Unintentional weight loss may occur spontaneously in older humans and animals. Further weight losses after surgery or illness in the older patients result in increased morbidity, mortality, and hospital readmission rate. A growing body of work has shown increased appetite and weight gain in response to administration of ghrelin, the “hunger hormone.” We conducted two studies in senescent male Brown Norway rats to assess the ability of peripheral administration of ghrelin to increase body weight and food intake. One study assessed the effect of 2 wk of daily subcutaneous ghrelin administration (1 mg·kg−1·day−1) to senescent rats in a baseline condition; a second study used the same administration protocol in an interventional experiment with aged rats subjected to a surgery with 10–15% blood loss as a model of elective surgery. In both studies, animals receiving ghrelin maintained their body weights, whereas control animals lost weight. Body weight stability was achieved in ghrelin-treated animals despite a lack of increase in daily or cumulative food intake in both experiments. Hormone and proinflammatory cytokine levels were measured before surgery and after 14 days of treatment. Ghrelin treatment appeared to blunt declining ghrelin levels and also to blunt cytokine increases seen in the surgical control group. The ability of peripheral ghrelin treatment to maintain body weights of senescent rats without concomitant increases in food intake may be due to its known ability to decrease sympathetic activity and metabolic rate, perhaps by limiting cytokine-driven inflammation.
- energy balance
- anorexia of aging
ghrelin, an orexigenic peptide hormone produced primarily by the stomach, has wide-ranging actions due to its activation of the growth hormone secretagogue receptor (see Ref. 62, for review). In addition to its ability to increase growth hormone (GH) secretion, ghrelin has other effects on hypothalamic and pituitary cells, pancreatic endocrine and exocrine function, gastric motility and acid secretion, and cardiovascular and immune systems (62). Ghrelin has been shown to stimulate appetite and increase food intake in healthy adult humans (4, 13, 14, 67), as well as patients with cancer (42), heart failure (38–40), kidney failure (70), and chronic obstructive pulmonary disease (37). Similarly, it can increase food intake and weight gain in healthy young and older rats (56, 58, 59) as well as those with experimentally induced cachexia due to cardiac failure (41), cancer (9), kidney disease (8), or anorexia due to injection of lipopolysaccaride (63) or the chemotherapeutic agent cisplatin (18).
Aging in humans and other mammals is associated with failure to maintain homeostasis in response to physiological and environmental disturbances (46, 47, 49, 51). One of the systems that is particularly affected with aging is that which regulates food intake and energy balance, leading to a decreased ability to maintain body weight and lean body mass in response to illness or disease (15, 16, 20, 23, 34, 36, 46, 47). One of the key factors appears to be failure of appetite to increase after weight loss (7, 36, 48, 49). Daily food intake declines over the life span of healthy individuals, whereas body fat and the appetite-suppressing hormone leptin increase, even in weight-stable individuals, leading to the speculation that dysregulation of energy balance unrelated to disease states contributes to the anorexia, wasting, and frailty that occur in the elderly (15, 16, 32, 36, 64, 66).
Studies have shown that healthy older people fail to respond to over- or underfeeding with the compensatory changes in eating that are observed in younger people (34, 46, 47, 49). In addition, older people report less hunger after an overnight fast and report greater degrees of satiation to meals than do younger subjects (7, 49). This apparent insensitivity to metabolic cues can lead to inappropriate weight loss in response to acute or chronic illness or other stressors, including surgery, resulting in greater morbidity and mortality associated with frailty in geriatric populations (15, 20, 36).
No consistent age-related decline of serum ghrelin has been reported during basal conditions in either humans or rodents (1, 45, 53, 54, 58, 65, 71). We have observed that 72 h of food deprivation caused serum ghrelin levels of young rats to nearly double but that this response was blunted in aged animals (65), leading us to speculate that inadequate ghrelin response may contribute to the anorexia of aging.
Americans over 75 years old are the fastest-growing segment of our population (61a). They are living longer and using an increasing number of medical and surgical services. According to the United States National Center for Health Statistics, 2.8 million patients between the ages 65 and 74 yr underwent surgery in 2002–2003, and 4.2 million patients over 75 yr old underwent surgery during the same period (6a). This is a 24% increase in the number of surgeries in patients over 75 yr old compared with a decade ago. Weight loss commonly occurs after surgery in humans and in animals (10, 11, 17, 19, 21). Lack of weight recovery after surgery is associated with increased risk for hospital readmissions (10, 11, 17). Furthermore, continued weight loss initiated by surgery leads to further illness and poor health (16). The importance of good nutrition postoperatively has been known for years (2, 3, 6, 52). Supplemental feedings can decrease morbidity and mortality in hospitalized older patients (44) although it can be difficult to achieve patient compliance with a nutrition program or supplemental feedings when the patient has a lack of appetite (5).
We conducted two studies in senescent male Brown Norway rats to assess the ability of peripheral administration of ghrelin to increase body weight and food intake. One study assessed the effect of daily subcutaneous ghrelin administration to senescent rats in a baseline condition; a second study used the same administration protocol in an interventional experiment with aged rats subjected to a surgery with 10–15% blood loss as a model of elective surgery. To date, there have been no interventional studies examining the ability of ghrelin to increase food intake and the rate of weight recovery after surgery in older animals. We therefore studied the effect of ghrelin administration to older Brown Norway rats after a simulated elective surgery. The “surgery with blood loss” model we developed mimics some of the stressors (general anesthesia, blood loss, and pain treated with analgesia) that humans experience as a result of elective surgery.
We hypothesized that one of the mechanisms by which ghrelin increases appetite is by decreasing proinflammatory cytokine production. Elevated levels of proinflammatory cytokines are associated with anorexia in humans and animals (26, 27, 35, 43, 50, 60). Ghrelin has been shown to decrease both proinflammatory cytokine levels in the plasma of an arthritic rat model as well as the chronic kidney failure model, and in proinflammatory cytokine production by cultured human cells (8, 12, 28, 63). Therefore, we hypothesized that older rats receiving ghrelin after surgery would regain weight more rapidly by increasing food intake relative to controls as a result of decreased proinflammatory cytokine levels.
Aged male inbred, specific-pathogen-free (SPF) Brown Norway BN/Bi (BN) rats were purchased under a National Institute on Aging contract from Harlan Sprague-Dawley (Indianapolis, IN). In this colony, the median life span of male BN rats is 32 mo, and tenth percentile survivors are 36 mo of age (30, 31, 61). These animals do not suffer from pituitary adenomas or glomerulonephrosis common in most strains of laboratory rats and which lead to considerable morbidity in aging individuals. The BN rat provides a good model for healthy human aging and has been used extensively in our laboratory (64–66). Animals were isolated in a dedicated room containing no other animals, in a modified SPF American Association for the Accreditation of Laboratory Animal Care-accredited facility at the Veterans Affairs Puget Sound Health Care System (Seattle, WA). Animals were individually housed in polysulfonate rat cages containing paper bedding (Tek-Fresh; Harlan, Madison, WI) and shredded paper nesting material (EnviroDri) in a light- and temperature-controlled room on a 12:12-h light-dark cycle (lights off from 1800 to 0600; room temperature 74 ± 4°F). Animals had ad libitum access to rodent chow (Teklad Laboratory Rodent Diet NIH-31; Harlan) and tap water. These senescent animals were closely observed by laboratory staff two times daily and were monitored for changes in physical condition and presence of age-related illness. Teeth were regularly examined for malocclusion and clipped when necessary, under brief isoflurane anesthesia. About one-half of these elderly animals did need to have their teeth clipped at least one time; these animals were reexamined every 10–14 days. All animal experiments were conducted in accordance with the principles and procedures outlined in the National Institutes of Health Guide for the Care and Use of Laboratory Animals and were approved by the Veterans Affairs Puget Sound Health Care System Institutional Animal Care and Use Committee.
Synthesized, highly purified (purity ≥95% by analytical HPLC; Global Peptide Services, Fort Collins, CO) rat ghrelin [H-Gly-Ser-Ser(n-Octanoyl)-Phe-Leu-Ser-Pro-Glu-His-Gln-Lys-Ala-Gln-Gln-Arg-Lys-Glu-Ser-Lys-Lys-Pro-Pro-Ala-Lys-Leu-Gln-Pro-Arg-OH] was dissolved in sterile saline, aliquoted, and stored at −20°C. Ghrelin efficacy was determined by intracerebroventricular administration in a separate group of rats. Aliquots in use were stored at 4°C for use within 5 days. Ghrelin (1 mg·kg−1·day−1∼0.3 μmol·kg−1·day−1), or saline as control, was injected subcutaneously in two divided doses daily for both experiments. Pilot studies conducted earlier (unpublished observations) indicated that subcutaneous treatment was almost as effective in increasing short-term food intake as intraperitoneal injections and was better tolerated by the animals. Additionally, this route of administration has more relevance to human treatment paradigms.
Experiment 1: Chronic ghrelin baseline study.
Twenty-one rats (31 mo of age) were allowed 1 mo of conditioning to single housing, daily handling, and daily food and body weight measurements before experimental procedures began. During the conditioning period, food was weighed 2 h after lights-on (at 8:00 A.M.), and body weights were measured 2 h before lights-off (at 4:00 P.M.).
Later (1 mo), these same elderly rats (now 32 mo old) were randomly divided into two groups. Animals received two subcutaneous injections a day (at 8:00 A.M. and 4:00 P.M.) of either saline (n = 10) or ghrelin (n = 11; 1 mg·kg−1·day−1) for 17 days. Food intake was measured at 8:00 A.M. 9:00 A.M., and 4:00 P.M. (1, 7, and 24 h after the first injection of the day). Body weights were measured at 4:00 P.M. Pretreatment body weights of animals in ghrelin groups were 478 ± 13 g; those of saline control animals were 481 ± 13 g. These animals were not killed at the end of the study, and no blood samples were taken.
Experiment 2: Surgical intervention with chronic ghrelin.
A separate group of 40 rats (28 mo of age) was allowed 2 wk of conditioning to single housing, daily handling, and daily food and body weight measurements, as in experiment 1. Animals were then randomly assigned to one of the following three experimental groups: surgery + ghrelin (n = 15; presurgical body wt 499 ± 10 g), surgery + saline (n = 14; presurgical body wt 506 ± 13 g), or nonsurgical control (n = 10; pretreatment body wt 481 ± 20 g). Animals in the two surgical groups underwent general anesthesia with metered isoflurane gas anesthesia, followed by cut down of the external jugular vein, from which blood was drawn (1% of body wt, equivalent to 15% of total blood volume; from 4 to 6 ml) via a temporary indwelling catheter. Total surgical time was 90 ± 30 min, and animals were kept warm by a heated surgical table. After the blood was drawn, the incision was closed with absorbable sutures (Dexon 5), and animals were kept warm on circulating heated water pads until awake and ambulatory (usually within 30 min). Postoperative pain relief (ketoprofen) was provided for 3 days, per facility protocol.
Experimental rats received subcutaneous injections of ghrelin (1 mg·kg−1·day−1) or saline two times a day for 14 days, beginning the afternoon of surgery. Nonsurgical control animals received twice-daily injections of saline to control for handling and injection stress. Body weights and food intake were monitored two times daily. After 2 wk, animals were killed by decapitation under brief isoflurane anesthesia. The trunk bloods were collected for serum and plasma hormone assays and processed within 2 h. All abdominal fat pads (epididymal, retroperitoneal, mesenteric, perirenal, and omental) were dissected and weighed, as were skeletal muscles of the rear leg (biceps femoris, quadriceps, and rear calf muscles), as previously described (66). Results are presented as the sums of these fat and muscle groups.
Plasma and serum samples were separated from blood of nonfasted rats obtained during the surgery and from trunk blood at the time of death and stored at −20°C. Samples were collected during the middle of the light cycle (generally between 10:00 A.M. and 2:00 P.M.). Hormone and proinflammatory cytokine assays were performed by our laboratory (author Marck) on duplicate samples and within the same assay. Plasma samples for RIA analysis of active ghrelin were acidified with 50 μl of 1 N HCl and 10 μl of phenylmethylsulfonyl fluoride per 1 ml of serum before being stored at −20°C, per manufacturer (Linco Research, St. Louis, MO) recommendations. Active and total ghrelin, leptin, insulin, and insulin-like growth factor (IGF)-I levels were determined by double-antibody RIA kits [rat active ghrelin: GHRA-89K; rat total ghrelin: GHRT-89K; rat leptin: RL-83K; rat insulin: RI-13K (Linco Research); DSL-2900 mouse/rat IGF-I (Diagnostic Systems Laboratories, Webster, TX)]. The assay detection limit was 10 pg/ml for active ghrelin and 0.16 ng/ml for rat total ghrelin. The detection limit of the rat leptin assay was 0.5 ng/ml; intra-assay variability was 7.2%. The assay detection limit for rat insulin was 0.1 ng/ml, and the intra-assay variability was 3%. The IGF-I assay was performed on ethanol-extracted serum; the detection limit was 21 ng/ml, and intra-assay variability was <10%.
Analyses of proinflammatory cytokines interleukin (IL)-1β, IL-6, and tumor necrosis factor (TNF)-α were performed using Bio-Plex multiplex bead-based assay protocols (Bio-Rad Laboratories, Hercules, CA). The assay detection limit for IL-1β, IL-6, and TNF-α was 2, 2, and 4 pg/ml, respectively. The intra-assay coefficient of variation for IL-1β, IL-6, and TNF-α was 10%, and the interassay coefficient of variation for IL-1β, IL-6, and TNF-α was 30%.
All data are presented as means ± SE. Comparisons between placebo and ghrelin-treated groups and between surgical and control treatment groups were made using ANOVA; the level of significance was set at P < 0.05. Post hoc testing by Fisher's protected least-significant difference was performed for between-treatment differences.
Sample size power calculation was based on pilot data which showed that 10 animals in each group would give 100% power to detect 2.5% difference in percent weight gain between the control and experimental animals with a significance level of 0.05 (α = 0.05). For the surgical experiment, 15 rats/surgical group were entered in the study to accommodate possible postsurgical or spontaneous deaths over time in these senescent rats. The statistical software package used was StatView Version 4.0 for Macintosh (Abacus Concepts and SAS Institute, Berkeley, CA).
Experiment 1: Chronic ghrelin baseline study.
Animals receiving subcutaneous ghrelin maintained their body weights throughout the 17-day period of treatment (Fig. 1, top), whereas animals receiving saline lost a maximum of ∼3% body weight. Body weights were significantly different between groups beginning on the 3rd day of treatment and remained significant for 3 days after treatment ended. After that time, body weights of the ghrelin-treated animals declined, reaching the weight of control animals within 10 days after treatment ended. Daily and cumulative food intakes were not affected by ghrelin administration (Fig. 1, bottom), although ghrelin caused a slight (16%) increase in food intake (FI) 1 h after injection (average 1 h FI ANOVA P = 0.06; ghrelin 1.63 ± 0.11 g, saline 1.36 ± 0.09 g).
Experiment 2: Surgical intervention with chronic ghrelin.
Animals receiving subcutaneous ghrelin for 2 wk after surgery maintained their body weights compared with both saline-injected groups (Fig. 2, top). Body weights, as a percentage of presurgical baseline, were significantly lower in surgical rats receiving saline compared with both other groups, beginning on the second day after surgery and continuing throughout the study, whereas animals receiving ghrelin after surgery had decreased body weights for only 2 days after surgery compared with nonsurgical control animals. Maximum weight loss (Fig. 2, bottom) was not significantly different between the surgical groups, and both were significantly greater compared with nonsurgical controls (ANOVA P < 0.0001). The final body weight change (Fig. 2, bottom) was negative for both saline-treated groups and positive for ghrelin-treated animals (ANOVA P < 0.0001). Nonsurgical control animals lost a small amount of weight (1.5%, ANOVA P < 0.02 compared with saline-treated surgical group; P = 0.052 compared with ghrelin-treated surgical group) over the course of the experiment because of their advanced age and the stress of twice-daily handling and injections.
Daily food intake was significantly lower (P < 0.01) in surgical groups compared with control animals for the first 3 days after surgery (Fig. 3, top). As in experiment 1, there were no significant differences in cumulative food intake (Fig. 3, bottom) between the ghrelin and control groups.
There were no significant differences in visceral body fat or hindlimb skeletal muscle mass (either in total weight or as a percentage of body weight) at the end of 2 wk of treatment in animals treated with ghrelin compared with surgical or nonsurgical control animals (Fig. 4).
Nonfasted hormone levels before and after treatment are presented in Table 1, with ANOVA values listed for each hormone at each time point. Samples taken before treatment were collected at the time of surgery and contained some heparin used to facilitate collection through the catheter; this may have altered the binding characteristics for some of the assays (specifically active and total ghrelin and perhaps IL-6 and TNF-α). Samples taken after treatment were from trunk blood ∼20 h after the last injection. There were no significant differences between surgical groups for any hormone before treatment began (Table 1). After treatment, both active and total ghrelin levels were significantly higher (∼2-fold) in the ghrelin-treated group compared with saline-treated groups, which were similar to each other. Ghrelin levels in the treated group decreased nonsignificantly from baseline despite ghrelin treatment. There were no significant differences in leptin, insulin, or IGF-I levels after 2 wk of treatment (Table 1); however, within-animal changes in IGF-I levels before and after treatment were significantly different between saline- and ghrelin-treated rats. IGF-I levels were increased in saline-treated animals, whereas those of ghrelin-treated animals did not change.
Proinflammatory cytokine levels.
Cytokine levels before and after treatments are presented in Table 2, with ANOVA values listed for each cytokine at each time point. There were no significant differences between treatment groups in any of the cytokines in blood samples taken at the time of surgery (before treatment), although animals in the group that was randomized to ghrelin treatment had nonsignificantly higher initial levels than those that would later receive saline (Table 2).
Levels of IL-1β decreased after 2 wk of ghrelin treatment but increased in postsurgical animals receiving saline. Serum IL-1β levels were similar in nonsurgical saline control animals and those receiving ghrelin. IL-6 and TNF-α levels increased in both surgical groups after 2 wk of treatment compared with pretreatment samples; the increase in IL-6 was significantly blunted in animals receiving ghrelin (Table 2).
The primary finding of this study is that chronic subcutaneous treatment with ghrelin prevented weight loss in senescent BN rats, without significantly increasing daily or cumulative food intake. Two experiments were performed. In the first, ghrelin treatment temporarily halted the slow, spontaneous decline in body weight that occurs in senescent animals; however, this effect disappeared soon after ghrelin treatment ended, and the body weight trajectory rapidly declined until matching the body weights of saline-treated animals. In the second experiment, interventional postsurgical treatment with ghrelin rapidly overcame weight loss that occurred after our surgery with blood loss model. These experiments show conclusively that peripherally administered ghrelin can halt the declining body weight common in senescent rats, both in the basal state and in response to the physiological stressor of a simulated elective surgery. Although we did not measure body composition per se, we assessed visceral adiposity by dissecting all abdominal fat pads and found that animals assigned to interventional ghrelin treatment initiated after surgery had neither more nor less intra-abdominal fat than animals treated with saline at the conclusion of the experiment. Hindlimb skeletal muscle mass was also not altered by treatment or surgery, despite overall changes in body weight. Although it appears that ghrelin treatment did not affect overall body composition, we can't rule out an effect of ghrelin to increase subcutaneous fat, which comprises a much larger pool of adipose tissue than the visceral fat pads that we did weigh.
Although ghrelin is known as “the hunger hormone,” its effects on food intake are frequently minimal when given peripherally. To facilitate translation of our findings to humans, we administered ghrelin subcutaneously rather than by intracerebroventricular injection, where the clearest and strongest effects on feeding have been shown (26, 29, 56, 59, 68). In the current experiments, daily and cumulative food intakes were not increased by ghrelin treatment, despite a modest 16% increase noted in the 1st h after the morning ghrelin injection (when rodents typically are not eating). Experiments in human subjects from the Bloom laboratory have also consistently shown acute increases in food intake after subcutaneous injection (13, 70) or intravenous infusion (42, 67) that are not sustained. Other animal studies have also shown a lack of effect of similar doses of peripheral ghrelin on overall food intake, despite beneficial effects on body weight (e.g., see Ref. 18).
Therefore, ghrelin's ability to maintain body weight without increasing food intake indicates that the effect may be due to a blockade of lipolysis, or to decreased physical activity or metabolic rate. That ghrelin can maintain body weight at a dose that did not lead to increased food intake points to ghrelin's known role in modulating energy expenditure. Ghrelin has been shown to decrease measures of sympathetic activity in animals and humans (29, 33, 37, 38, 40, 57) and to decrease spontaneous physical activity in rats (56). Further studies of the effect of ghrelin on activity and metabolism are warranted.
The lack of effect of ghrelin treatment on IGF-I levels was somewhat surprising, in light of the GH secretagogue effect of ghrelin, since GH tends to increase IGF-I levels. A similar result was seen in a different group of senescent animals, but not in younger individuals undergoing a similar treatment paradigm (unpublished data, Wolden-Hanson), and other studies of peripheral ghrelin administration in rats (8, 9, 18) also failed to elicit the expected increase in IGF-I levels.
Interventional treatment with ghrelin appeared to blunt the declines in active and total ghrelin and to blunt the increases in cytokine levels that resulted from surgery. It must be noted that the source of sampled blood was different between the pretreatment samples obtained at surgery and those collected from trunk blood at the end of the experiment. Thus some of the pre- and posttreatment differences in active ghrelin and the proinflammatory cytokines may be due to effects of heparin or longer-term anesthesia used during the surgical blood collection. Further study of the effects of ghrelin on proinflammatory cytokines is warranted.
Elevations in proinflammatory cytokines (IL-1β, IL-6, and TNF-α) have been associated with experimental and clinical anorexia (26, 35, 43, 60) and with increased resting energy expenditure in patients with end-stage renal disease and human immunodeficiency virus infection (27, 50). Ghrelin has been used with success to treat both animals and humans with conditions that are associated with elevated cytokine levels, such as cancer cachexia (9, 22, 42), cardiac cachexia (14, 38–41), lipopolysaccharide injection (63), sepsis (69), renal failure (55), and chronic kidney disease (8). Thus ghrelin's effects on body weight and survival may be secondary to salutary effects on cytokine levels.
Perspectives and Significance
We found that subcutaneous injections of ghrelin completely abolished the 2–3% body weight loss experienced by control senescent BN rats after surgery. Although this effect may appear to be small, our surgical model was minimally invasive, and our period of observation was relatively short. In geriatric patients who undergo more extensive surgery, or who experience postoperative complications leading to greater degrees of weight loss, any effect of ghrelin to speed the recovery of preoperative weight may confer a significant clinical benefit. Therefore, the translation of our results into the clinical arena should receive high priority. Because we observed no differences in food intake between treated and control animals, further study will be necessary to determine the mechanism by which ghrelin abolishes postoperative weight loss. Our data suggest that the proinflammatory cytokine response to surgery may be diminished by ghrelin treatment, and it is conceivable that diminished inflammation may promote weight regain by reducing overall energy expenditure. This might be another salutary effect of ghrelin that deserves investigation in older patients undergoing surgery.
This work was supported by a Veterans Affairs MERIT Review grant (T. Wolden-Hanson) and National Institute on Aging Grant 1K23AG-020196-01A2 (M. Yukawa).
We acknowledge research development and design support by Dr. Alvin M. Matsumoto of the Veterans Affairs Puget Sound Health Care System Seattle Division and of the Division of Gerontology and Geriatric Medicine, University of Washington School of Medicine. We are also grateful for the laboratory assistance provided by Barbara Anderson and Dr. LeBris Quinn of the Veterans Affairs Puget Sound Health Care System American Lake Division.
Work in this report has been presented and published, in part, at the 89th Annual meeting of the Endocrine Society, Toronto, Ontario, Canada (June 2–5, 2007).
The costs of publication of this article were defrayed in part by the payment of page charges. The article must therefore be hereby marked “advertisement” in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.
- Copyright © 2008 the American Physiological Society