INTRODUCTION

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

UROCORTIN (UCN) is a recently identified peptide with 45% sequence identity to corticotropin-releasing hormone (CRH) (39). High-affinity binding of UCN to the CRH-R2 receptor has led to the hypothesis that UCN may be the endogenous ligand for the CRH-R2 receptor. CRH receptors are widely distributed throughout the CNS (17). Within the hypothalamic paraventricular nucleus (PVN), UCN-immunoreactivity (UCN-ir) has been demonstrated both in perikarya and fibers (22), and the predominant CRH receptor subtype in the PVN is CRH-R2 (18), providing a neuroanatomic basis for UCN action in the PVN. Intracerebroventricular and PVN injection of UCN potently and dose dependently decreases feeding in food-deprived, free-feeding, and neuropeptide Y (NPY)-injected rats (37, 40). Food deprivation results in decreased UCN fibers and varicosities in the PVN (22), which is consistent with anorectic properties of UCN. Together these data suggest an important role for UCN in PVN circuitry regulating feeding.

The PVN is a major source of autonomic outflow (29), and stimuli that influence feeding within this area often result in changes in sympathetic nerve firing rate to brown adipose tissue (BAT) (10). NPY and cocaine-amphetamine-related transcript (CART), which affect food intake after site-specific brain stimulation, produce complementary effects on uncoupling protein (UCP) gene expression in BAT (5, 40). A proposed PVN-BAT signaling pathway has been demonstrated by use of the transneuronal viral-labeling technique (2), which provides a basis for effects on UCP in BAT after PVN manipulation.

UCP-2 and -3 are newly identified UCPs present in BAT, although the highest concentrations of UCP-2 and -3 are found in white adipose and muscle tissue, respectively (9, 19). UCPs may mediate thermogenesis and fuel partitioning, and thus alterations in the activity of these proteins may influence energy expenditure and metabolism (6, 7). However, knowledge of the brain circuitry involved in the regulation of these UCPs is sparse. We recently demonstrated that whereas PVN injection of NPY, which is orexigenic, has differential effects on the three UCPs (28), PVN injection of CART, which is anorectic, increases gene expression of all three UCPs (40). Thus the inverse relationship between direction of change in feeding and UCP-1 response after brain stimulation with appetite-modulating compounds (11) does not pertain consistently to UCP-2 and -3 and may reflect functional differences in these UCPs.

The purpose of the current studies was to determine the effect of UCN administration in the PVN on energy metabolism. On the basis of the anorectic effect of PVN UCN administration (40) and the central role of the PVN in energy metabolism, we hypothesized that anorectic response to UCN in the PVN may be accompanied by simultaneous alterations in gene expression of UCPs.

	MATERIALS AND METHODS

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

Experimental design. Twenty-five PVN-cannulated rats were injected at 0800, 2000, and again at 0800 within a 24-h treatment period (see Table 1). This injection scheme is based on previous studies of PVN NPY stimulation effects on the UCPs (28), in which we injected NPY every 6 h for 24 h. In this study, we reduced the number of injections from five to three within 24 h on the basis of the observation that UCN anorectic effects last longer than the feeding effects of NPY. Animals were killed 2 h (at 1000) after the final set of injections. The treatments were as follows: 1) artificial cerebrospinal fluid (aCSF) with food allowed ad libitum; 2) 200 pmol UCN with food allowed ad libitum; and 3) aCSF with food intake matched (pair fed) to the level of the UCN-treated rats. The UCN dose represents an amount that we estimated would significantly and reliably reduce feeding for up to 8 h after injection. The half-life of UCN within the central nervous system (CNS) is unknown. However, previous studies in our laboratory indicate that 100 pmol UCN significantly inhibits feeding (deprivation, nocturnal and NPY-induced) for up to 4 h after injection, and in one study, 100 pmol UCN in the PVN did not affect feeding after 4 h (41). To enhance the probability that UCN effects would last for the 8-h interval between injections in the current study, we increased the dose to 200 pmol. This experiment was conducted over 2 days, with UCN treatment and aCSF ad libitum treatment represented as equally as possible over the 2 days. The aCSF pair-fed group was necessarily included entirely on the 2nd day because we needed food-intake information from UCN-treated animals on the 1st day before we could match this level of food intake appropriately. The data from the 2 days were combined after determining that there was no effect of day on 24-h feeding response (by ANOVA: F_1,19 = 1.222, P = 0.2827) in the aCSF ad libitum control groups.

Animals. In all experiments, male Sprague-Dawley rats (Harlan, Madison, WI, or Charles River, Portage, MI) weighing 250-350 g were individually housed in conventional hanging cages with a 12:12-h light-dark photoperiod (lights on at 0700) in a temperature-controlled room (21-22°C). Teklad-verified lab chow and water were allowed ad libitum, except where noted. We received approval from the local Institutional Animal Care and Use Committee before performing these studies.

Cannulation. Rats were anesthetized with Nembutal (40 mg/kg) and fitted with 26-gauge stainless steel guide cannulas (Plastics One, Roanoke, VA) placed 1 mm above the PVN target for all experiments. Stereotaxic coordinates were determined from the rat brain atlas by Paxinos and Watson (32) and were as follows for the PVN: 0.5 mm lateral, 1.9 mm posterior to bregma, and 7.3 mm below the skull surface. For all cannulations, the incisor bar was set at 3.3 mm below the ear bars. At least 7 days elapsed after surgery before experimental trials.

Injections. Injections into the PVN were given in a 0.5-µl volume over 45 s by the use of a 33-gauge internal cannula (Plastics One). The injector extended 1 mm beyond the end of the guide cannula.

Verification of cannula placement. After the experiments, brains were dissected out and stored in a 10% formaldehyde solution for later placement verification by histological examination. Data from four animals (3 in the aCSF group and 1 in the UCN group) with incorrectly placed cannulas were excluded from the final analysis.

Drugs. UCN was purchased from Phoenix Pharmaceuticals (Mountain View, CA). All drugs were dissolved in aCSF just before use.

Food intake measurements. Food was allowed ad libitum in hoppers. Just before injection, food was removed from the hoppers. Immediately after injection, preweighed pellets of chow were placed inside the rat cage. At 12 and 24 h, pellets and spillage were weighed and subtracted from the initial weight to quantify the amount of food eaten. Animals with food intake restricted to the level of that in the UCN-treated animals were given food amounts estimated from that observed in the UCN-treated, ad libitum-fed rats. An additional 1-2 g of chow was then added to this amount to account for assumed spillage, and the total amount was given to these animals.

Leptin radioimmunoassay. Trunk blood samples were centrifuged for 20 min at 2,000 g, and sera were stored at 4°C until use. On the day of the RIA, samples were slowly thawed, and 100 µl sera was added to the RIA tube for use in the rat leptin RIA kit (Linco Research, St. Louis, MO). Standard concentrations ranged between 0.5 and 50 ng/ml, which is within the limit of linearity. All samples were within this range. The assay sensitivity is 0.5 ng/ml (100-µl sample size). The cross-reactivity test, provided by Linco Research, indicated no cross-reaction with insulin, glucagon, or somatostatin release inhibitory factor (SRIF).

UCP-1, UCP-2, UCP-3, and -actin mRNA determination. Rats were killed by decapitation 2 h after the final set of injections, and tissues were taken for gene expression measurements. Interscapular brown fat was dissected free from surrounding tissue. Acromiotrapezius and biceps femoris muscle and epididymal white fat were dissected out. Total RNA from all tissues was extracted by the rapid guanidine thiocyanate-phenol-chloroform method (14). Tissue was homogenized in a buffer containing 4 M guanidine thiocyanate with added -mercaptoethanol and water-saturated molecular biology-grade phenol. Sarcosyl, 2 M sodium acetate, and chloroform were then added. After centrifugation, the aqueous phase was precipitated with isopropanol, resuspended in guanidine thiocyanate buffer, and reprecipitated with isopropanol. The pellet was washed with 75% ethanol. The resulting RNA was stored in 100% ethanol at 80°C.

1

1

Statistical analysis. Four rats were removed from the analysis because of misplaced cannulas, and thus data from 21 animals remained in the analysis. Data were analyzed by a one-factor ANOVA followed by Fisher's least significant difference t-test to compare means. Where indicated, regression analysis was performed.

	RESULTS

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

Food intake. There was a main effect of treatment on food intake in the 12- to 24-h interval and on cumulative food intake from 0-24 h (12-24 h: F_2,18 = 8.815, P = 0.0021; 0-24 h: F_2,18 = 8.815, P = 0.0021; Fig. 1). In the 0- to 12-h interval, the effect of treatment on food intake approached significance (F_2,19 = 3.3293, P = 0.0589; Fig. 1). In the 12- to 24-h interval, and at 24 h after injection, intake of the UCN-treated and pair-fed rats was significantly lower than controls [12-24 h: P = 0.0041 and P = 0.0010, respectively (Fig. 1); 0-24 h: P = 0.0026 and P = 0.0009, respectively (Fig. 1)].

View larger version (12K): [in this window] [in a new window]   Fig. 1.   Effect of paraventricular (PVN) urocortin (UCN) administration on feeding at 12 and 24 h. * Significant difference from control [artificial cerebrospinal fluid (aCSF)] value, P < 0.001.

Gene expression of UCP-1, UCP-2, and UCP-3. In BAT, there was a main effect of treatment on UCP-1 gene expression (F_2,21 = 3.708, P = 0.0461; Fig. 2A). Post hoc analysis indicates no difference between the aCSF-injected, ad libitum-fed rats and the UCN-treated rats but a significant UCN-induced increase in UCP-1 mRNA above the levels observed for the aCSF-treated pair-fed rats (P = 0.0236; Fig. 2A). Food intake between UCN-treated and aCSF pair-fed rats was not significantly different (P = 0.6629; Fig. 1) over the 24-h experimental period (UCN treated, 12.1 g; aCSF pair fed, 10.8 g; Fig. 1). Thus the comparison of UCP gene expression data between these two groups is most appropriate because the only difference in treatment was UCN administration. There were no main effects of treatment on UCP-2 gene expression in BAT or WAT (F_2,17 = 0.667, P = 0.5263 and F_2,17 = 0.911, P = 0.4210, respectively; Fig. 2, B and C). In the biceps femoris muscle, there was no main effect of treatment on UCP-3 gene expression (F_2,16 = 0.531, P = 0.5980; Fig. 2D). In the acromiotrapezius muscle, there was a main effect of treatment on UCP-3 gene expression (F_2,17 = 5.203, P = 0.0173; Fig. 2E). Post hoc analysis indicates that UCN significantly decreased UCP-3 gene expression relative to pair-fed aCSF controls (P = 0.0246; Fig. 2E). Similar to the UCP-1 gene expression data, there was no difference in UCP-3 mRNA levels between UCN-treated animals and the aCSF-injected ad libitum-fed animals.

View larger version (31K): [in this window] [in a new window]   Fig. 2.   Effect of PVN UCN administration on uncoupling protein-1 (UCP-1; A), UCP-2 (B and C), and UCP-3 (D and E) gene expression in brown adipose tissue (BAT; A and B), white adipose tissue (WAT; C), and muscle tissues [biceps femoris muscle (BFM; D) and acromiotrapezius muscle (TM; E)]. * Significant difference from control (aCSF) and UCN-alone values, P < 0.03.

Plasma leptin. There was a main effect of treatment on plasma leptin levels (F_2,17 = 7.994, P = 0.0036; Fig. 3). UCN induced a significant increase in leptin levels compared with pair-fed aCSF controls (P = 0.0228; Fig. 3). Leptin levels in the UCN-treated animals were not significantly different from those of the aCSF-treated ad libitum-fed controls. Regression analysis indicates that plasma leptin levels are significantly correlated with 24-h food intake (R = 0.622, F_1,19 = 8.864; P = 0.0034). If data from the UCN-treated animals are removed from this analysis (so that the only independent variable is food intake and not drug treatment), the correlation coefficient improves (R = 0.854, F_1,11 = 26.961; P = 0.0004).

View larger version (51K): [in this window] [in a new window]   Fig. 3.   Effect of PVN UCN administration on plasma leptin. Columns with differing superscripts (a, b) indicate that representative values are significantly different from each other (P < 0.03)

	DISCUSSION

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

The present data demonstrate for the first time that UCN in the PVN influences gene expression of UCP-1 in BAT, UCP-3 in muscle, and plasma leptin. Short-term fasting results in decreased gene expression of UCP-1 in BAT (13) and increased gene expression of UCP-3 in muscle (8). As shown in Fig. 2, our data concur with those results. In food-restricted aCSF-injected rats, BAT UCP-1 gene expression was decreased (Fig. 2A) and UCP-3 gene expression in acromiotrapezius muscle was increased (Fig. 2E) relative to that of ad libitum-fed controls. Administration of UCN into the PVN reversed these conditions: in both instances, UCN restored UCP gene expression values to control levels (Fig. 2, A and E). The lack of difference from controls with ad libitum access to food indicates that UCN effects on UCP gene expression are state dependent, and one state where it appears to be active is in food restriction. Food restriction in the current study was relatively mild: the food-restricted animals were given the same amount of food eaten by the UCN-treated animals, which represented ~50% of normal 24-h food intake. This was done to ensure that differences observed were not due to metabolic changes induced by decreased food consumption in UCN-treated animals. The results observed suggest that UCN may act to reverse metabolic and sympathetic nervous system (SNS) conditions existing during food restriction, such as decreased plasma leptin and insulin (16, 30, 33) and decreased SNS activity (33).

Induction of UCP-1 in BAT by UCN in the PVN is consistent with the hypothesis that feeding and thermogenic properties of centrally administered neuropeptides are inversely related (10, 11) and result in whole body anabolic or catabolic responses, depending on the stimulus delivered. The current data indicate that during food restriction, UCN in the PVN can increase the capacity for thermogenesis by elevating UCP-1 levels in BAT (Fig. 2A). UCN increases SNS outflow (11, 37), and thus the induction of UCP-1 in BAT during food restriction (Fig. 1A) may be via this mechanism. The effect of UCN in PVN on UCP-1 in BAT is similar to the effect of CART in the PVN, which increases gene expression of UCP-1 in BAT (40). However, in that study, CART increased expression of UCP-2 in WAT and UCP-3 in muscle, which is inconsistent with the observed UCN effects (Fig. 2, C-E) in the current study.

The increase of UCP-3 in acromiotrapezius muscle in response to food restriction in the current study (Fig. 2E) is consistent with the findings of others (8). It is hypothesized that increases in muscle UCP-3 during fasting or food restriction may be due to elevated free fatty acids, which stimulate UCP-3 gene expression (43). UCN in the PVN decreased the fasting-induced rise in UCP-3 in acromiotrapezius muscle, but there was no effect of fasting or UCN treatment on UCP-3 gene expression in biceps femoris muscle (Fig. 2D). Depot differences in UCP-3 responses likely reflect differences in muscle type, level of sympathetic innervation, and/or local fuels available. The functional significance of UCN-induced reduction of UCP-3 in acromiotrapezius muscle (Fig. 2E) is unclear, as the role of UCP-3 in muscle is undefined. Although it has been demonstrated that muscle tissue from UCP-3-knockout mice is more highly coupled (15) and that mice overexpressing UCP-3 are lean compared with controls (31), UCP-3-knockout mice are normal weight (21). Furthermore, although linear relationships between UCP-1 gene expression and UCP-1 activity are usually observed (4, 5, 26), discordant relationships between gene expression of UCP-2 and UCP-3 and uncoupling activity of these proteins have been described (12). Thus the relative contributions of the UCPs to whole body thermogenesis, and the combined effect of their response to PVN UCN administration on overall thermogenesis, are not known.

The relationship between PVN UCN administration, food intake levels, and plasma leptin supports UCN involvement in energy regulation. As shown in Fig. 3, leptin levels of pair-fed aCSF controls were significantly lower than those of aCSF-treated controls, which is consistent with previous demonstrations of decreased leptin due to fasting (42). UCN-treated rats had the same level of food intake as pair-fed aCSF-treated rats, yet leptin levels in this group were significantly higher (Fig. 3). Regression analysis of data including all treatment groups indicates a significant positive correlation between plasma leptin and 24-h food intake (R = 0.622, P = 0.0034), a finding consistent with the hypothesis that food intake results in elevated leptin levels (24), presumably due to insulin-stimulated leptin release from adipose tissue (3, 23). However, when data from UCN-treated rats are removed from this analysis, the correlation coefficient improves (R = 0.854, P = 0.0004). Furthermore, there is no significant correlation between food intake and plasma leptin in an analysis of data only from the UCN-treated rats (R = 0.376, P = 0.3584). Thus UCN administration in the PVN diminishes the linear relationship between food ingestion and leptin release, suggesting that PVN UCN may impose a regulatory influence on energy signaling pathways.

Another explanation for the present results may be that PVN UCN anorectic and metabolic effects are mediated through stimulation of leptin release from adipose tissue. Neuroanatomic data indicate a PVN-WAT signaling pathway (1), and in the current study, UCN in the PVN resulted in increased plasma leptin (Fig. 3). Although it is unlikely that UCN anorectic effects are mediated via leptin because of the immediacy of UCN effects on feeding (40), it has been previously demonstrated that leptin induces UCP-1 (25, 34), which is consistent with our finding of increased UCP-1 in BAT (Fig. 2A). However, the current UCP-2 and UCP-3 data do not fit with a UCN-leptin mediated-pathway: leptin has been reported to increase UCP-2 in BAT (35) and to either increase or not affect muscle UCP-3 (20, 36). Thus a UCN-initiated leptin signal would not explain the observed lack of effect on UCP-2 in BAT (Fig. 2B) or UCN-induced decrease in muscle UCP-3 (Fig. 2E).

In summary, we found that during food restriction, UCN in the PVN increased plasma leptin and UCP-1 in BAT and decreased UCP-3 gene expression in muscle, which may have important consequences for thermogenesis and/or fuel partitioning. These data indicate that UCN decreases feeding while also preventing some of the metabolic conditions induced by fasting. Furthermore, these data implicate the PVN as a potential site of UCN influence on energy metabolism, which is consistent with the major role of the PVN in the regulation of energy balance.