Data with differing superscripted letters differ from each other by P ≤ 0.05 by post hoc Bonferroni test after intergroup differences were found by one-way ANOVA.

DISCUSSION

Compared with DR rats, selectively bred DIO rats have reduced ARC and VMN leptin signaling, which is associated with impaired physiological and biochemical responses to leptin before exposure to HFD that make them obese (24, 29, 30, 38, 39). Here we show that DIO rats also have reduced ¹²⁵I-labeled amylin binding in the DMN and VMNdm, an area that overlaps with their decreased ¹²⁵I-labeled leptin binding (29) and leptin-induced pSTAT3 expression (13, 38, 58). Amylin enhances Lepr-b expression (35), ¹²⁵I-labeled leptin binding (80), and leptin-induced STAT3 activation (71) and produces a synergistic effect on lowering food intake and body weight in obese rodents and humans (71). Thus we postulated that reducing endogenous amylin signaling in the VMH by depleting the CTR component of the amylin receptor complex would impair leptin signaling and make obesity-resistant DR rats “leptin resistant” and prone to become obese on HFD. In fact, depleting ARC and VMN CTR_A and CTR_B by 80–95% in DR rats did impair their leptin signaling (ARC and VMN ¹²⁵I-labeled leptin binding and VMN leptin-induced pSTAT3). In addition to the marked depletion of VMH CTR_A and CTR_B, RAMP1 mRNA expression was also decreased in the ARC, which should reduce the affinity of amylin for the CTR/RAMP complex (76). However, we have no direct proof that reductions in CTR and RAMP1 actually reduced amylin binding and signaling since, due to technical problems, we were unable to carry out ¹²⁵I-labeled amylin receptor binding studies. In addition, both calcitonin gene-related peptide (CGRP) (75) and amylin peptide (41) are expressed in the hypothalamus, and both can serve as ligands for CTR and/or the CTR-RAMP complex (3, 52, 76). Amylin peptide is present in Lepr-b-expressing neurons in many of the same hypothalamic areas where ¹²⁵I-labeled amylin binds, where its expression varies as a function of sex, obesity, and the presence of leptin (41). Thus, although amylin is expressed in low abundance in the hypothalamus, it, rather than peripheral amylin, might serve as the endogenous ligand that regulates leptin signaling within the VMH. Finally, although injections of the AAVs likely did spread to other areas, such as the DMN, our focus on the ARC and VMN as mediators of the physiological effects on energy and glucose homeostasis seems warranted, since this is where changes in leptin signaling occurred as a result of ARC and VMN CTR depletion.

While chow-fed DIO rats gained more weight and adiposity and were insulin resistant [by OGTT and ISI (46)] compared with chow-fed control DR rats, chow-fed VMH CTR-depleted DR rats also had increased adiposity and insulin resistance compared with DR controls. However, these changes occurred in the absence of increased body weight gain or food intake in VMH CTR-depleted DR rats. Most importantly, neither chow-fed DIO controls nor VMH CTR-depleted DR rats had particularly high levels of carcass adiposity (6 and 4% total fat pad and carcass weight, respectively) compared with that seen after the latter was fed HFD for 6 wk (16%). Unfortunately, we had no HFD-fed DIO rats to compare to those CTR-depleted DR rats fed HFD in experiment 1, so it is not possible to say whether impairing their VMH CTR and leptin signaling would have caused them to become as obese and insulin resistant as DIO rats fed HFD. Nevertheless, it is clear that impairing signaling mediated by VMH CTR does significantly reduce leptin signaling and predispose DR rats to obesity and insulin resistance when fed HFD. Thus the main conclusion from these studies is that endogenous signaling via VMH CTR is required for full leptin signaling and protection from developing obesity and insulin resistance on a HFD. However, the endogenous ligand responsible for this signaling, be it peripheral or hypothalamic amylin or CGRP, remains to be clarified.

VMH CTR-depleted DR rats had increased adiposity without increased intake when fed only chow. When fed HFD for 6 wk, they increased their body weight gain and adiposity in the absence of increased food intake compared with DR controls. These findings suggest that reduced VMH amylin signaling decreased their energy expenditure to account for weight gain without hyperphagia. This is in keeping with previous studies demonstrating that both acute systemic (31, 35, 54, 82) and centrally administered amylin increase energy expenditure, even though this occurs in the presence of decreased food intake (54, 83). Chronic central administration does not affect overall energy expenditure, but does reduce the respiratory quotient (83). Thus, although very large doses of amylin injected directly into the hypothalamus can reduce food intake in rats (16, 17), our data suggest that the primary effect of amylin-leptin interactions in the VMH may be on energy expenditure (either increased thermogenesis and/or activity) rather than food intake. On the other hand, most available evidence points to the AP and underlying NTS as the primary sites of amylin's acute satiating effects (45, 64, 68), as well as the longer term anorectic effects of chronic amylin infusions (44). Besides its direct effects, amylin also interacts in the hindbrain with other gut peptides like cholecystokinin (11, 42, 51), as well as leptin (33, 35, 48), to enhance their satiating effects. However, in addition to the AP/NTS, short-term amylin-induced reductions in feeding can also be elicited by injection of salmon calcitonin (as an amylin surrogate) into the ventral tegmental area (VTA) (where it interacts with leptin) (48) and the nucleus accumbens where amylin reduces food intake in an opiate-dependent fashion (4, 5).

This discrepancy between the acute satiating effects of amylin and its chronic effects on reducing both food intake and body weight in obese rodents and humans (71, 79) raises the possibility of at least two different mechanisms of action for amylin. The acute satiating effects in the AP/NTS appear to be dependent on phosphorylation of ERK1/2 (63). Our present data and several other studies (35, 69, 71, 79, 80) suggest that the chronic effects of amylin administration on reducing body weight in obese rodents and humans (2, 18, 20, 44, 55, 66, 71, 72, 77, 78) are dependent on long-term amylin interactions with leptin signaling at its receptor and downstream signaling cascade in the VMH (69, 71, 79). Our laboratory previously demonstrated that exogenous amylin given over several days, in vivo or in vitro, stimulates VMH microglia to produce IL-6, which interacts with its IL-6 receptor-α/gp130 receptor complex on Lepr-b-expressing neurons to coactivate STAT3 downstream of both receptors (35). This increased pSTAT3 expression and its presumptive stimulation of gene transcription are the likely mechanisms by which amylin increases VMH Lepr-b mRNA expression (35), leptin receptor binding (80), as well as ARC POMC expression (70).

While we did not directly assess amylin's interactions with its VMH receptors, the decrease in IL-6 seen here in the ARC of VMH CTR-depleted DR rats (at least compared with control AAV-injected rats) suggests that amylin signaling might have been reduced in those rats as a proximate cause of their decreased VMH leptin signaling. Our laboratory's previous studies (35) strongly suggest that IL-6 is the critical messenger linking exogenous amylin administration with VMH leptin signaling. For example, amylin fails to reduce food intake or body weight or to enhance VMH leptin-induced pSTAT3 expression in IL-6 knockout mice (35). Thus our present results are compatible with the contention that endogenous amylin signaling via the CTR, together with amylin-induced IL-6 production, are required for full VMH leptin signaling and protection from HFD-induced obesity. However, utilizing the same AAV CTR shRNA as we used here, Mietlicki-Baase et al. (49) showed that knockdown of CTR in the VTA exacerbated both the hyperphagia and the obesity seen in rats fed HFD, suggesting that other brain areas besides the VMH are involved in amylin's long-term effects on energy homeostasis.

Our present studies raise the issue of how endogenous amylin signaling and pharmacological amylin therapy differ and why amylin “resistance” does not seem to develop in obese individuals, despite their elevated plasma amylin levels (14, 15). The present studies, and those in which the CTR portion of the amylin receptor complex was knocked down in the VTA (49), provide evidence that endogenous CTR-mediated signaling, possibly utilizing peripheral amylin transported into the brain or endogenous amylin or CGRP, is an important mediator of ongoing regulation of energy and glucose homeostasis. Although we did not measure plasma amylin levels here, obese rodents are well documented to have elevated levels (14, 15, 62), much as their plasma leptin levels are elevated. Yet such elevations of endogenous amylin and leptin levels are not sufficient to combat the development of hyperphagia and obesity on HFD exposure. On the other hand, even though DIO rats have reduced VMH amylin binding and leptin signaling, even before they become obese (29), they still respond to pharmacological doses of amylin alone, or together with leptin, by increasing VMH leptin signaling and reducing their food intake and body weight gain (35, 71, 80).

Although our laboratory previously showed that exogenous amylin activates VMH microglial IL-6 production, which then acts downstream of the IL-6 receptor complex to increase pSTAT3 expression, this action of exogenous amylin on IL-6 production is limited to the first few days of administration (35). This suggests that a possible “hit and run” string of epigenetic effects are set in motion, which prolong the responses to chronic amylin administration. Nevertheless, at least in the ARC, interfering with endogenous amylin signaling reduced IL-6 expression in our DR rats. Together, these findings suggest that endogenous amylin signaling is required to maintain a tonic level of IL-6 and IL-6-mediated enhancement of leptin signaling in the VMH, but, that the high endogenous amylin levels associated with obesity are insufficient to overcome the leptin resistance seen in obese individuals. Thus a form of amylin resistance at the level of the VMH may occur in obese individuals. This is possibly due to the fact that amylin-induced IL-6 signaling, like leptin signaling, stimulates pSTAT3 (9, 12, 22, 81) and the resultant expression of SOCS3, an endogenous inhibitor of leptin signaling (56, 57) that contributes to leptin resistance in obesity (28, 53). On the other hand, pharmacological doses of amylin may, by further stimulating IL-6 production, act to increase VMH leptin signaling, such that the enhancing effects of amylin on Lepr-b expression and leptin binding outweigh the inhibitory effects of SOCS3. Clearly, amylin or amylin receptor agonists remain effective in decreasing food intake and body weight gain in obese rodents (15, 71) and overweight/obese humans (18, 66), even in the face of persistent hyperamylinemia, with or without obesity (15).

Finally, although selectively bred DIO rats have reduced Lepr-b expression (39), leptin-induced pSTAT3 (13, 38, 58), and/or ¹²⁵I-labeled leptin binding in the VMH, these parameters do not differ between DIO and DR rats in the VTA (29) or NTS (38). This suggests that DIO rats have selective leptin resistance (65) and that chronic amylin administration might additionally act in concert with leptin on VTA (48) and AP/NTS neurons (55) to enhance reductions in food intake and body weight in obese individuals.

In summary, our results strongly support the hypothesis that endogenous amylin signaling via its CTR/RAMP receptor complex in the VMH is required for full leptin signaling and protection from becoming obese and insulin resistant on HFD. However, from the present studies, we cannot definitively determine whether amylin transported from the periphery into the brain or endogenous hypothalamic amylin or CGRP are the relevant ligands or whether RAMPs are required for such full expression of leptin signaling. Clearly, there are other sites in the brain where amylin acts, sometimes in concert with leptin, to reduce feeding and inhibit weight and adipose gain on such diets. However, the actions in the VMH appear to be mostly involved with energy expenditure and glucose homeostasis, since impairing VMH amylin signaling reduces leptin signaling and produces increased adiposity and insulin resistance, but not hyperphagia, on HFD. Thus the reduced amylin binding in the DMN/VMNdm of DIO rats may contribute to both their impaired leptin signaling and propensity to become obese on HFD.

GRANTS

This work was funded by the Research Service of the Veterans Administration (B. E. Levin, A. A. Dunn-Meynell), National Institute of Diabetes and Digestive and Kidney Diseases (NIDDK) Grant DK-RO1-30066 (B. E. Levin), Swiss National Science Foundation and the Center for Integrative Human Physiology (T. A. Lutz), and NIDDK Grant DK-096139 (M. R. Hayes).

DISCLOSURES

No conflicts of interest, financial or otherwise, are declared by the author(s).

AUTHOR CONTRIBUTIONS

Author contributions: A.A.D.-M. and B.E.L. conception and design of research; A.A.D.-M., C.L.F., and M.D.J. performed experiments; A.A.D.-M., C.L.F., M.D.J., T.A.L., M.R.H., and B.E.L. analyzed data; A.A.D.-M., C.L.F., T.A.L., M.R.H., and B.E.L. interpreted results of experiments; A.A.D.-M. drafted manuscript; A.A.D.-M., C.L.F., M.D.J., T.A.L., M.R.H., and B.E.L. edited and revised manuscript; A.A.D.-M., C.L.F., M.D.J., T.A.L., M.R.H., and B.E.L. approved final version of manuscript; C.L.F. and B.E.L. prepared figures.