INTRODUCTION

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

NUMEROUS VERTEBRATE SPECIES, held under natural conditions, undergo an annual cycle of metabolism characterized by transitions between the lean, insulin-sensitive and obese, insulin-resistant states (33). Seasonal fattening supports survival during long periods of low food availability. The annual selective increase in fat stores is not always accompanied by sustained hyperphagia, but rather with increased lipogenesis and hyperinsulinemia (8, 13, 33). Seasonal fattening is subsequently associated with increased mobilization and utilization of fat stores to fuel energy demands of migration, overwintering, or hibernation (33, 35, 51). Increased plasma free fatty acid (FFA) levels from increased adipose lipolysis promote insulin resistance in muscle and liver (6, 40). Reduced muscle utilization of increased hepatic glucose output maintains a requisite glucose supply to the central nervous system during these long periods of low food availability. Considerable evidence indicates that these orchestrated physiological events shifting peripheral glucose and lipid metabolism to induce the obese glucose-intolerant state (and support survival) are governed by central (hypothalamic) activities (14, 25).

Inasmuch as the insulin-resistant state is associated with increased insulin secretion (hyperinsulinemia), as well as increased hepatic glucose output and adipose lipolysis, we surmised that changes in the ventromedial hypothalamus (VMH), which regulates these three activities, might be operative in the development of the obese glucose-intolerant (metabolic) syndrome. Specifically in this regard, we have observed increased extracellular levels of the norepinephrine (NE) metabolite methoxy-hydroxy-propylglycol in the VMH of seasonally glucose-intolerant vs. glucose-tolerant hamsters, as measured via in vivo microdialysis (31). These findings suggest that increased VMH noradrenergic activities are associated with the glucose-intolerant condition. Furthermore, a variety of other nonseasonal (induced or genetic) animal models of the metabolic syndrome all have increased hypothalamic and/or VMH NE levels relative to their normal counterparts (15, 20, 29, 31, 37). Acute administration of NE in the VMH of normal rats induced a rapid increase in plasma glucose, FFA, insulin, and glucagon levels (9, 47, 49), all correlates of the metabolic syndrome.

We therefore reasoned that the increased endogenous VMH NE levels so commonly associated with the metabolic syndrome across animal models may participate in the development and maintenance of this condition. As such, we investigated the effects of chronic increases in VMH NE levels in normal rats on carbohydrate and lipid metabolism and related endocrine functions.

	MATERIALS AND METHODS

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

Experimental design. Female Sprague-Dawley rats (body weight 220 ± 10 g, 10 wk old) (Charles River) were allowed to feed and drink ad libitum and were maintained on 12:12-h light-dark daily photoperiods from birth. Female rats of this strain and age maintain the euinsulinemic lean condition for several months at least (43). Rats were infused continuously with either NE (25 nmol/h) or vehicle (10 mM sodium bisulfite isotonic saline solution) into the right VMH through osmotic minipumps placed subcutaneously. The NE dose was based on previous studies (9, 48) demonstrating hyperglycemic responses (20% increase) to acute VMH NE administration, hyperinsulinemic responses (4-fold increase) to such chronic administration, and preliminary dose-response studies done in our lab. Moreover, in a recent similarly designed study of hamsters, by our lab, 25 but not 5 nmol/h VMH NE infusion induced glucose intolerance (27). At the infusion rate of 25 nmol/h, the NE delivery per minute is about equivalent to the endogenous VMH tissue NE level. After 2 wk of NE infusion, the VMH tissue NE level increased 18-fold from 1.8 ± 0.1 to 33.3 ± 6 ng.

Surgery. Each animal was anesthetized with a mixture of ketamine, xylazine, and acepromazine (80:5:0.75 mg/kg body wt ip) and placed on a stereotaxic apparatus (David Kopf). A 33-gauge stainless steel cannula with an arm connected to an osmotic minipump was implanted into the right VMH with the following coordinates: 2.8 mm posterior to bregma, 0.7 mm right lateral to the midsagittal suture, and 9.2 mm ventral to the surface of the skull with the incisor bar set 2 mm below the interaural line. The cannula was secured permanently to the skull with three stainless steel screws penetrating the skull and acrylic cement. The osmotic minipump (model 2002; Alza) attached to the cannula and filled with sterile solutions of drugs being tested was placed subcutaneously on the back of the animal. NE was dissolved in sterile isotonic saline solution contained 10 mM sodium bisulfate as an antioxidant, and the solutions were passed through a membrane filter (0.2 µm; Gelman Sci. Acrodisc) for sterilization before use. The osmotic minipump ensures constant delivery of the test substance for 2 wk at the infusion rate of 0.5 µl/h. The minipumps were changed every 2 wk with animals lightly anesthetized with metofane. To test the stability of neurotransmitters in this condition, neurotransmitters were measured in the solution left in minipumps from hamsters after 14 days of implantation. The stability of the monoamines after 14 days was >75% as determined by high-performance liquid chromatography.

Histology. At the end of the experiments, animals were killed, and brains were removed and stored in 10% buffered neutral Formalin for 1 day. Serial coronal brain sections were cut on a cryostat at 50 µm. The sections were stained with hematoxylin according to standard histological procedures. Each histological section was observed under a microscope to verify the position of the infusion cannulas relative to the intended infusion sites. Rats with incorrect placement of the cannula (outside the VMH) were excluded from the study analyses. Twenty-micrometer sections were also cut and stained with hematoxylin to assess neuronal integrity within the VMH shell.

Study 1. Female rats were infused with NE (25 nmol/h) or vehicle into the right VMH for up to 14 days. At different time points after initiation of infusion, different subsets of animals from each treatment group (n = 6-8/group) were killed for the analyses of isolated adipocyte lipogenic and lipolytic activities, isolated islet insulin secretory response to glucose, and plasma glucose, insulin, glucagon, TG, and FFA levels.

Isolation of adipose cells. Adipose tissue was taken from retroperitoneal fat pads of individual rats for the determination of cell size as well as lipogenic and lipolytic rate. Adipose cells were isolated by a collagenase digestion method (46). Digestion was performed with 1 g of minced fat pad in 2 ml of 1 mg/ml collagenase for 45 min in 2.5% fraction V BSA in Krebs-Ringer phosphate buffer at 37°C in a shaking water bath. Cells were then filtered through nylon mesh. Cell diameter was measured microscopically (46), and lipid content was measured by lipid extraction. Cell size was expressed as micrograms lipid per cell. A 10% suspension of isolated adipocytes was prepared in Krebs-Ringer solution buffered with 20 mM HEPES, pH 7.4, containing 200 nM adenosine, 2 mM glucose, and 2.5% BSA (media).

Lipolysis in isolated adipose cells. Cell incubations were initiated by introducing 100 µl of the adipocyte suspension into incubation vials containing 400 µl of media with 1 U/ml adenosine deaminase (Sigma, St. Louis, MO) and lipolytic stimulator (isoproterenol). Incubations were carried out at 37°C in a shaking water bath for 20 min. Lipolysis was determined by spinning cells through dinonyl phthalate oil. The glycerol content in 150-µl aliquots of cell free media was measured by using a colorimetric enzymatic assay kit (Sigma). Lipolysis was quantified by calculating the release of glycerol from adipocytes in the presence or absence of isoproterenol (expressed as pmol · h¹ · cell¹).

Lipogenesis in isolated adipose cells. Lipogenesis was assayed by measuring TG production after the conversion of D-[U-¹⁴C]glucose to the total [¹⁴C]TG pool. Cell incubations were initiated by introducing 200 µl of the adipocyte suspension into incubation vials containing 800 µl of media with 1 U/ml adenosine deaminase and 2 mM glucose with 1 µCi/ml D-[U-¹⁴C]glucose. Adipocytes were incubated under 95% O₂-5% CO₂ for 60 min at 37°C in a metabolic shaker at a rate of 80 strokes/min. Incubations were terminated by decanting the contents of the incubation vials into glass screw-cap tubes containing 5 ml of extraction mixture of Dole solution (10). Samples were held in the extraction mixture overnight at room temperature to ensure complete recovery of [¹⁴C]TG. Aliquots were transferred to 20-ml glass scintillation vials to which 10 ml of scintillatin fluid were added and were counted in a liquid scintillation counter. Experimental blanks contained D-[U-¹⁴C]glucose but no adipocytes. Results were normalized for D-[U-¹⁴C]glucose specific activity and are expressed as picomoles glucose incorporation into lipids per hour per adipocyte.

Study of insulin release in vitro. Islets were isolated from pancreas using a collagenase digestion and Ficoll gradient centrifugation as previously described (24). Isolated islets were preincubated in Krebs-Ringer buffer for 20 min, and then insulin release stimulated by various glucose concentrations was tested by static incubation as described previously (24).

Study 2 . On the basis of the findings of study 1, a second similar study was conducted to investigate the relation between body fat, whole body resiratory quotient (RQ), and plasma leptin levels in VMH NE-infused rats. Rats were infused continuously into the VMH with NE (25 nmol/h) or vehicle (n = 8/group) for up to 23 days. Different subsets of rats from each treatment group were assayed for body fat content, RQ, and plasma leptin concentrations during the course of the infusion period.

Indirect calorimetry measurement. Whole body RQ was measured by indirect calorimetry. Animals were held in air-tight Oxymax chambers (Columbus Instrument, Columbus, OH) at 18 ± 2°C with an oxygen flow rate of 0.85 l/min. Animals were allowed to adapt to the metabolic cages for 30 min before the initiation of measurements. Oxygen consumption (O₂) and CO₂ production (CO₂) were measured during six 30-s sampling periods. During measurements, animals had no access to food and water. Energy expenditure (EE) was calculated according to the Weir equation: kJ/min= 16.32 × O₂ + 4.6 × CO₂. The measurements of metabolic activity were carried out at a daily time period (at 4-6 h after light onset) that was the same as that when plasma and tissue samples were collected at the termination of the experiments.

Study 3. On the basis of these previous findings, a final study was conducted to assess the longer term influence of VMH NE infusion on glucose tolerance. Rats were infused for 35 days with NE (25 nmol/h) or vehicle into the right VMH. After 28 days, rats were subjected to a glucose tolerance test (GTT). Forty-eight hours before the GTT, rats were anesthetized, and the carotid artery was cannulated. Awake rats were injected intravenously with a 50% glucose solution (1.5 g/kg body wt) at 4 h after light onset. Blood samples were taken from the carotid before and at 7, 15, 30, and 45 min after the glucose injection. Blood glucose concentrations were determined immediately by a blood glucose monitor.

Assay of blood samples. Blood glucose concentrations were determined by a blood glucose monitor (Accu-Chek Advantage, Boehringer, Indianapolis, IN). Plasma TG concentration was measured enzymatically using Sigma Diagnostics kit. Plasma FFA concentration was determined by enzymatic colorimetric assay utilizing acyl-CoA synthetase and acyl-CoA oxidase coupled to peroxidase (Wako Chemicals, Richmond, VA). Plasma insulin, glucagon, and leptin were assayed by RIA using commercially available assay kits utilizing rabbit ani-rat serum and rat insulin, glucagon, and leptin as standards (Linco Research, St. Charles, MO). Plasma NE and Epi were assayed by RIA using commercially available assay kits utilizing goat anti-rabbit serum and rabbit NE and Epi as standard (American Laboratory Products, Windham, NH).

Statistical analysis . All data are expressed as means ± SE. Significant differences among treatments were determined by t-tests, two-way ANOVA with treatment and time of the day as the two main factors, or one-way ANOVA with repeated measurement or without, followed by a multiple-range test (Newman-Keuls) as appropriate. The null hypotheses of no difference between treatments were rejected at P < 0.05.

	RESULTS

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

Although physical damage was imposed along the cannula track, histological analysis of the right VMH after 14 or 35 days of NE infusion revealed few fragmented or pyknotic nuclei or obviously damaged neurons within the VMH shell and outside the cannula track, similar to previous reports (27, 48). Data from rats wherein the cannula placement was outside the VMH were not used in any analyses. Indeed, we observed that the typical metabolic response to VMH NE infusion was markedly attenuated if placement of the cannula was outside of the VMH (data not shown, Ref. 27). VMH NE infusion induced only a transient (30%) increase in food consumption, persisting for only 5 days [overall, there were treatment (P < 0.001), time (P < 0.01), and interaction (P < 0.001) effects, Fig. 1A]. However, VMH NE infusion induced a rapid (after 4 days, P < 0.01) and marked increase in retroperitoneal fat weight that averages more than threefold greater than vehicle controls after 14 days of such treatment (P < 0.01) (treatment effect: P < 0.0001 by ANOVA) without affecting total body weight change (Fig. 1, B and C). Intrascapular brown fat weight, measured only at day 14 of infusion, was also increased by treatment from 80 ± 19 mg in controls to 273 ± 33 mg (P < 0.01). Figure 2 shows a representative histological comparison of brown fat in vehicle-treated and NE-treated rats. NE-infused rats had a significant increase in fat droplet deposition in these adipocytes.

View larger version (16K): [in this window] [in a new window]   Fig. 1.   Effects of norepinephrine (NE; ) or vehicle () infusions into the unilateral ventromedial hypothalamus (VMH) on daily food consumption (A), body weight (B), and retroperitoneal fat weight (C) in female Sprague-Dawley rats. Values are means ± SE of n = 6 or 8 rats/group. * P < 0.05 and ** P < 0.01 at least, vs. respective vehicle treatment value (see RESULTS for ANOVA).

View larger version (167K): [in this window] [in a new window]   Fig. 2.   Histological comparison of brown fat in vehicle-treated (A) and NE-treated (B) rats. Representative fields of paraformaldehyde-fixed, routinely processed and paraffin-embedded hematoxylin and eosin-stained sections. Note the significant increase in fat droplet deposition of adipocytes in the NE-treated tissue. Scale bar represents 20 µm.

The increased body fat store content of VMH-treated animals was associated with a parallel (~3-fold) increase in adipocyte cell size (treatment effect: P < 0.0001 by ANOVA, Fig. 3A) and lipogenic activity (treatment effect: P < 0.0001, Fig. 3B) over time. Again, these effects were demonstrated as early as 4 days after treatment initiation (P < 0.05). In agreement with increased lipogenic activity, plasma TG was significantly increased (2-fold) after 4 days of infusion and was sustained at this level until day 14 (treatment effect: P < 0.0001). However, plasma FFA was significantly reduced (P < 0.01, Fig. 4, A and B), which may reflect the hepatic reesterification to TG. Serum insulin levels, which may be expected to support lipogenesis, also increased relative to controls as early as 2 days after the start of VMH NE infusion (P < 0.05, Fig. 5) and reached a maximum at day 14 (3-fold increase, P < 0.01) (treatment effect: P < 0.0001 by ANOVA). In agreement with the hyperinsulinemia observed in vivo, VMH NE infusion induced an increase in isolated islet insulin secretory responses to glucose by 50-100% (P < 0.05, Table 1).

View larger version (19K): [in this window] [in a new window]   Fig. 3.   Effects of NE () or vehicle () infusions into the unilateral VMH on adipocyte cell size (A) and lipogenesis in isolated adipocytes (B) of female Sprague-Dawley rats. Lipogenesis was assayed by measuring the incorporation of [14C]glucose into total lipids. Values are means ± SE of n = 6 or 8 rats/group. * P < 0.05 and ** P < 0.01 at least, vs. respective vehicle treatment value (see RESULTS for ANOVA).

View larger version (18K): [in this window] [in a new window]   Fig. 4.   Effects of NE () or vehicle () infusions into the unilateral VMH on plasma free fatty acids (FFA; A) and triglyceride (B) levels of female Sprague-Dawley rats. Values are means ± SE of n = 6 or 8 rats/group. * P < 0.05 and ** P < 0.01 at least, vs. respective vehicle treatment value (see RESULTS for ANOVA).

View larger version (13K): [in this window] [in a new window]   Fig. 5.   Effects of NE () or vehicle () infusions into the unilateral VMH on plasma glucose (A), insulin (B), and glucagon (C) levels of female Sprague-Dawley rats. Values are means ± SE of n = 6 or 8 rats/group. * P < 0.05 and ** P < 0.01 at least, vs. respective vehicle treatment value (see RESULTS for ANOVA).

VMH NE infusion induced progressive increases in plasma glucagon levels over the 14-day treatment period (treatment effect: P < 0.0005, Fig. 5). After 2 days, glucagon levels were 40% greater in treated vs. control rats (P < 0.01), which rose to 50% greater at 14 days of treatment. However, plasma glucose did not change during the experiment (Fig. 5). Furthermore, after 14 days of treatment, plasma NE and Epi concentrations were increased 176 and 53%, respectively, relative to control animals (P < 0.001 and P < 0.05, respectively, Fig. 6).

View larger version (12K): [in this window] [in a new window]   Fig. 6.   Effects of NE (solid bars) or vehicle (open bars) infusions into the unilateral VMH for 14 days on plasma NE (A) and epinephrine (B) levels of female Sprague-Dawley rats. Values are means ± SE of n = 6 or 8 rats/group. * P < 0.05 and ** P < 0.01 at least, vs. respective vehicle treatment value.

VMH NE infusion also resulted in rapid and sustained increases in basal (P < 0.001 by ANOVA) and isoproterenol-stimulated lipolysis (P < 0.0001 by ANOVA) in isolated adipocytes concurrent with the increases in lipogenesis (Fig. 7). Two- to threefold increases in basal (P < 0.05) and isoproterenol-stimulated (P < 0.01) lipolysis were observed as early as 4 days after treatment initiation and were maintained thereafter to 14 days of treatment (P < 0.05, Fig. 7).

View larger version (14K): [in this window] [in a new window]   Fig. 7.   A: effects of NE () or vehicle () infusions into the unilateral VMH on basal lipolysis. B: isoproterenol-stimulated lipolysis in isolated adipocytes of female Sprague-Dawley rats treated with VMH NE for 2 (), 4 (), 9 (), or 14 () days or vehicle (). Values are means ± SE of n = 6 or 8 rats/group. * P < 0.05 and ** P < 0.01 at least, vs. respective vehicle treatment value (see RESULTS for ANOVA).

On the basis of the above findings, a second study was conducted to investigate the relation between body fat, RQ, and plasma leptin levels in VMH NE-infused animals. The results of study 2 indicate that VMH NE infusion induced immediate increases in whole body RQ relative to control (treatment effect: P < 0.001; time effect: P < 0.005 by ANOVA) which was above 1.0 at 1-9 days of infusion, while induced increases in body fat plateaued at day 14 (P < 0.0001; ANOVA). Moreover, such sustained increases in body fat were associated with parallel sustained four- to sevenfold increases in circulating leptin (treatment effect: P < 0.0001 by ANOVA) (Fig. 8, A-C).

View larger version (14K): [in this window] [in a new window]   Fig. 8.   Effects of NE () or vehicle () infusions into the unilateral VMH on plasma leptin (A), retroperitoneal fat pad weight (B), and respiratory quotient (RQ; C) of female Sprague-Dawley rats. Values are means ± SE of n = 6 or 8 rats/group. * P < 0.05 and ** P < 0.01 at least, vs. respective vehicle treatment value (see RESULTS for ANOVA).

A third study examined the effect of long-term (4-5 wk) VMH NE infusion on glucose tolerance and metabolism. After 4 wk of treatment, VMH NE infusion induced glucose intolerance in rats (Fig. 9). The area under the GTT curve was increased from 11,450 ± 702 min · mg¹ · dl¹ in controls to 15,870 ± 1,782 min · mg¹ · dl¹ in NE-treated animals (P < 0.01). After 5 wk of NE infusion, EE and heat production were significantly reduced relative to controls (~17%, P < 0.01), whereas plasma FFA, insulin, and fat pad were increased 33, 69, and 174%, respectively (P < 0.01). Furthermore, RQ values were reduced from 0.903 ± 0.01 to 0.873 ± 0.02 by treatment (P < 0.05) (Table 2).

View larger version (15K): [in this window] [in a new window]   Fig. 9.   Effects of NE () or vehicle () infusions into the unilateral VMH on plasma glucose during a glucose tolerance test. Animals were administered glucose (1.5 g/kg body wt iv) at 4 h after light onset after 5 wk of infusion, and glucose levels were determined over the next 45 min. Values are means ± SE of n = 6 or 8 rats/group. NE infusion increased the area under the glucose tolerance test curve relative to vehicle infusion (P < 0.05, t-test).

	DISCUSSION

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

Chronic increases in the unilateral VMH NE level orchestrate widespread neuroendocrine and biochemical changes that induce the major physiological aspects of the obese glucose-intolerant (metabolic) syndrome (16, 23, 38, 44). The present time course study of responses to such treatment has unveiled the sequence of several metabolic events involved in the genesis of this metabolic syndrome. Early and sustained concurrent increases in lipogenesis and lipolysis potentiate increases in body fat store levels and serum TG, respectively, for ~14 days. This period is followed by a transition phase wherein fat oxidation increases, which may facilitate the curtailment of further fat accretion, and a new metabolic steady state is established characterized by increased body adiposity, increased whole body fat oxidation, and glucose intolerance. The neuroendocrine and physiological events delineated in this study that are involved in this metabolic metamorphosis, when taken in the context of other available information, may offer new insights into the neuroendocrine organization of metabolism as follows.

First, the progressive increases in body fat stores induced by treatment are not accompanied by sustained increases in food consumption. The VMH NE-stimulated increase in fat stores is also not associated with any change in body weight. These findings indicate a repartitioning of body composition in response to treatment, which favors an increase in fat-to-lean mass ratio; that is, exogenous and endogenous metabolic substrates are shifted toward lipid synthesis and accretion. This result correlates well with observations that pharmacological intervention that reduces elevated VMH NE metabolism in obese, insulin-resistant hamsters decreases lipogenesis and fattening without altering food consumption and increases conservation of body protein (i.e., decreases amino acid deamination) (8, 31).

Treatment-induced increases in fattening are associated with increased lipogenesis as evidenced by increased in vitro adipocyte conversion of glucose to lipid and in vivo whole body RQ values >1.0. Treatment-induced increases in adiposity correlate with and are preceded by greater than twofold increases in plasma insulin levels, which may in part potentiate adipose lipogenesis. The increase in plasma insulin levels in turn correlates well with an increased pancreatic islet insulin secretory response to glucose from treated vs. control animals. A similar islet response to chronic VMH NE infusion has been observed in hamsters and may involve a desensitization of islet responsiveness to the inhibitory effects of local NE (25). VMH NE may also activate vagal efferents to the pancreatic islets to stimulate insulin release (9, 49).

The effects of VMH NE infusion to increase brown adipose tissue fat mass are in accordance with and may have a mechanistic basis in previous studies demonstrating that 1) glutamate stimulation of VMH neurons selectively increases sympathetic outflow to brown fat to increase lipid oxidation (1, 2, 50) and 2) iontophoretic application of NE to the VMH inhibits glutamate-stimulated neuronal activity and this NE effect is more pronounced in obese vs. lean animals (22). Therefore, VMH NE may selectively block VMH sympathetic stimulation of brown fat lipid oxidation and thermogenesis. However, other aspects of the sympathetic nervous system may actually be stimulated by VMH NE (see below).

For example, in white adipose tissue, the increase in lipogenic activity is associated with an increase in basal and isoproterenol-stimulated lipolysis that plateaus early (day 4) after treatment initiation. The increase in lipolysis may be attributable, in part, to increased cell size; however, other neuroendocrine factors may be operative. The rise in serum glucagon levels precedes the onset of increased lipolysis in response to treatment, and at day 14 of treatment, the plasma NE and E concentrations are both substantially elevated, which collectively would potentiate increases in lipolysis. Second, the isoproterenol response of adipocytes from treated rats is substantially enhanced. Therefore, VMH NE infusion induces 1) neuroendocrine correlates of increased sympathetic tone (i.e., increased circulating levels of glucagon, NE, and Epi), possibly via connections to the paraventricular nucleus (PVN) (19, 26), as well as 2) increased responsiveness of white adipocytes to sympathetic stimuli that interact to increase lipolytic activity. It therefore appears from available evidence that chronic VMH NE treatment may reduce sympathetic drive to brown fat and increase it to white fat. Despite increases in white adipose lipolytic activity during the first 14 days of treatment, plasma FFA levels are not elevated but actually reduced. An RQ of 1.0 or greater among treated animals during this time period suggests that FFA are more reesterified than oxidized. This postulate is further supported by the immediate and sustained twofold increase in plasma TG during this period. However, by day 35 of treatment, RQ values of treated rats (0.87) are now less than those of controls and indicate an increase in net lipid oxidation relative to days 1-14 of treatment (RQ of 1.1-1.0). Furthermore, the plasma FFA levels at day 35 are now increased relative to controls. This transition to increased lipid oxidation may explain the plateau in lipid accretion observed after day 14 of treatment. As such, by day 35, a new steady state has been established wherein fat stores are increased in response to preceding neuroendocrine changes and are maintained by a balance between increased lipid synthesis and oxidation. For reasons mentioned above, it is unlikely (though not certain) that this increase in fat oxidation is predominately from brown fat. Also, reduced EE and heat production at day 35 of treatment support this hypothesis.

Increases in plasma FFA levels and lipid oxidation are known to potentiate insulin resistance in both muscle and liver (6, 11, 40). Hyperinsulinemia and increased tissue adiposity also contribute to insulin resistance and glucose intolerance (11, 21, 39). After 28 days of treatment, glucose tolerance was impaired (Fig. 9); however, when measured at 14 days of treatment, no effect on glucose tolerance was observed (data not shown). The induction of the hyperinsulinemic-euglycemic condition early after treatment initiation (2 days) and its maintenance thereafter indicates the presence of insulin resistance (4, 12), which normally precedes the glucose-intolerant state (4, 12, 44) as observed in this study. Insulin clamp studies at different days after VMH NE infusion are required to fully delineate the time course of transition from the normal to insulin-resistant state. The observed subsequent glucose intolerance may be the ultimate cumulative effect of the insulin resistance, fattening, increased hepatic glucose output, and/or neuroendocrine alterations induced by VMH NE infusion. Acute administration of NE to the VMH is known to stimulate an immediate increase in plasma glucose (9, 47, 49), which may be maintained by chronic increases in circulating glucagon, NE, and Epi as demonstrated herein. In this environment, increases in insulin resistance will potentiate glucose intolerance.

The VMH is a primary site of leptin action to reduce weight gain and increase glucose disposal in muscle (18, 34). It is therefore particularly interesting that NE infusion to the VMH, which induces the obese, glucose-intolerant state also induces hyperleptinemia and apparent leptin resistance. The marked early and sustained increases in plasma leptin of treated rats do not prevent either the subsequent fattening or glucose intolerance. VMH NE may block the action of leptin at this site, and preliminary evidence from our lab suggests this may be the case.

Notwithstanding the above discussion, it must be appreciated that the neurophysiological and anatomic aspects involved in this widespread response to VMH NE require further study. In this regard, however, a few salient points are worthy of mention. The body composition response to unilateral VMH NE is the same as that to bilateral VMH electrolytic lesions which remain within the borders of the nucleus (i.e., increased adiposity without hyperphagia or weight gain) (41). We therefore considered the possibility that VMH NE administration was damaging the nucleus, and this was responsible for the observed metabolic responses. However, histological hematoxylin and eosin staining of brain tissue revealed no major damage to VMH neurons at the unilateral site of administration or of the contralateral VMH. Furthermore, we conducted a study wherein radiofrequency lesions were made to either the unilateral or bilateral ventromedial nucleus of rats, and metabolic responses were monitored for 2 wk thereafter. Compared with sham-operated controls, bilateral VMH-lesioned rats had increased (by 56%) retroperitoneal fat weight (P < 0.002). However, adiposity of unilateral VMH-lesioned animals was the same as sham controls. This result is similar to a previously reported study of unilateral vs. bilateral VMH electrolytic-lesioned rats (17). Therefore, it is unlikely that the effects of chronic unilateral VMH NE infusion result from NE-induced destruction of these neurons. In a similar study by Shimazu et al. (48) as described herein, although neither neuroendocrine nor metabolic biochemical activities were measured, similar increases in plasma insulin and brown fat weight were observed in response to chronic VMH NE infusion. However, these similar increases were associated with disparate increases in feeding and body weight gain, possibly due to differences in genetic background of this Japanese rat strain.

The in vivo ionotophoretic application of NE to the VMH inhibits neuronal activity therein (22), and this may be mediated via GABA release (5). Moreover, this inhibitory responsiveness to NE is more pronounced in obese vs. lean mice (22), and this increased responsiveness may be the result of a concurrent increase in noradrenergic postsynaptic ₁- and -receptors and decrease in _2A (likely presynaptic)-receptors within the VMH (7). These findings taken together with consistent observations of increased levels or metabolism of NE within the VMH of obese, glucose-intolerant vs. lean, glucose-tolerant animals (15, 20, 29, 31, 37) reveal a neurophysiological situation wherein endogenous VMH NE activities may be amplified in the obese, glucose-intolerant state.

Perspectives