| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
Amylin Pharmaceuticals, Inc., San Diego, California 92121
Address all correspondence to: Jonathan Roth, Ph.D., Amylin Pharmaceuticals, Inc., 9360 Towne Centre Drive, San Diego, California 92121. E-mail: jonathan.roth{at}amylin.com.
| Abstract |
|---|
|
|
|---|
| Introduction |
|---|
|
|
|---|
Amylin is a 37-amino-acid peptide hormone produced in pancreatic ß-cells and cosecreted with insulin in response to nutrient ingestion. Plasma amylin concentrations rise rapidly in response to meals, peak within 60 min, and remain elevated for up to 4 h (1). Once in the periphery, amylin binds with high affinity to receptors in the central nervous system, particularly within the area postrema (AP) (2, 3, 4). Peripheral administration of amylin to lean or diet-induced obese (DIO) rats decreased food intake and weight (2, 5, 6, 7). In DIO rats, pair-feeding studies revealed that whereas food intake reduction was the predominant mode of action for overall weight loss, weight loss in amylin-treated rats was entirely attributable to a reduction in fat mass, with relative preservation of lean mass. In contrast, pair-fed control animals experienced reductions in both fat and lean body mass. Importantly, amylin-induced weight loss was not associated with counterregulatory decreases in EE (7). Several of these findings in rodents have been mirrored in various clinical studies. Pramlintide, a synthetic analog of human amylin, has been shown to increase satiety, resulting in decreased 24-h food intake and weight loss in obese subjects (8, 9, 10, 11). Randomized, double-blind, phase 2 studies in obese subjects have demonstrated placebo-corrected weight loss of 3.1% over 16 wk (12) and up to 6–7% over 1 yr with pramlintide monotherapy (13).
PYY is a 36-amino-acid peptide hormone that is secreted from intestinal L-cells after a meal and cleaved by dipeptidyl peptidase-IV (DPP-IV) to generate PYY[3–36], the major circulating form (14). After nutrient ingestion, plasma PYY[3–36] levels increase within 15 min, reach a peak at approximately 90 min, and remain elevated for up to 6 h (15). PYY[3–36] levels reflect meal size and content, with fat being the most potent stimulator of PYY secretion (16). PYY[3–36] has been suggested to bind to Y2 receptors expressed by neurons within the hypothalamic arcuate nucleus (17). Experiments in rodents have shown that PYY[3–36] inhibits food intake, reduces body weight gain, and increases utilization of fat stores for energy (17, 18). Obese humans may have lower levels of PYY than lean subjects, and PYY levels have been shown to be inversely correlated with body mass index (19). Infusion of PYY[3–36] into obese (19) and lean (19, 20) subjects reduced caloric intake, and a single sc injection of PYY[3–36] resulted in a trend toward increased hunger suppression and satiety quotient after a meal (21, 22). However, a phase 2 randomized, placebo-controlled study found that nasally administered PYY[3–36] did not significantly reduce weight, compared with placebo, after 12 wk of treatment (23).
The full potential of peptide hormone therapeutics for obesity treatment may not be realized if each neurohormone is evaluated only in isolation. It is conceivable that greater weight loss could be achieved with combinatorial regimens of neurohormones, especially with those that have complementary mechanisms of action and naturally occurring synergies. Furthermore, the metabolic counterregulatory adaptations (increased appetite, decline in metabolic rate, reduced thyroid hormone levels, and increased muscle efficiency) elicited by reduced caloric intake and body weight (24) may be easier to override with combination therapies. We hypothesized that combination therapy with amylin and PYY[3–36] would elicit additive, and possibly synergistic, anorexigenic and weight-lowering effects. We report our findings on the interactive effects of amylin and PYY[3–36] in multiple rodent models.
| Materials and Methods |
|---|
|
|
|---|
Twenty-four-hour food intake monitoring
HSD rats were habituated to the testing chamber, food hopper, and diet for 10 d before the study. Testing chambers (10.5 x 19 x 8 in., BioDAQ Food Intake Monitor; Research Diets) were equipped with a tunnel containing a food hopper at the end. Rats received daily ip injections of vehicle (10% dimethylsulfoxide in sterile water) before lights off on d 8, 9, and 10 of habituation. After habituation, rats (initial body weight was
441 g at time 0) were divided into treatment groups (seven to eight rats per group) of equal food consumption. On test day, rats received a single ip injection of vehicle or peptide (10 µg/kg amylin and/or 1000 µg/kg PYY[3–36]) 15 min before lights off and were placed immediately back into the BioDAQ Food Intake Monitor cages. These doses were chosen because previous data showed near-maximal anorexigenic effects in this model (data not shown). Food hoppers were weighed automatically every 5 sec throughout the experiment. Cumulative food intake (in grams) was monitored for 24 h.
Locomotor activity
To examine the acute effects of amylin and PYY[3–36] on locomotor activity, HSD rats (six rats per group) received a single ip injection of vehicle or peptide (10 µg/kg amylin and/or 1000 µg/kg PYY[3–36]) just before lights out. Immediately after injection, rats were placed into individual cages within SmartFrame Cage Racks (Kinder Scientific, San Diego, CA) with ad libitum access to food and water and were undisturbed for 24 h. This cage system measures locomotor activity via photo beam breaks. Total zone distance (meters) was recorded for 12-h intervals. MotorMonitor software (Kinder Scientific) was used to analyze the data.
DIO mouse and rat: single-dose combinations
For chronic infusion studies, each drug (or vehicle) was delivered by a separate surgically implanted sc osmotic minipump (Durect Corp., Cupertino, CA) containing either drug or vehicle (50% dimethylsulfoxide in sterile water). DIO mice (11–14 mice per group) were weight-matched into treatment groups (initial body weight was
27 g on d 0) and implanted with minipumps containing vehicle, 300 µg/kg·d amylin, 1000 µg/kg·d PYY[3–36], or amylin plus PYY[3–36]. Similarly, DIO-prone rats (five rats per group) were weight-matched into treatment groups (initial body weight was
500 g on d 0) and implanted with minipumps containing vehicle, 100 µg/kg·d amylin, 200 µg/kg·d PYY[3–36], or amylin plus PYY[3–36]. Thus, both studies had a total of four treatment groups: vehicle plus vehicle, vehicle plus amylin, vehicle plus PYY[3–36], or PYY[3–36] plus amylin. Because the weight-loss efficacy of amylin and PYY[3–36] varies somewhat across DIO mice and DIO-prone rats (unpublished observations), a higher dose range was tested in mice. Body weight [expressed as percent change (vehicle-corrected)] and cumulative food intake [expressed as percent change [vehicle-corrected)] were measured weekly for 2 wk. At termination, plasma from DIO-prone rats that were fasted for approximately 3 h (postabsorptive;
0800 h) before plasma collection were assayed for amylin and PYY[3–36] by ELISA (Millipore, Billerica, MA). Amylin and PYY[3–36] concentrations are expressed as mean ± SE.
Indirect calorimetry
On d 4 of treatment, DIO-prone rats (five rats per group) were placed in an indirect calorimeter (Oxymax Equal Flow System; Columbus Instruments, Columbus, OH). Rats were habituated in the Oxymax cages for 24 h before testing. Measurements were taken over a 24-h period during which the animals had ad libitum access to food and water. Respiratory quotient (RQ) and EE were calculated and averaged across the 24-h measurement session.
DIO-prone rat: response surface methodology (RSM)
In an independent group of animals, DIO-prone rats were weight-matched (initial body weight was
500 g on d 0) into one of the 12 groups (five rats per group) resulting from a full factorial design using four doses of amylin (0, 4, 20, and 100 µg/kg·d) and three doses of PYY[3–36] (0, 200, and 400 µg/kg·d) for a total of 60 rats. Body weight and cumulative food intake were measured weekly for 2 wk.
Statistical analyses
Treatment differences for locomotor activity, RQ, EE, 24-h feeding, and vehicle-corrected changes from baseline in food intake and body weight were analyzed using one-way ANOVA followed by contrasts. Due to the small sample size, multiple comparisons were not performed. Statistical analyses were done in SAS version 8.2 (SAS Institute, Inc., Cary, NC) or Prism 4 for Windows (GraphPad Software, San Diego, CA). Graphs were generated using Prism 4 for Windows. Data are presented as mean ± SE.
For RSM, a first-order plus interaction model was fit to the observed percent vehicle-corrected body weight. The model contained effects for the intercept, linear effect of amylin, linear effect of PYY[3–36], and linear interaction effect for the amylin plus PYY[3–36] combination. To improve the fit of the statistical model, the doses of amylin were expressed as the base 10 logarithm, where 1 was added to avoid problems with the zero dose [i.e. dose amylin = log10(dose amylin + 1)]. A reduced second-order model was fit to the observed percent vehicle-corrected food intake to improve the fit. This model contained effects for the intercept, linear, and quadratic effects of amylin, linear effect of PYY[3–36], and linear interaction effect for the amylin plus PYY[3–36] combination. The doses of amylin were again expressed using the base 10 logarithm modified for the zero dose. Lack of fit for the model was tested by the usual F test derived by partitioning the error sum of squares into pure error and lack of fit. Using this lack of fit test, other quadratic and higher-order effects were not necessary and were excluded from the model.
Using these models, a response surface was created to predict weight loss and food intake over the entire dose region. A formal test for synergy or antagonism was performed by comparing the P value for the interaction term of each model to 0.05. A significant interaction indicates that the slope of the response surface changes as the combination doses are increased. The sign of the interaction term coefficient is indicative of the type of departure from additivity; a negative coefficient represents synergistic reductions in body weight or food intake. P values
0.05 suggest additivity. For first-order models, the P value from the interaction term has been shown (26) to be the correct P value for testing departures from additivity using Berenbaums interaction index derived from isobolograms (27), a widely accepted method of rigorously evaluating combinatorial relationships.
| Results |
|---|
|
|
|---|
|
Chronic administration of amylin and PYY[3–36] in DIO mice and DIO-prone rats
The observation that acute administration of amylin plus PYY[3–36] was anorexigenic at 24 h after administration led us to test the durability and potential weight-reducing effects of the amylin plus PYY[3–36] combination using continuous drug administration. After 2 wk of therapy in DIO mice, cumulative food intake (vehicle corrected) was reduced by 3.5 ± 1.5%, 9.7 ± 0.9% (P < 0.05), and 11.4 ± 1.9% (P < 0.05) in amylin, PYY[3–36] and combination-treated groups, respectively (Fig. 2A
). Similarly, body weight (vehicle corrected) was reduced in all treatment groups: amylin –3.1 ± 0.7% (P < 0.05), PYY[3–36] –6.1 ± 0.9% (P < 0.05), and combination –9.2 ± 0.9% (P < 0.05) (Fig. 2B
). Combination therapy also significantly reduced body weight compared with either monotherapy at 2 wk (P < 0.05).
|
|
Indirect calorimetry in DIO rats treated with amylin and PYY[3–36] combination therapy
Indirect calorimetry was performed in DIO-prone rats after the first week of treatment to test whether weight changes observed with monotherapy and/or combination therapy were associated with increased fatty acid oxidation (RQ) and/or changes in EE (Table 1
). RQ was significantly reduced compared with vehicle in rats treated with combination therapy (P < 0.05) but not in rats treated with amylin or PYY[3–36] alone. EE was unchanged across treatment groups.
|
0.05 and synergy if P < 0.05. In addition to formally testing for antagonism or synergy, RSM can be used to predict the response over an entire combination dose region. Using this predictive model, optimal dose combinations for a specific response can be easily identified. DIO-prone rats were treated for 2 wk with a range of doses of amylin (0, 4, 20, and 100 µg/kg·d) and PYY[3–36] (0, 200, and 400 µg/kg·d). At wk 2, RSM analysis of cumulative food intake (Fig. 4
|
|
| Discussion |
|---|
|
|
|---|
Several groups have described the anorexigenic effects of a single peripheral injection of either amylin (32) or PYY[3–36] (33) on 24-h food intake. Similar to the results of the current study, the anorexigenic effects of individually administered amylin and PYY[3–36] were most evident during the first few hours after injection. Food intake rebounded after the initial hypophagia, most likely due to the relatively short half-lives of these peptides and/or to compensatory responses related to energy balance (34). For example, decreased food intake can result in a net reduction in both gastric distention and secretion of nutrient-stimulated satiety signals (35), which could in turn prevent the full manifestation of the anorexigenic properties of an individually administered peptide. Pharmacologically replacing these neurohormonal signals through the use of peptide combinations represents one potential approach to overcoming such compensatory mechanisms and may have contributed to the more robust, sustained reduction in food intake. For example, in the current study, the combination of amylin plus PYY[3–36] maintained an anorexigenic effect for the entire 24-h study period. The increased effect observed with amylin plus PYY[3–36] is consistent with an emerging literature regarding the acute synergy of pancreatic, intestinal, and adipose signals (33, 36, 37, 38, 39, 40, 41).
Although our initial finding of an additive effect of amylin and PYY[3–36] on short-term food intake in normal rats was encouraging, it was important to demonstrate the durability of the effect and determine whether or not the peptide combination could elicit meaningful weight loss in rodent models of obesity. Therefore, DIO mice and DIO-prone rats were treated with amylin and/or PYY[3–36] via continuous infusion for 14 d. In both DIO models, food intake and body weight were reduced to a greater extent by the combination relative to either monotherapy. Previous studies indicated that monotherapy with either amylin or PYY[3–36] was associated with a decrease in RQ and a maintenance of metabolic rate in the face of decreased caloric intake and body weight loss (7, 42). Adding to these previous findings, our experiments showed that metabolic rate was maintained not only during amylin or PYY[3–36] monotherapy but also when both peptides were given in combination. Additionally, RQ was reduced relative to vehicle in DIO-prone rats treated with amylin plus PYY[3–36], but not monotherapy, suggesting an increase in fat oxidation with this treatment regimen. Of note, relative to the rats, the mice showed a less robust anorexigenic response to the combination of amylin and PYY[3–36] with a similar percent reduction in body weight. Although indirect calorimetry was not captured in these animals, these results suggest that mice may be more sensitive than rats to the metabolic effects of the combination of amylin plus PYY[3–36].
Having established in two DIO models that a combinatorial regimen with amylin plus PYY[3–36] elicited greater weight loss than was achievable with either peptide alone (at the doses tested), our next set of experiments sought to formally describe the interaction between the two peptide hormones. Given the complexity of the neurohormonal systems contributing to the regulation of food intake and body weight, it is difficult to capture the extent/efficacy of drug interactions using single-dose methodology. Using RSM (3 x 4 factorial design), we demonstrated that amylin and PYY[3–36] interacted in a synergistic manner to reduce food intake in DIO-prone rats, whereas body weight was reduced in an additive manner. This discordance between synergistic reduction of food intake and additive weight-loss suggests that some counterregulation was evident even with combination therapy. For example, at the time point queried, EE was not maintained as well with combination therapy as it was with vehicle control. Additional counterregulatory adaptations that could also explain the discordance include reduced chronic activity levels, increased skeletal muscle work efficiency, increased fat absorption (reduced fecal energy loss), decreased sympathetic nervous system tone, and changes in the thyroid axis (24, 43). These mechanisms should be evaluated in future studies. However, the fact that combination therapy reduced weight to a greater extent than either PYY[3–36] or amylin alone indicates that combination therapies are a promising avenue for the development of neurohormones as potential antiobesity drugs.
With both monotherapy and combination therapy, it is important to consider the contribution of malaise or competing behaviors to the observed reduction in food intake and body weight. Typically, potential nonhomeostatic effects of compounds (e.g. conditioned taste aversion, locomotor effects, and pica) are evaluated in acute preclinical models. Because there are also clinical data regarding the tolerability and weight-reducing effects of the amylin analog pramlintide and the anorexigenic properties of PYY[3–36], it is now possible to draw parallels between these observations. Findings from several studies clearly suggest that acute amylin administration in rodents is not associated with malaise (conditioned taste aversion) or competing locomotor behaviors (44, 45, 46, 47). Consistent with preclinical models, the effects of pramlintide on food intake and body weight were dissociable from the occurrence of nausea. Approximately 76% of patients who received pramlintide achieved a statistically significant reduction in body weight and did not report any malaise (11). Findings with PYY[3–36] in rodents are less clear, with some reports suggesting that PYY[3–36] supports taste aversion learning (48, 49), whereas other groups have not been able to confirm this observation (37, 50). PYY[3–36] administration decreased energy intake in lean and obese humans (after iv infusion) without incidence of nausea (17, 19), or with a range of nausea, fullness, and abdominal discomfort depending on the route of administration (iv, sc, or intranasal) (20, 23, 51, 52). In the present studies, the acute effects of a single-dose combination of amylin and PYY[3–36] on 24-h food intake, alone and in combination, were shown to be dissociable from any effects on locomotor activity in rats. With sustained administration, our plasma data (in rats) indicate that the effects on food intake and body weight were clearly achieved at pharmacological levels of each compound (
40- and
200-fold above endogenous levels of amylin and PYY[3–36], respectively). Ongoing clinical research studies assessing the safety and tolerability of pramlintide and PYY[3–36] will hopefully define the therapeutic window for decreasing food intake and reducing body weight without unwanted side effects. The observation that additivity/synergy was achieved across a wide range of dose combinations in DIO rats suggests that it may be possible to select an optimally tolerated combination while retaining the desired effects on food intake and body weight.
Although the neural substrates mediating the observed anorexigenic synergy and weight loss additivity between amylin and PYY[3–36] were not measured in the present studies, potential neuroanatomical bases for convergence of their signaling pathways exist. Both amylin (53) and PYY[3–36] (54) bind receptors located within the AP in the caudal brainstem. Amylins anorexigenic properties are attributable to its binding in the AP and subsequent activation of the nuclei tractus solitarii, lateral parabrachial nucleus, and the central nucleus of the amygdala (55, 56). PYY[3–36] mediates its effects on gastrointestinal motility through brainstem Y2 receptors (57, 58). However, unlike amylin, animals with lesions of the AP are still responsive to the anorexigenic effects of peripherally administered PYY[3–36] (59). The anorexigenic effects of PYY[3–36] are thought to be mediated through Y2 receptors located on neuropeptide Y neurons within the arcuate nucleus neurons of the hypothalamus (17). Interactions between amylin and PYY[3–36] may also take place via indirect projection pathways. For example, amylin-specific effects within the AP may influence upstream hypothalamic signaling. Amylin inhibited fasting-induced neuronal activation within the lateral hypothalamic area and reduced mRNA levels of orexin (56, 60), a food intake-stimulating peptide. Sustained administration of amylin (but not pair feeding) increased arcuate mRNA levels of proopiomelanocortin (7), a peptide linked to the anorexigenic actions of PYY[3–36] (61). However, because the acute anorexigenic effects of amylin are retained in agouti mice (7) and the acute effects of PYY[3–36] are retained in mice lacking proopiomelanocortin (62), intact melanocortinergic signaling may not be required for the expression of these effects. Studies examining the interconnecting neuronal pathways regulating food intake and body weight will be useful in identifying other naturally occurring neurohormonal synergies.
Food intake and/or body weight have been successfully reduced after peripheral administration of other peptide/protein combinations. For example, PYY[3–36] and glucagon-like peptide-1[7–36] additively reduced food intake in both lean and obese mice as well as in lean human volunteers (33). Combination therapy with PYY[3–36] and the glucagon-like peptide-1 receptor agonist exendin-4 over a 2-wk time period resulted in synergistic reductions in food intake (37). More recently, we showed that conjunctive treatment with amylin and leptin resulted in synergistic reductions in both food intake and body weight in DIO-prone rats (40). Additionally, calorie restriction (by pair feeding to amylin) alone did not reduce weight to the same extent as amylin plus leptin treatment, suggesting that weight loss synergy was at least partially dissociable from amylins anorexigenic effects (40).
Collectively, the results of these studies highlight the importance of studying neurohormones for potential interactions and suggest that combinatorial neurohormonal approaches may offer a promising, innovative approach to obesity pharmacotherapy.
| Acknowledgments |
|---|
| Footnotes |
|---|
Disclosure Statement: J.R., T.C., C.J., H.M., J.A., C.M., C.W., and D.P. are employed by and own stock in Amylin Pharmaceuticals, Inc.
First Published Online August 30, 2007
Abbreviations: AP, Area postrema; DIO, diet-induced obese; EE, energy expenditure; HSD, Harlan Sprague Dawley; PYY, peptide YY; RQ, respiratory quotient; RSM, response surface methodology.
Received July 3, 2007.
Accepted for publication August 20, 2007.
| References |
|---|
|
|
|---|
This article has been cited by other articles:
![]() |
T. H. Moran and M. J. Dailey Gut Peptides: Targets for Antiobesity Drug Development? Endocrinology, June 1, 2009; 150(6): 2526 - 2530. [Abstract] [Full Text] [PDF] |
||||
![]() |
E. Karra, K. Chandarana, and R. L. Batterham The role of peptide YY in appetite regulation and obesity J. Physiol., January 1, 2009; 587(1): 19 - 25. [Abstract] [Full Text] [PDF] |
||||
![]() |
J. L. Trevaskis, T. Coffey, R. Cole, C. Lei, C. Wittmer, B. Walsh, C. Weyer, J. Koda, A. D. Baron, D. G. Parkes, et al. Amylin-Mediated Restoration of Leptin Responsiveness in Diet-Induced Obesity: Magnitude and Mechanisms Endocrinology, November 1, 2008; 149(11): 5679 - 5687. [Abstract] [Full Text] [PDF] |
||||
![]() |
R. D. Reidelberger, A. C. Haver, P. K. Chelikani, and J. L. Buescher Effects of different intermittent peptide YY (3-36) dosing strategies on food intake, body weight, and adiposity in diet-induced obese rats Am J Physiol Regulatory Integrative Comp Physiol, August 1, 2008; 295(2): R449 - R458. [Abstract] [Full Text] [PDF] |
||||
![]() |
A. L. Mark Dietary Therapy for Obesity: An Emperor With No Clothes Hypertension, June 1, 2008; 51(6): 1426 - 1434. [Full Text] [PDF] |
||||
![]() |
S. R. Bloom, F. P. Kuhajda, I. Laher, X. Pi-Sunyer, G. V. Ronnett, T. M.M. Tan, and D. S. Weigle The Obesity Epidemic: Pharmacological Challenges Mol. Interv., April 1, 2008; 8(2): 82 - 98. [Abstract] [Full Text] [PDF] |
||||
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| Endocrinology | Endocrine Reviews | J. Clin. End. & Metab. |
| Molecular Endocrinology | Recent Prog. Horm. Res. | All Endocrine Journals |