This Article Abstract Full Text (PDF) Purchase Article View Shopping Cart Alert me when this article is cited Alert me if a correction is posted Citation Map Services Email this article to a friend Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Request Copyright Permission Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Wisse, B. E. Articles by Cummings, D. E. Search for Related Content PubMed PubMed Citation Articles by Wisse, B. E. Articles by Cummings, D. E.

Reversal of Cancer Anorexia by Blockade of Central Melanocortin Receptors in Rats

Division of Metabolism, Endocrinology and Nutrition (B.W., R.S.F., D.E.C.), Department of Medicine, Veterans Affairs Puget Sound Health Care System, and Harborview Medical Center (M.W.S.), University of Washington, Seattle, Washington 98108

Address all correspondence and requests for reprints to: David E. Cummings, M.D., VA Puget Sound Health Care System, 1660 South Columbian Way, Endo/111, Seattle, Washington 98108-1597. E-mail: davidec{at}u.washington.edu

	Abstract

Top Abstract Introduction Materials and Methods Results Discussion References

 
Anorexia is a debilitating manifestation of many malignancies. The etiology of cancer anorexia is poorly understood, and effective treatment options are limited. To investigate the role of central melanocortin receptor signaling in the pathogenesis of cancer anorexia, we assessed the effects on food intake of the melanocortin receptor antagonist SHU9119 administered into the third cerebral ventricle of Lobund-Wistar rats that were anorexic from prostate cancer. In anorexic tumor-bearing rats, daily treatment with SHU9119 (0.35 nmol, intracerebroventricularly) increased food intake from 71 ± 3% to 110 ± 6% of preanorectic baseline and caused significant weight gain (13 ± 5 vs. 5 ± 1 g/3 d, SHU9119 vs. baseline in tumor-bearing rats). In control rats pair-fed to the intake of tumor-bearing animals, SHU9119 was ineffective at increasing food intake. The specificity of the SHU9119 feeding response was assessed using two other orexigenic peptides, NPY and the novel hormone ghrelin. Treatment of tumor-bearing rats with intracerebroventricular ghrelin (10 µg) increased food intake, but the effect was blunted relative to that in controls. Intracerebroventricular injections of NPY (1 µg) also failed to reverse anorexia in tumor-bearing rats. Because SHU9119 completely reverses cancer anorexia in this model, whereas ghrelin and NPY do not, increased central nervous system melanocortin signaling is implicated in the pathogenesis of this disorder. This suggests that new targets for the treatment of cancer anorexia may be found in the melanocortin pathways.

	Introduction

Top Abstract Introduction Materials and Methods Results Discussion References

 
CHRONIC DISEASE STATES, including malignancies, infections, and inflammatory conditions, are frequently associated with decreased appetite and food intake (1). Anorexia occurs in up to 40% of people with cancer, significantly decreasing quality of life and increasing morbidity and mortality (2, 3, 4). Due in part to our limited understanding of the pathogenesis, few treatment options exist for cancer anorexia that are supported by controlled trials, and no interventions have been shown to reverse the syndrome (5).

Both central and peripheral mediators have been implicated in cancer anorexia in animal models (1, 6, 7). Much of this work has focused on proinflammatory cytokines, including TNF, IL-6, and IL-1ß (1), but no single cytokine explains the entire syndrome (8, 9). Classical central nervous system (CNS) neurotransmitters including serotonin, dopamine, and noradrenaline are also thought to be involved in cancer anorexia, although interventions aimed at these mediators have been relatively ineffective (10).

There has recently been tremendous improvement in our understanding of the central mechanisms regulating energy balance (11). The neuropeptides involved can be classified as anorexigenic or orexigenic. Neurons containing anorexigenic neuropeptides are activated by the adipocyte hormone leptin and promote decreased food intake and weight loss. Orexigenic neuropeptides are produced by neurons that are suppressed by leptin, and promote increased food intake and weight gain. Among the anorexigenic neuropeptides, the hypothalamic melanocortin -MSH, a product of POMC, is the most strongly implicated in the normal control of food intake (11). -MSH induces anorexia by activating two distinct melanocortin receptors, Mc3r and Mc4r, expressed in the hypothalamus and other brain regions (12). Inactivation of both Mc3r and Mc4r can be achieved with the potent, synthetic antagonist SHU9119, and central administration of this compound produces long-lasting increases in food intake in normal animals (13).

The role of some of the more recently identified regulators of energy balance in cancer anorexia has been investigated. Leptin is decreased in tumor-bearing (TB) animals, consistent with their decreased body fat (14), and leptin-induced paraneoplastic syndromes resulting in cancer anorexia have not been reported. In anorexic animals, hypothalamic mRNA and peptide levels of the orexigen NPY are elevated appropriately for the energy deficit, suggesting that increased hypothalamic NPY signaling is inadequate to prevent defective feeding in this setting (15). Consistent with this, NPY does not elicit normal increases of food intake in animals with cancer anorexia (16). Thus, our improved understanding of the complex systems regulating energy homeostasis has yet to illuminate the pathogenesis of anorexia associated with malignancy.

Ghrelin is a recently described orexigenic hormone secreted primarily by the stomach in response to fasting, whose potency appears second only to NPY (17). The role of ghrelin in cancer anorexia is unexplored, but expression of ghrelin mRNA in the stomach is decreased by IL-1ß (17), a cytokine known to be elevated in malignancy (1). If ghrelin plays a physiological role to stimulate food intake (18, 19), diminished ghrelin signaling could be considered in the etiology of cancer anorexia, a possibility with important therapeutic implications, as orally bioavailable ghrelin receptor agonists exist (20).

The ability to study the pathogenesis and treatment of cancer anorexia requires animal models relevant to human disease. Prostate adenocarcinoma is a common cause of cancer death in men, and among patients with this disease, anorexia is well described and associated with decreased survival (21). Lobund-Wistar rats spontaneously develop prostate cancer, and have been extensively characterized as a model of the human malignancy (22). Cells from these tumors can be inoculated sc into young rats to cause an aggressive, predictable course of tumor progression that includes the development of cancer anorexia (23).

To investigate whether this form of anorexia arises as a consequence of increased CNS melanocortin signaling, we infused the melanocortin antagonist, SHU9119, into the third cerebral ventricle of rats with prostate cancer anorexia. We also investigated whether ghrelin deficiency occurs in TB animals and compared the feeding responses of the orexigens ghrelin and NPY to that of SHU9119 in this form of anorexia. The results indicate that CNS melanocortin receptor blockade completely reverses cancer anorexia in this model, whereas intracerebroventricular (icv) ghrelin and NPY do not, and that the condition is not caused by decreased circulating ghrelin.

	Materials and Methods

Top Abstract Introduction Materials and Methods Results Discussion References

 
Animals
All experimental animals were male Lobund-Wistar rats, aged 12–18 wk, purchased from Dr. Morris Pollard (Lobund Laboratory, University of Notre Dame, South Bend, IN) after sc implantation of either freshly harvested prostate adenocarcinoma cells (TB rats), or vehicle (control rats) over the right flank. Rats were housed individually in a temperature-controlled room (23 ± 2 C) with artificial lighting from 0600–1800 h. The use of animals in this experiment was approved by the Animal Care Committee of the Puget Sound VAMC (Seattle, WA). All animals were handled and maintained in accordance with the NIH Guide for the Care and Use of Laboratory Animals.

Definition of cancer anorexia
TB animals had ad libitum access to pelleted rat chow (Purina Rodent Chow No. 5001) and tap water throughout the two studies. Control (C) rats were either ad libitum-fed or pair-fed to the same amount of food consumed by the anorexic TB rats (see Study Protocol for details). Uneaten food was weighed daily, 30 min before the beginning of the dark cycle, and immediately replaced with a measured quantity of chow for the next 24-h period. Cages were searched for fragments of uneaten pellets. Each animal’s individual baseline 24 h food intake was defined as the average daily food intake over a period of 10 d, following recovery from cannulation surgery. Subsequent food intake data are expressed as a percent of individual, baseline daily food intake. In TB animals, cancer anorexia was defined as either daily food intake below 80% of the individual rat’s baseline food intake for a period of 3 d consecutive, or as a single value below 75% of baseline occurring after a steady decline of at least 3 d duration. Data are presented as group averages of percent of individual baseline food intake. Body weight was determined on a weekly basis until the onset of cancer anorexia, after which animals were weighed every 2–3 d, or daily during SHU9119 treatment.

Surgical procedures
After the rats had acclimatized to our animal facilities for 5 d, a 26-gauge cannula (Plastics One, Roanoke, VA) was implanted into the third cerebral ventricle of each animal under general anesthesia, achieved using inhaled isoflurane as previously described (24). Animals received 0.2 ml of gentamicin (40 mg/ml, Fujisawa USA, Inc., Deerfield, IL) immediately before surgery.

Angiotensin II testing
To verify cannula placement in the third ventricle, all cannulated animals were tested with icv angiotensin II (Sigma, St. Louis, MO), 1 µl of a 1 ng/µl solution over 1 min, using a hand-held 25 µl microsyringe (Hamilton, Reno, NV). This was done 5–7 d following the procedure. Animals failing to consume >5 ml of water within 30 min of the injection were subjected to a second test after a 3- to 4-d interval. Animals failing both tests were rejected as cannulation failures.

Tumor size
Tumor size on the right flank was measured serially in two dimensions, and the overall health of the TB animals was assessed daily by veterinary staff. Two TB rats were deemed too ill to finish the experiment at the time they met criteria for anorexia.

Exp 1: effect of icv SHU9119 on food intake and body weight in TB and C animals
Study protocol (Fig. 1A). Icv administration of SHU9119 was conducted in two phases. Phase 1 determined the effect of 3 d consecutive of SHU9119 treatment on food intake and body weight in C and anorexic TB animals fed ad libitum. Phase 2 was designed to control for the effect of reduced energy intake on the response to SHU9119. Accordingly, the same C animals were subsequently group pair-fed to the diminished food intake of the TB group, and then subjected to SHU9119 treatment. This pair-feeding was accomplished by providing each C animal access to an amount of chow equal to the mean percent of baseline food intake consumed by anorexic TB rats. Food intake in these animals was then determined following a single icv treatment with SHU9119 or vehicle, with subsequent ad libitum access to food (see Icv injections below), or by relieving the food restriction in the absence of icv treatment. After return to baseline food intake and body weight (7 d), this pair-feeding paradigm was repeated in the same C rats, with all animals crossed over to receive an icv injection different from that given during the first pair-feeding period. Although the energy restriction in these C animals was matched to the average food intake of the TB group rather than to any given individual TB animal, this C group is referred to as "pair-fed" throughout the text for the sake of clarity.

View larger version (37K): [in this window] [in a new window]   Figure 1. Study design for Exp 1 and 2. A, In Exp 1 all C animals were initially treated with icv SHU9119 or vehicle (aCSF) during the same time period as the anorexic TB animals. Subsequently, C rats were subjected to a pair-feeding protocol designed to duplicate the energy restriction experienced by the anorexic TB rats. This protocol was followed by treatment with icv SHU9119, icv vehicle, or no icv treatment. Pair-feeding was repeated using a crossover of icv treatment. B, The timing points of orexigen injection and food intake determination in Exp 2 are shown for NPY, ghrelin, and SHU9119. These time points were the same for both TB and C animals.

Phase 2. The pair-feeding paradigm in C animals replicated the food intake of anorexic TB rats during the 9-d period that immediately preceded SHU9119 treatment. At the end of each of the pair-feeding periods (see Study protocol above), C animals were given a single icv dose of SHU9119 (0.35 nmol in a volume of 2 µl of aCSF), aCSF (2 µl) alone, or were handled for 3 min without receiving an icv treatment.

Exp 2: orexigenic effects of icv SHU9119, NPY, and ghrelin in TB and C animals
A second set of Lobund-Wistar rats was prepared using the same procedure as in Exp 1, to compare the feeding response of TB (n = 28) and C (n = 19) rats to icv injection of SHU9119, ghrelin, or NPY. These animals were provided ad libitum access to pelleted rat chow and drinking water at all times, and after third ventricular cannulation had daily food intake measured as described above. In addition, at baseline and during the progression of anorexia in TB animals, food intake determinations were made every 2 h during the final 4 h of the light cycle in both groups for comparison to the 2 h and 4 h food intake measured after icv injections (Fig. 1B). Anorexia in TB animals was defined as above.

Icv injections
Anorexic TB and C animals were given single icv injections of one of two doses of SHU9119 (0.2 or 0.35 nmol; TB n = 11, C n = 9), NPY (1 µg, American Peptide Co., Sunnyvale, CA; TB n = 10, C n = 6) or ghrelin (10 µg, kindly provided by Dr. C. Y. Bowers, Tulane University, New Orleans, LA; TB n = 6, C n = 5). In both TB and C animals, the feeding response to the two SHU9119 doses was not significantly different, so food intake data for both doses were combined. SHU9119 was administered just before the onset of the dark cycle, as described above. Ghrelin and NPY were given 4 h before the onset of the dark cycle in an effort to maximize the effect of these orexigens. NPY is a short-acting peptide with a peak effect on food intake approximately 2 h after injection and is most effective when given during the light cycle, when both endogenous hypothalamic NPY and spontaneous food intake are low. Although the actions of ghrelin are still being elucidated, it appears to exert a maximum effect on food intake during the initial 4 h after an icv injection, and as a short-acting orexigen, would also be expected to have greatest efficacy during the light cycle (17). Food intake responses to SHU9119 are qualitatively different, as the effect is long-lived (>24 h (13)] and relatively modest at early time points. For these reasons, food intake data are presented at 24 h for SHU9119, and at 2 h and 4 h for NPY and ghrelin. Vehicle injections for SHU9119 and NPY (aCSF), and ghrelin (endotoxin-free normal saline, Abbott Laboratories, Chicago, IL) were given to C animals, at the same time of day as the injection of the respective test substance. Injected animals were killed 1 d after treatment or were observed over several days while the treatment effect abated. Food intake determinations continued in all animals until they were killed.

Blood collection
All animals were killed 5–8 h after the onset of the light cycle, following a dark cycle where food was available. After decapitation, trunk blood was collected from each animal in gel separator tubes (Vacutainer, Becton Dickinson and Co., Franklin Lakes, NJ). Serum was isolated by centrifugation at 4 C and stored at -80 C until analyzed.

Serum analyses
Serum glucose (Sigma, St. Louis, MO), triglycerides (Roche Molecular Biochemicals, Indianapolis, IN) and FFA (Wako Chemicals, Inc., Richmond, VA) were measured using trinder-type enzymatic colorimetric assays. Serum leptin (rat leptin RIA, Linco Research, Inc., St. Charles, MO), insulin (rat insulin RIA, Linco Research, Inc.), and ghrelin (rat ghrelin RIA, Phoenix Pharmaceuticals, Inc., Mountain View, CA) were measured by specific RIAs.

Statistical analyses
Comparisons between mean values of food intake, body weight, and serum components were performed using an unpaired t test. Univariate ANOVA tests were used to determine whether treatment effects were significantly different between TB and C groups. The null hypothesis of no difference between groups was rejected at P < 0.05. All values are presented as the mean ± SEM.

	Results

Top Abstract Introduction Materials and Methods Results Discussion References

 
Exp 1: effect of icv SHU9119 on food intake and body weight in TB and C animals
Phase 1. Daily (24 h) food intake did not differ significantly between TB (n = 16) and C (n = 16) rats during the initial 10 d following recovery from third ventricular cannulation (21.6 ± 0.1 g/d and 21.7 ± 0.1 g/d, respectively). In the TB group, food intake decreased to values that met the predetermined criteria for anorexia (see Materials and Methods) in 15 of 16 animals. Two of these rats were deemed too ill to continue the experiment and were euthanized. The time to onset of anorexia after tumor implantation varied between 51 and 62 d. On average, 9 d elapsed between the onset of significant decline in food intake below the preanorexic baseline, until criteria for anorexia were met and icv injections administered. The C animals showed no sustained changes in food intake during this period.

Third ventricular injection of the melanocortin antagonist SHU9119 (0.35 nmol) significantly increased food intake in both TB and C animals fed ad libitum (Fig. 2). Average daily food intake in the TB group rose from 71 ± 3% to 96 ± 5% of preanorexic baseline after the first injection (absolute values 15.8 ± 0.6 to 21.4 ± 1.1 g/d; respectively), and to levels above baseline (110 ± 6% and 106 ± 7%) on the two subsequent treatment days. Among the 13 TB rats treated with SHU9119, every animal showed increased food intake in response to the compound. Food intake also increased in C animals treated with SHU9119, reaching a peak of 139 ± 9% of baseline values on d 1 of the washout period following 3 d of SHU9119 administration. The maximum treatment effect of SHU9119 on food intake, relative to values immediately before treatment, was not different according to tumor status, being 39% in TB animals and 41% in C animals. Icv injection of vehicle (aCSF), in both TB and C groups produced small, transient decreases in food intake (data not shown).

View larger version (19K): [in this window] [in a new window]   Figure 2. Response to icv SHU9119 in TB and C rats. Daily food intake values, expressed as a percent of the preanorexic baseline, are shown for the 19 d before treatment, the 3 consecutive days of SHU9119 therapy, and the 3-d washout period following treatment. Anorexia was defined as food intake <80% of baseline for 3 consecutive days or <75% of baseline for 1 d following a steady decline lasting at least 3 d. Baseline food intake was calculated as the average daily intake during a 10-d period following recovery from icv cannulation (e.g. from d -40 to d -30). Animals received icv injections of SHU9119 (0.35 nmol in 2 µl of artificial CSF) 30 min before the onset of the dark cycle. Data are means ± SEM for TB (, n = 13) and C (, n = 13) groups, both fed ad libitum.

SHU9119 treatment caused significant weight gain in both C and TB animals (Fig. 3). During the 3-d period of SHU9119 treatment, C rats gained 12 ± 5 g (from 353 ± 12 g to 365 ± 13 g) and TB rats gained 13 ± 5 g (from 360 ± 7 g to 373 ± 10 g). This compared with a healthy baseline weight gain of 5 ± 1 g/3 d for both groups (P < 0.01 for both). Tumor growth is unlikely to have contributed substantially to this weight gain, because two-dimensional measurements of the primary tumors did not significantly change during this period (data not shown).

View larger version (15K): [in this window] [in a new window]   Figure 3. Weight change in SHU9119-treated animals. Weight gain is shown for a 3-d period during the preanorexic baseline () and during SHU9119 treatment () in C (n = 13) and TB (n = 13) animals. Data are means ± SEM. The rate of weight gain during SHU9119 treatment compared with an equivalent period before the onset of anorexia was significantly greater in both groups (b vs. a, P = 0.01) but not different between groups. Two-dimensional measures of implanted tumor did not change significantly during the SHU9119 treatment period.

View larger version (16K): [in this window] [in a new window]   Figure 4. Response to SHU9119 in tumorbearing vs. pair-fed control animals. Control rats were pair-fed to the same percent of preanorexic baseline food intake consumed by TB animals. The pair-feeding protocol was performed over a 9-d period before treatment with SHU9119. The last 4 d of the pair-feeding and the result of treatment are shown. Results are presented for pair-fed C animals either treated with SHU9119 (, n = 10) or handled but given no icv treatment (, n = 6), along with the results of SHU9119 treatment in TB animals (, n = 13). Data are means ± SEM. Response to SHU9119 was not significantly different among the three groups.

A single icv dose of SHU9119 dramatically increased food intake in anorexic TB animals (n = 11), from 69 ± 3% to 109 ± 7% of their preanorexic baseline, a response not significantly different from the SHU9119 response that occurred in Exp 1 (P = 0.2). These percent changes reflect an increase of daily food intake from 17.3 ± 0.9 g on the day before treatment to 26.4 ± 2.0 g following SHU9119 (P < 0.0001, Fig. 5A). In C animals (n = 9), treatment with SHU9119 increased 24 h food intake from 21.5 ± 0.6 to 26.5 ± 1.3 g (P = 0.001). The relative change in food intake due to SHU9119 treatment was 53% in TB rats but only 17% in the C animals. The single icv injection of SHU9119 did not significantly alter body weight (data not shown).

View larger version (27K): [in this window] [in a new window]   Figure 5. Comparison of icv orexigens in TB and C rats. A, Shown are the effects of NPY, ghrelin, SHU9119, and vehicle on food intake in C (white bar, pretreatment food intake; diagonal stripe, treatment effect) and anorexic TB rats (checkered pattern, pretreatment food intake; black bar, treatment effect). The time interval over which food intake was assessed is indicated on the abscissa, and was chosen to be appropriate for the kinetics of each peptide. Data are means ± SEM. The significance levels of differences in the icv treatment effect between TB and C animals, as determined by univariate ANOVA, are P < 0.0001 for NPY, P = 0.03 for ghrelin, and P = 0.16 for SHU9119. B, Change in food intake due to treatment in C (white bar) and TB rats (black bar). All treatment differences are significant at P = 0.05 or greater. Data are means ± SEM.

Although icv ghrelin injection stimulated food intake in TB rats, this response was also blunted compared with that in C animals (Fig. 5A). In anorexic TB rats (n = 6), ghrelin treatment resulted in a 4-h food intake of 5.2 ± 1.0 g, compared with 2.6 ± 0.4 g in rats that were handled but received no icv treatment (n = 20, P = 0.04). In C animals (n = 5), ghrelin treatment increased 4-h food intake more potently to 8.5 ± 1.3 g, whereas handled but untreated animals (n = 16) consumed 3.5 ± 0.3 g (P < 0.0001). A decreased orexigenic effect of ghrelin in TB rats was also seen at the 2-h time point. Thus, icv ghrelin increased food intake in anorexic TB animals, but the overall response was blunted in comparison to that of C animals (P = 0.03, ghrelin in TB vs. C). As with NPY, ghrelin did not increase 24-h food intake or body weight in either TB or C groups (data not shown).

In summary, SHU9119 caused a much greater increase in food intake in TB relative to C animals than did NPY or ghrelin. The orexigenic effect of both NPY and ghrelin was blunted in TB rats, whereas the effect of SHU9119 was as great or greater in TB than in C animals (Fig. 5B).

Serum concentrations of ghrelin, leptin, insulin, glucose, FFA, and triglycerides measured during the middle of the light cycle are shown in Fig. 6 for both C and TB rats. Ghrelin levels were increased to the same degree in both anorexic TB and pair-fed C rats (2.1 ± 0.4 ng/ml and 2.4 ± 0.2 ng/ml, respectively; P = 0.25), and both were significantly higher than levels in ad libitum-fed C animals (1.4 ± 0.1 ng/ml; both groups P < 0.02). Treatment of anorexic rats with SHU9119 resulted in lower ghrelin concentrations (1.2 ± 0.3 ng/ml) that were no different from those seen in ad libitum-fed C rats (P = 0.9). The reciprocal pattern was seen with serum leptin concentrations in the four groups. Untreated anorexic TB animals had lower leptin levels (0.9 ± 0.04 ng/ml) than even the pair-fed C rats (1.9 ± 0.3 ng/ml, P = 0.0003). SHU9119-treated TB rats had significantly higher serum leptin concentrations (4.4 ± 0.8 ng/ml, P = 0.0002), equivalent to values seen in ad libitum-fed C rats (3.8 ± 0.4 ng/ml, P = 0.14). Insulin levels showed a similar relationship among the four groups to that seen with leptin. FFA concentrations were elevated in pair-fed C animals, as expected from lipolysis due to calorie restriction, as well as in both TB groups, as expected from active tissue cachexia.

View larger version (30K): [in this window] [in a new window]   Figure 6. Serum values in TB and C rats. Shown are the concentrations of ghrelin, leptin, insulin, glucose, FFA, and triglycerides in the four treatment groups, standardized to the ad libitum-fed C group. Data are means ± SEM. Actual values in the ad libitum-fed C group were: ghrelin 1.4 ± 0.06 ng/ml, leptin 3.8 ± 0.4 ng/ml, insulin 0.8 ± 0.07 ng/ml, glucose 123 ± 15 mg/dl, FFA 0.27 ± 0.06 mEq/liter, and triglycerides 123 ± 8 mg/dl. The four groups of animals are: ad libitum-fed C (white bar, n = 7), pair-fed C (diagonal stripes, n = 6), untreated TB (checkered pattern, n = 9), and SHU9119-treated TB (black bar, n = 9). Ghrelin and leptin values were not significantly different between SHU9119-treated TB and ad libitum-fed C animals, as determined by univariate ANOVA.

	Discussion

Top Abstract Introduction Materials and Methods Results Discussion References

The normalization of food intake by SHU9119, and its greater efficacy in TB rats than in similarly calorie- and leptin-deficient controls, suggests that CNS melanocortin signaling may be inappropriately increased in cancer anorexia. Because hypothalamic melanocortin signaling is reduced in conditions associated with weight loss and low leptin levels (25), we reasoned that SHU9119 would have less impact on food intake in normal animals that are energy-restricted than in ad libitum-fed animals. Our findings confirm this hypothesis. Whereas food intake was markedly increased by SHU9119 treatment in ad libitum-fed controls, pair-fed controls showed no greater increase in food intake with SHU9119 than was seen by discontinuing the food restriction in the setting of no treatment. In TB rats, leptin levels were even lower than in pair-fed controls, which in normal animals should yield low CNS melanocortin signaling and minimal responsiveness to melanocortin antagonism. The fact that TB rats showed a potent feeding response to SHU9119 in the face of low leptin levels suggests that cancer anorexia in this model arises, at least in part, from melanocortin signaling that is inappropriately activated through leptin-independent mechanisms.

The report that hypothalamic mRNA for the melanocortin precursor POMC is not increased in this model of cancer anorexia relative to pair-fed animals (23), argues against increased hypothalamic POMC transcription as a cause of anorexia. Several other mechanisms of increased CNS melanocortin signaling could be invoked. Cancer anorexia may occur through decreases in agouti related peptide (Agrp), the endogenous CNS melanocortin antagonist, rather than increases in -MSH. Increased activity of prohormone convertases could generate greater amounts of -MSH from POMC protein without increases in POMC mRNA levels. SHU9119 could block -MSH derived from POMC neurons in regions outside the hypothalamus, such as the brain stem (26). Finally, the Mc4r has very recently been shown to have endogenous activity in the absence of ligand that can be suppressed by a melanocortin receptor antagonist, acting in this case as an inverse agonist rather than as a traditional competitive antagonist for -MSH (27). Although our data implicate the melanocortin system in the etiology of cancer anorexia, it is feasible that other parallel pathways are also involved, and that SHU9119 may interact with these pathways in an as yet undefined manner.

We hypothesize that cancer anorexia may result from increased central melanocortin signaling driven by proinflammatory cytokines, such as IL-1ß, IL-6, and TNF (Fig. 7). All of these peptides can produce profound anorexia (28), and a variety of tumors and peri-tumoral cells can release them into the circulation (9, 29). These cytokines are also produced in the brain, and hypothalamic expression of IL-1ß has been shown to be elevated in the prostate adenocarcinoma model used in our studies (23). However, blockade of any one of these cytokines using antagonists or gene knockouts fails to eliminate anorexia in various animal models of cancer and inflammation (30, 31, 32, 33, 34, 35), and serum levels of cytokines correlate poorly with cancer anorexia in humans (8). These findings suggest that tumor-related increases of any one cytokine are not sufficient to explain cancer anorexia (9).

View larger version (15K): [in this window] [in a new window]   Figure 7. Proposed model of CNS interaction between cytokines and melanocortin signaling in cancer anorexia.

A similar mechanism may mediate anorexia in nonmalignant inflammation. Administration of SHU9119 was recently shown to be partially effective in reversing anorexia induced by lipopolysaccharide, a component of bacterial cell walls (39). This prototypic inflammatory agent induces anorexia by triggering a profound cytokine response (40), and therefore cytokine-mediated central effects on melanocortins may be a common mechanism underlying the anorexia in both cancer and chronic inflammation.

Data from our laboratory and others argue against a pathogenic role for impaired production of NPY in cancer anorexia. Arcuate nucleus NPY biosynthesis is elevated in animal models of this condition (41), consistent with an adaptive response intended to compensate for anorexia, rather than cause it. Furthermore, in our prostate cancer-bearing rats, as well as in a sarcoma model of anorexia (16), the orexigenic potency of NPY is markedly attenuated compared with its effects in healthy controls, suggesting a functional resistance to the action of NPY.

Ghrelin has not previously been administered to tumor-bearing animals. This potent orexigen is of particular interest because of its hypothesized role as a circulating hormone that can act centrally to stimulate food intake during periods of energy deficit (18, 19). Inadequate secretion of ghrelin and/or resistance to its actions are therefore potential mechanisms of cancer anorexia, especially because ghrelin expression was recently shown to be down-regulated by exogenous IL-1ß, a cytokine often elevated in cancer anorexia (17). However, we found that serum ghrelin levels were increased in anorexic TB animals. As with NPY, this suggests an ineffective compensatory role for ghrelin in cancer anorexia, rather than a causative one. Ghrelin levels were also elevated in pair-fed, healthy controls. This is the first demonstration that circulating ghrelin is positively regulated by chronic energy restriction, as has been shown for acute fasting (18). Whether such increases of ghrelin play a physiological role to increase food intake remains an important, unanswered question.

Like NPY, ghrelin displayed markedly blunted orexigenic potency in TB compared with C animals, suggesting that anorexia in this setting is mediated via mechanisms that compete with, or are downstream of, the actions of NPY and ghrelin. Taken together, these data suggest that alterations of ghrelin and NPY signaling are not causes of cancer anorexia. However, NPY and ghrelin are much shorter-acting orexigens than SHU9119, and though their efficacy was blunted, single icv injections of NPY and ghrelin did increase food intake in anorexic TB rats. Therefore, we cannot discount the possibility that multiple daily injections or continuous administration of these compounds could increase daily food intake and body weight in cancer anorexia. This possibility has important medical implications, as orally bioavailable ghrelin receptor agonists have been developed and administered safely to humans in GH-related studies (20).

Delineating the mechanisms by which SHU9119 reverses cancer anorexia requires an understanding of the biological roles of Mc4r and Mc3r because this compound antagonizes both receptors. Pharmacologic and genetic studies have shown that Mc4r activation reduces food intake and may also increase metabolic rate, whereas Mc3r activation decreases feed efficiency and may mediate the sensation of illness with consequent food aversion (Fig. 7) (12, 42). Mc4r-mediated signaling may predominate in cancer cachexia as demonstrated recently by Marks et al. (43) in a mouse adenocarcinoma model. However, food aversion is common in both humans and animals with cancer. Anorexic TB rodents presented with a novel diet initially increase food intake, presumably due to the absence of learned aversion to the new food (44). Subsequently, food intake decreases rapidly as these animals develop an aversion to the second diet. In summary, the orexigenic effects of SHU9119 in TB animals may arise from a decrease in the sensation of illness and food aversion (Mc3r antagonism), a true increase in appetite (Mc4r antagonism), or both, making melanocortin antagonists a potentially ideal class of agents to increase food intake in patients with cancer.

Administration of SHU9119 to TB rats in our experiments caused remarkable and rapid weight gain, a clinically relevant endpoint in malignancy. The lack of measurable change in tumor size during SHU9119 treatment indicates that tumor growth was not a major contributor to weight gain during this period, although we do not know whether SHU9119-induced weight gain arose primarily from expansion of the fluid, fat, or lean tissue compartments. Importantly, SHU9119 treatment of TB rats reversed the effect of cancer to lower leptin and insulin, and to raise ghrelin concentrations in blood, restoring them to the levels found in ad libitum-fed C animals. The increased food intake resulting from SHU9119 appears to have been sufficient to change the hormonal secretion patterns of white adipose tissue, pancreatic ß cells, and gastrointestinal endocrine cells, from one characteristic of a fasted state to that of a fed state. This occurred even though the elevated FFA values in TB animals suggest ongoing cachexia. Cachexia is often associated with anorexia in malignancy and represents active tissue catabolism causing greater weight loss than can be accounted for by anorexia alone (45). In our cancer model, treatment with SHU9119 increased energy balance enough to replenish nutrient stores at least partially in nontumor tissues, despite presumably ongoing cachexia.

Treatment options for cancer anorexia are few (5, 46). The progesterone-like hormone megestrol acetate is the most extensively characterized agent and has been shown in rodent cancer models to cause modest increases in food intake and body weight, but no increase in lean mass (47, 48). Few studies of other therapeutic agents have been conducted in animal models of cancer anorexia, and the human literature contains a dearth of randomized, placebo-controlled trials testing treatments other than megestrol acetate. Among these, the placebo effect on food intake is often substantial and indistinguishable from that of the study drug. If our findings can be reproduced in other models of cancer anorexia, and can be successfully translated into the clinical setting, melanocortin antagonists may hold promise as new candidates for the treatment of cancer anorexia.

Further studies are required to evaluate the overall clinical impact of ameliorating cancer anorexia with melanocortin antagonists. It is possible that adverse effects of central -MSH antagonists, such as increased fever, may outweigh the positive impact on food intake. Even if anorexia can be safely eliminated with these agents, it remains to be seen whether this would increase survival, as anorexia is but one factor contributing to mortality in patients with neoplastic diseases. Recent studies showing that total parenteral nutrition prolonged life in cancer patients offer some hope in this regard (49). Whether or not survival is extended, there is little question that reversal of anorexia would improve quality of life for those affected.

	Acknowledgments

 
We would like to thank Dr. Morris Pollard and Kay Stewart at the University of Notre Dame for supplying us with Lobund-Wistar rats, and Dr. Cindy Pekow for her expert veterinary care. We also thank Dr. C.Y. Bowers for generously donating the ghrelin used in this experiment.

	Footnotes

 
These studies were supported by grants from the University of Washington Clinical Nutrition Research Unit (NIH DK-35816), a Burroughs Wellcome Fund Career Award (No. 80-1519, to D.E.C.), Diabetes, Metabolism and Endocrinology Training Grant (NIH/NIDDK T32-DK-07247), an Endocrine Fellows Foundation grant, and the Medical Research Service of the Department of Veterans Affairs.

Abbreviations: aCSF, Artificial cerebrospinal fluid; C, control; CNS, central nervous sytem; icv, intracerebroventricular(ly); Mc3r and Mc4r, two distinct melanocortin receptors; TB, tumor-bearing.

Received March 5, 2001.

Accepted for publication April 16, 2001.

	References

Top Abstract Introduction Materials and Methods Results Discussion References

 

1. Plata-Salaman CR 1996 Anorexia during acute and chronic disease. Nutrition 12:69–78[CrossRef][Medline]
2. Tisdale MJ 1997 Biology of cachexia. J Natl Cancer Inst 89:1763–1773[Abstract/Free Full Text]
3. Dewys WD, Begg C, Lavin PT, et al. 1980 Prognostic effect of weight loss prior to chemotherapy in cancer patients. Eastern Cooperative Oncology Group. Am J Med 69:491–497[CrossRef][Medline]
4. Bruera E 1997 ABC of palliative care. Anorexia, cachexia, and nutrition. BMJ 315:1219–1222[Free Full Text]
5. Bruera E 1998 Pharmacological treatment of cachexia: any progress? Support Care Cancer 6:109–113[CrossRef][Medline]
6. Plata-Salaman CR 2000 Central nervous system mechanisms contributing to the cachexia-anorexia syndrome. Nutrition 16:1009–1012[CrossRef][Medline]
7. Inui A 1999 Cancer anorexia-cachexia syndrome: are neuropeptides the key? Cancer Res 59:4493–501[Abstract/Free Full Text]
8. Maltoni M, Fabbri L, Nanni O, et al. 1997 Serum levels of tumour necrosis factor and other cytokines do not correlate with weight loss and anorexia in cancer patients. Support Care Cancer 5:130–135[Medline]
9. Noguchi Y, Yoshikawa T, Matsumoto A, Svaninger G, Gelin J 1996 Are cytokines possible mediators of cancer cachexia? Surg Today 26:467–475[CrossRef][Medline]
10. Meguid MM, Fetissov SO, Varma M, et al. 2000 Hypothalamic dopamine and serotonin in the regulation of food intake. Nutrition 16:843–857[CrossRef][Medline]
11. Schwartz MW, Woods SC, Porte Jr D, Seeley RJ, Baskin DG 2000 Central nervous system control of food intake. Nature 404:661–671[Medline]
12. Cummings DE, Schwartz MW 2000 Melanocortins and body weight: a tale of two receptors. Nat Genet 26:8–9[CrossRef][Medline]
13. Hagan MM, Rushing PA, Schwartz MW, Yagaloff KA, Burn P, Woods SC, Seeley RJ 1999 Role of the CNS melanocortin system in the response to overfeeding. J Neurosci 19:2362–2367[Abstract/Free Full Text]
14. Lopez-Soriano J, Carbo N, Tessitore L, Lopez-Soriano FJ, Argiles JM 1999 Leptin and tumor growth in rats. Int J Cancer 81:726–729[CrossRef][Medline]
15. Chance WT, Balasubramaniam A, Fischer JE 1995 Neuropeptide Y and the development of cancer anorexia. Ann Surg 221:579–587; discussion 587–589[Medline]
16. Chance WT, Balasubramaniam A, Thompson H, Mohapatra B, Ramo J, Fischer JE 1996 Assessment of feeding response of tumor-bearing rats to hypothalamic injection and infusion of neuropeptide Y. Peptides 17:797–801[CrossRef][Medline]
17. Asakawa A IA, Kaga T, Yuzuriha H, et al. 2001 Ghrelin is an appetite-stimulating signal from stomach with structural resemblance to motilin. Gastroenterology 120:337–345[CrossRef][Medline]
18. Tschop M, Smiley DL, Heiman ML 2000 Ghrelin induces adiposity in rodents. Nature 407:908–913[CrossRef][Medline]
19. Wren AM, Small CJ, Ward HL, et al. 2000 The novel hypothalamic peptide ghrelin stimulates food intake and growth hormone secretion. Endocrinology 141:4325–4328[Abstract/Free Full Text]
20. Ghigo E, Arvat E, Camanni F 1998 Orally active growth hormone secretagogues: state of the art and clinical perspectives. Ann Med 30:159–168[Medline]
21. Emrich LJ, Priore RL, Murphy GP, Brady MF 1985 Prognostic factors in patients with advanced stage prostate cancer. Cancer Res 45:5173–5179[Abstract/Free Full Text]
22. Pollard M 1998 Lobund-Wistar rat model of prostate cancer in man. Prostate 37:1–4[CrossRef][Medline]
23. Plata-Salaman CR, Ilyin SE, Gayle D 1998 Brain cytokine mRNAs in anorectic rats bearing prostate adenocarcinoma tumor cells. Am J Physiol 275:R566–R573
24. Chavez M, Seeley RJ, Woods SC 1995 A comparison between effects of intraventricular insulin and intraperitoneal lithium chloride on three measures sensitive to emetic agents. Behav Neurosci 109:547–550[CrossRef][Medline]
25. Schwartz MW, Seeley RJ, Woods SC, et al. 1997 Leptin increases hypothalamic pro-opiomelanocortin mRNA expression in the rostral arcuate nucleus. Diabetes 46:2119–2123[Abstract]
26. Grill HJ, Ginsberg AB, Seeley RJ, Kaplan JM 1998 Brainstem application of melanocortin receptor ligands produces long-lasting effects on feeding and body weight. J Neurosci 18:10128–10135[Abstract/Free Full Text]
27. Nijenhuis WA, Oosterom J, Adan RA 2001 AgRP(83–132) acts as an inverse agonist on the human-melanocortin-4 receptor. Mol Endocrinol 15:164–171[Abstract/Free Full Text]
28. Plata-Salaman CR, Sonti G, Borkoski JP, Wilson CD, French-Mullen JM 1996 Anorexia induced by chronic central administration of cytokines at estimated pathophysiological concentrations. Physiol Behav 60:867–875[Medline]
29. Lonnroth C, Moldawer LL, Gelin J, Kindblom L, Sherry B, Lundholm K 1990 Tumor necrosis factor- and interleukin-1 production in cachectic, tumor-bearing mice. Int J Cancer 46:889–896[Medline]
30. Torelli GF, Meguid MM, Moldawer LL, Edwards 3rd CK, Kim HJ, Carter JL, Laviano A, Fanelli FR 1999 Use of recombinant human soluble TNF receptor in anorectic tumor- bearing rats. Am J Physiol 277:R850–R855
31. Smith BK, Kluger MJ 1993 Anti-TNF- antibodies normalized body temperature and enhanced food intake in tumor-bearing rats. Am J Physiol 265:R615–R619
32. Sherry BA, Gelin J, Fong Y, et al. 1989 Anticachectin/tumor necrosis factor- antibodies attenuate development of cachexia in tumor models. FASEB J 3:1956–1962[Abstract]
33. Leon LR, Kozak W, Peschon J, Kluger MJ 1997 Exacerbated febrile responses to LPS, but not turpentine, in TNF double receptor-knockout mice. Am J Physiol 272:R563–R569
34. Fantuzzi G, Zheng H, Faggioni R, et al. 1996 Effect of endotoxin in IL-1ß-deficient mice. J Immunol 157:291–296[Abstract]
35. Fattori E, Cappelletti M, Costa P, et al. 1994 Defective inflammatory response in interleukin 6-deficient mice. J Exp Med 180:1243–1250[Abstract/Free Full Text]
36. Lipton JM, Catania A 1998 Mechanisms of antiinflammatory action of the neuroimmunomodulatory peptide -MSH. Ann N Y Acad Sci 840:373–380[CrossRef][Medline]
37. Huang QH, Entwistle ML, Alvaro JD, Duman RS, Hruby VJ, Tatro JB 1997 Antipyretic role of endogenous melanocortins mediated by central melanocortin receptors during endotoxin-induced fever. J Neurosci 17:3343–3351[Abstract/Free Full Text]
38. Rajora N, Boccoli G, Burns D, Sharma S, Catania AP, Lipton JM 1997 -MSH modulates local and circulating tumor necrosis factor- in experimental brain inflammation. J Neurosci 17:2181–2186[Abstract/Free Full Text]
39. Huang QH, Hruby VJ, Tatro JB 1999 Role of central melanocortins in endotoxin-induced anorexia. Am J Physiol 276:R864–R871
40. Gayle D, Ilyin SE, Plata-Salaman CR 1999 Feeding status and bacterial LPS-induced cytokine and neuropeptide gene expression in hypothalamus. Am J Physiol 277:R1188–R1195
41. Chance WT, Sheriff S, Kasckow JW, Regmi A, Balasubramaniam A 1998 NPY messenger RNA is increased in medial hypothalamus of anorectic tumor-bearing rats. Regul Pept 75–76:347–353
42. Benoit SC, Schwartz MW, Lachey JL, et al. 2000 A novel selective melanocortin-4 receptor agonist reduces food intake in rats and mice without producing aversive consequences. J Neurosci 20:3442–3448[Abstract/Free Full Text]
43. Marks DL, Ling N, Cone RD 2001 Role of the central melanocortin system in cachexia. Cancer Res 61:1432–1438[Abstract/Free Full Text]
44. Bernstein IL, Sigmundi RA 1980 Tumor anorexia: a learned food aversion? Science 209:416–418[Abstract/Free Full Text]
45. Tisdale MJ 1999 Wasting in cancer. J Nutr 129:243S–246S
46. Loprinzi CL, Kugler JW, Sloan JA, et al. 1999 Randomized comparison of megestrol acetate versus dexamethasone versus fluoxymesterone for the treatment of cancer anorexia/cachexia. J Clin Oncol 17:3299–306[Abstract/Free Full Text]
47. McCarthy HD, Crowder RE, Dryden S, Williams G 1994 Megestrol acetate stimulates food and water intake in the rat: effects on regional hypothalamic neuropeptide Y concentrations. Eur J Pharmacol 265:99–102[CrossRef][Medline]
48. Beck SA, Tisdale MJ 1990 Effect of megestrol acetate on weight loss induced by tumour necrosis factor and a cachexia-inducing tumour (MAC16) in NMRI mice. Br J Cancer 62:420–424[Medline]
49. Bozzetti F, Cozzaglio L, Gavazzi C, et al. 1998 Nutritional support in patients with cancer of the esophagus: impact on nutritional status, patient compliance to therapy, and survival. Tumori 84:681–686[Medline]

This article has been cited by other articles:

				J. M. Scarlett, X. Zhu, P. J. Enriori, D. D. Bowe, A. K. Batra, P. R. Levasseur, W. F. Grant, M. M. Meguid, M. A. Cowley, and D. L. Marks Regulation of Agouti-Related Protein Messenger Ribonucleic Acid Transcription and Peptide Secretion by Acute and Chronic Inflammation Endocrinology, October 1, 2008; 149(10): 4837 - 4845. [Abstract] [Full Text] [PDF]

				S. Strassburg, S. D. Anker, T. R. Castaneda, L. Burget, D. Perez-Tilve, P. T. Pfluger, R. Nogueiras, H. Halem, J. Z. Dong, M. D. Culler, et al. Long-term effects of ghrelin and ghrelin receptor agonists on energy balance in rats Am J Physiol Endocrinol Metab, July 1, 2008; 295(1): E78 - E84. [Abstract] [Full Text] [PDF]

				E. R. Ropelle, J. R. Pauli, K. G. Zecchin, M. Ueno, C. T. de Souza, J. Morari, M. C. Faria, L. A. Velloso, M. J. A. Saad, and J. B. C. Carvalheira A Central Role for Neuronal Adenosine 5'-Monophosphate-Activated Protein Kinase in Cancer-Induced Anorexia Endocrinology, November 1, 2007; 148(11): 5220 - 5229. [Abstract] [Full Text] [PDF]

				M. D. DeBoer, X. X. Zhu, P. Levasseur, M. M. Meguid, S. Suzuki, A. Inui, J. E. Taylor, H. A. Halem, J. Z. Dong, R. Datta, et al. Ghrelin Treatment Causes Increased Food Intake and Retention of Lean Body Mass in a Rat Model of Cancer Cachexia Endocrinology, June 1, 2007; 148(6): 3004 - 3012. [Abstract] [Full Text] [PDF]

				J. R. Nicholson, G. Kohler, F. Schaerer, C. Senn, P. Weyermann, and K. G. Hofbauer Peripheral Administration of a Melanocortin 4-Receptor Inverse Agonist Prevents Loss of Lean Body Mass in Tumor-Bearing Mice J. Pharmacol. Exp. Ther., May 1, 2006; 317(2): 771 - 777. [Abstract] [Full Text] [PDF]

				G. Bronner, A. M. Sattler, A. Hinney, M. Soufi, F. Geller, H. Schafer, B. Maisch, J. Hebebrand, and J. R. Schaefer The 103I Variant of the Melanocortin 4 Receptor Is Associated with Low Serum Triglyceride Levels J. Clin. Endocrinol. Metab., February 1, 2006; 91(2): 535 - 538. [Abstract] [Full Text] [PDF]

				J.N. Gordon, S.R. Green, and P.M. Goggin Cancer cachexia QJM, November 1, 2005; 98(11): 779 - 788. [Abstract] [Full Text] [PDF]

				S. Markison, A. C. Foster, C. Chen, G. B. Brookhart, A. Hesse, S. R. J. Hoare, B. A. Fleck, B. T. Brown, and D. L. Marks The Regulation of Feeding and Metabolic Rate and the Prevention of Murine Cancer Cachexia with a Small-Molecule Melanocortin-4 Receptor Antagonist Endocrinology, June 1, 2005; 146(6): 2766 - 2773. [Abstract] [Full Text] [PDF]

				D. L. Williams and D. E. Cummings Regulation of Ghrelin in Physiologic and Pathophysiologic States J. Nutr., May 1, 2005; 135(5): 1320 - 1325. [Abstract] [Full Text] [PDF]

				K. E. Foster-Schubert, A. McTiernan, R. S. Frayo, R. S. Schwartz, K. B. Rajan, Y. Yasui, S. S. Tworoger, and D. E. Cummings Human Plasma Ghrelin Levels Increase during a One-Year Exercise Program J. Clin. Endocrinol. Metab., February 1, 2005; 90(2): 820 - 825. [Abstract] [Full Text] [PDF]

				A. J. van der Lely, M. Tschop, M. L. Heiman, and E. Ghigo Biological, Physiological, Pathophysiological, and Pharmacological Aspects of Ghrelin Endocr. Rev., June 1, 2004; 25(3): 426 - 457. [Abstract] [Full Text] [PDF]

				N. M. Neary, C. J. Small, A. M. Wren, J. L. Lee, M. R. Druce, C. Palmieri, G. S. Frost, M. A. Ghatei, R. C. Coombes, and S. R. Bloom Ghrelin Increases Energy Intake in Cancer Patients with Impaired Appetite: Acute, Randomized, Placebo-Controlled Trial J. Clin. Endocrinol. Metab., June 1, 2004; 89(6): 2832 - 2836. [Abstract] [Full Text] [PDF]