Investigating the Effect of Enterally Administered Capromorelin on Body Weight in Mice (Mus musculus)

Abstract

Significant weight loss in mice (Mus musculus) is a welfare concern and can alter physiology and behavior in ways that may confound research aims. In this study, factorial design was used to investigate the effect of enterally administered capromorelin on changes in mouse body weight overall and with various research-related interventions, such as administration of analgesics, anesthesia, or surgery. BALB/c mice (n = 61 [27 males/34 females] for analysis) were randomized into 8 intervention-treatment groups with 2 treatment allocations: capromorelin (10 mg/kg) or control, and 4 intervention allocations: no intervention; buprenorphine extended-release (XR) alone; buprenorphine XR, meloxicam, and anesthesia; or surgery under anesthesia with buprenorphine XR, meloxicam, and bupivacaine administered. Mice were habituated to handling, weighing, and voluntary consumption of condensed milk, which was used as the control solution and later a vehicle for capromorelin delivery, for 5 d (days 0 to 4). Then, mice received their interventions followed by 3 days of daily treatment or control administration (days 7 to 9). Body weights were measured daily (days 8 to 11 and day 14) to compare with baseline weights (days 0 to 4 and day 7) and evaluate for treatment and intervention effects on body weight. The interventions resulted in a decrease in group body weights 3 and 4 d after the interventions were conducted. Overall, body weights increased more in mice given capromorelin compared with control, and mice treated with capromorelin returned to, or exceeded, baseline weights faster. The weight loss was mitigated by capromorelin administration in all interventions except for the buprenorphine XR-only group. It is recommended to clinically consider enterally administered capromorelin to mitigate research-induced weight loss in mice.

Introduction

An estimated 111.5 million mice and rats are used in research each year in the United States, comprising approximately 95% to 99% of all research animals used.^5,21 Mice have a high rate of metabolism due to the metabolic demand of their surface area-to-body mass ratio and metabolically active tissues, such as the liver, kidney, and brown fat deposits.^21,34,45 The high rate of metabolism in mice combined with their small body mass leaves them vulnerable to significant, rapid weight loss. Variations in body weight and composition can alter physiology and behavior, potentially masking or confounding the primary research findings.^30,45 Natural or induced stressors can cause weight change with normal appetite or weight change from altered food consumption.^{7,17,18,20,24,43,45,46} Common interventions that may cause stress to mice in a research setting include induction of disease or trauma, administration of a medication or treatment, surgery, changes in husbandry practices, and numerous others. Specifically, label warnings for a recently indexed extended-release buprenorphine injectable suspension (Ethiqa XR; Fidelis, North Brunswick, NJ) include decreased gastrointestinal motility and weight loss in mice treated postprocedurally,¹⁵ which is consistent with other reports of side effects after administration of opioids to rodents.^1,2,33,39 Therapeutic procedures for weight loss in mice include treating the underlying cause, if identified, fluid therapy, dietary enrichment, and thermal support.¹⁸ Appetite stimulants are not routinely used in the medical care of mice. A body-weight reduction of greater than 15% to 20% in mice is often used as an endpoint to prompt humane euthanasia,⁴⁶ potentially leading to early termination before the study end.⁴³

Capromorelin is a centrally penetrant ghrelin receptor agonist (GhrRA). Ghrelin is secreted predominantly by the stomach and acts systemically via neuroendocrine mechanisms to modulate appetite and stimulate food intake.^6,12,31,41 The weight gain associated with capromorelin via ghrelin is postulated to be due to more than just excess calorie consumption; alterations in energy expenditure, substrate utilization, and hormonal regulation likely play a significant role in the observed weight gain.^8,37,41,44 Recently, capromorelin was FDA approved for appetite stimulation in dogs and for management of weight loss in cats with chronic kidney disease, and it is now available commercially under the product names Entyce and Elura, respectively. Entyce and Elura are both labeled for once-daily oral dosing.^13,14 Studies using capromorelin and other GhrRAs in mice and rats demonstrated short-term effects on gastrointestinal emptying times, increased food consumption, and increased fecal output.^10,28,29,36 Further, GhrRAs have been shown to mitigate the effect of opioid administration on gastrointestinal motility in mice.¹⁰ The use of capromorelin has been explored in several other species, including rabbits, chickens, and nonhuman primates, with various dosing regimens, for its effects on food intake, gastrointestinal motility, and body weight.^4,9,11 The study in rabbits evaluated the effect of capromorelin on body weight when given once daily for 3 d.¹¹ Capromorelin and other GhrRAs have been studied in humans for various indications related to weight and gastrointestinal motility, among others.^3,41,42,48

Prior research has not addressed the influence of enterally administered capromorelin on body weight in mice nor its ability to counteract weight loss arising from research interventions. It was hypothesized that capromorelin would have a significant effect on mouse weight with the capromorelin-treated mice gaining more weight than the untreated controls. It was postulated that the positive effect of capromorelin on body weight would be evident in mice with no intervention and those with various treatments including buprenorphine XR administration, isoflurane anesthesia with buprenorphine XR and meloxicam administration, and surgery under anesthesia with buprenorphine XR, meloxicam, and bupivacaine administered.

Materials and Methods

This research was conducted under a United States Army Medical Research Institute of Infectious Diseases (USAMRIID) IACUC-approved protocol in compliance with the Animal Welfare Act, Public Health Service Policy on Humane Care and Use of Laboratory Animals, and other Federal statutes and regulations relating to the use of animals in research. Animals were housed in an AAALAC, International accredited facility that adheres to principles stated in the Guide for the Care and Use of Laboratory Animals.²³

Animals.

Sixty-four BALB/c mice (32 male and 32 female; age 70 to 84 d) were acquired from Charles River Laboratories (Wilmington, MA). Shortly after arrival, 2 mice, one male and one female, died of unknown causes, and one male mouse was euthanized due to fight wounds. An additional 6 mice (2 male, 4 female, age 84 to 90 d) were acquired due to attrition from the original order. Sixty-seven mice were used in the habituation phase. After the habituation phase, one male mouse was found dead, and 2 more were euthanized due to fight wounds. Sixty-four mice received interventions: 29 males and 35 females. Three mice died or were humanely euthanized during the intervention phase of the study: one male mouse from the treated group B^t, one female mouse from the treated group I^t, and one male mouse from the treated group I^t (see Interventions for group allocations). All mice that were found dead or euthanized during the intervention phase were sent for pathologic analysis with an undetermined cause of death. The mice that were lost during the intervention phase were excluded from statistical analysis so as to not confound a change in weight due to clinical decline with a change in weight due to intervention or treatment. Therefore, 61 mice were included in the data analysis: 27 males and 34 females.

The mice arrived to the Institute with documentation that they originated from a colony free of murine respirovirus (Sendai), pneumonia virus of mice, mouse hepatitis virus, murine norovirus, mouse parvoviruses, reovirus, epizootic diarrhea of infant mice, mouse encephalomyelitis virus, ectromelia virus, lymphocytic choriomeningitis virus, murine adenovirus 1 and 2, murine cytomegalovirus, K virus, polyoma virus, hantavirus, lactate dehydrogenase-elevating virus, murine chapparvovirus, Bordetella brochiseptica, Citrobacter rodentium, Corynebacterium bovis, Filobacterium rodentium, Mycoplasma pulmonis, Salmonella spp., Clostridium piliforme, Streptobacillus moniliformis, Streptococcus pneumoniae, Rodentibacter pneumtropicus, Corynebacterium kutscheri, Rodentibacter heylii, Helicobacter bilis, Helicobacter hepaticus, and pathogenic endoparasites, helminths, and protozoa. There are no known pathogens endemic in the USAMRIID mouse colony. The mice were held on protocol with no manipulation for 4 wk (first cohort) and 2 wk (additional 6 animals) to allow for acclimation and growth to a more adult body size. Body weights after acclimation and before randomization ranged from 20.51 g to 31.50 g (mean ± SD = 25.767 ± 3.419 g [males: 28.920 ± 1.762 g/females: 22.883 ± 1.395]). After randomization, the mice were housed in groups of 4 to 6 mice, except when fighting necessitated removing males for individual housing (19 males); all intervention-treatment groups had at least one male removed resulting in remaining male groups of 2 to 3 mice. The microisolators were maintained in the same position on the rack after randomization, unless mice were separated after randomization.

Mice were housed in individually ventilated microisolation caging (Techniplast, West Chester, PA) with pressed cellulose bedding (Alpha-dry; Shepherd Specialty Papers, Kalamazoo, MI). Their cages included an assortment of enrichment items such as crinkled paper (Enviro-dri; Shepherd Specialty Papers, Kalamazoo, MI), tissue paper (Certified Nesting Sheets; Bio-Serv, Flemington, NJ), Manzanita gnaw sticks (Bio-Serv, Flemington, NJ), cotton nesting squares (Nestlets; Ancare, Bellmore, NY), and plastic igloos and forts (Bio-Serv, Flemington, NJ). Microisolator cage bottoms and disposable enrichment items were changed weekly. Microisolator lids, durable enrichment devices, and cage card holders were changed every 2 wk. The mice were fed a standard laboratory rodent diet (5001 Rodent Diet; Lab Diet, Brentwood, MO) ad libitum. They were provided filtered (5 μm) domestic water via Lixits ad libitum. Water from animal rooms is tested quarterly (DoD Food Analysis and Diagnostic Laboratory, San Antonio, TX) for heavy metals, chlorinated hydrocarbons, nitrates, and microbial contamination, with no aberrant results on all reports reviewed for the year prior to this study. Mice were offered a standard volume of forage mix consisting of mixed grain cereals twice weekly. Housing rooms were maintained on a 12:12-h light:dark cycle (lights on at 0600 h, lights off at 1800 h) with fluorescent lights. Humidity was controlled between 30% and 70% and temperature was controlled between 68 and 76 °F (set point 74.5 °F). The mice were checked no less than twice daily by animal care and veterinary staff to ensure adequate health and welfare.

Experimental design.

A prospective, randomized factorial-design experiment was conducted to study the effects of capromorelin on body weight in mice that received specific medical and/or surgical interventions.

One week before study activities started, the mice were weighed, ear tagged, and randomized using SAS Proc Plan (SAS 9.4; SAS Cary, NC) into one of 8 groups balanced by sex and weight. The statistician who performed randomization was blinded to the intervention and treatment groups. The 4 intervention groups were defined based on interventions the mice would receive on day 7.

There were 2 treatment allocations: control (o) and capromorelin (t). Each intervention group was balanced with equal numbers of animals assigned to receive either control or capromorelin, for a total of 8 intervention-treatment groups. Control-assigned animals received 0.1 mL condensed milk only, and capromorelin-assigned animals received 0.1 mL condensed milk with capromorelin oral solution (Elura, 20 mg/mL; Elanco, Greenfield, IN) 10 mg/kg mixed into the condensed milk.

Initially, males and females were equally represented in each group. When a mouse died from a group before interventions, another mouse of the appropriate sex from the pool of animals that had not been assigned was randomized to the corresponding group that had lost a member. However, more males died or were euthanized due to fighting during the initial acclimation and habituation periods without enough unassigned males remaining to assign to the corresponding group, resulting in imbalanced groups. To compensate for male attrition and to maintain statistical power between intervention groups, additional females were added to the corresponding female intervention groups from which the males were lost. The imbalance of the male-to-female ratio due to disparities in losses between the sexes barred analysis of the intervention or treatment effect based on sex.

Habituation.

During the 5-d habituation period (days 0 to 4), mice were individually weighed and subsequently offered 0.1 mL condensed milk from a 20-gauge, 38-mm plastic gavage needle (Instech, Plymouth Meeting, PA). Habituation started at approximately 0900 daily. Each mouse’s daily acceptance of condensed milk was documented. Mice that did not voluntarily consume condensed milk during the habituation period did not receive the dose via gavage. The mice were weighed individually on the same digital scale, and weights were recorded to the hundredths. There was a scheduled 2-d period between the habituation period and when interventions and treatments started on day 7.

Interventions.

Daily body weights were measured and used to calculate intervention dosages (day 7) and to calculate capromorelin dosages (days 7 to 9), as applicable. The individuals obtaining weights were blinded with regards to treatment assignment. Mice in intervention group N (none) received no interventions. Mice in intervention group B (buprenorphine XR) received 3.25 mg/kg buprenorphine XR injectable suspension (Ethiqa XR, 1.3 mg/mL; Fidelis Animal Health, North Brunswick, NJ) subcutaneously. Mice in intervention group I (isoflurane) received 3.25 mg/kg buprenorphine XR injectable suspension subcutaneously and 6 mg/kg meloxicam (Alloxate, 5 mg/mL; Pivetal, Liberty, MO) subcutaneously after anesthetic induction with isoflurane. Mice in intervention group S (surgery) received 3.25 mg/kg buprenorphine XR injectable suspension subcutaneously and 6 mg/kg meloxicam subcutaneously immediately after anesthetic induction and a 3-mg/kg bupivacaine hydrochloride (2.5 mg/mL; Hospira, Lake Forest, IL) splash block after surgical incision.

Anesthetic and surgical procedure.

Mice in groups I and S were anesthetized with isoflurane in 100% oxygen (3% to 5% isoflurane induction; 0.5% to 3% isoflurane maintenance). Isoflurane was delivered using a precision vaporizer via a rodent anesthesia machine. Induction occurred in a clear viewing chamber. Following induction, anesthesia was maintained via a nose cone for group S, and in the induction chamber for group I. Lubricating eye ointment was applied to group S mice, but not group I mice, so as to not open the induction chamber; no ocular issues were noted in any anesthetized mice.

Group S anesthetized mice were aseptically prepared for surgery from the xiphoid process to the pubis. Using an aseptic technique, an approximately 1-cm midline incision was made extending from the level of the umbilicus caudally through the skin, the bupivacaine splash block was administered, and a stab incision was made through the body wall to expose the peritoneal cavity. No procedures were performed in the peritoneal cavity. The body wall was sutured closed with one or 2 simple interrupted sutures with 5-0 polydioxanone suture swaged on a 17-mm, 1/2c taper-point needle (Ethicon; Johnson-Johnson, Bridgewater, NJ), and the skin was closed with surgical skin glue to fully approximate tissues. The veterinarian who performed all surgical procedures was blinded to treatment assignment on the day of surgery.

Group S animals remained anesthetized for the time required to complete the surgical procedure. Surgery time ranged from 8 to 20 min, with a median surgery time of 13 min. Each mouse in intervention group I was anesthetized for the average number of minutes that the group S mice were under anesthesia, rounded to the nearest whole minute; therefore, group I mice were maintained under anesthesia for 13 min. All mice were monitored and provided thermal support during the anesthetic period and postanesthesia until fully ambulatory. Mice all recovered from anesthesia within 3 to 6 min. The mice were observed approximately 4 h after surgery to ensure the integrity of the surgery site and overall recovery.

Treatments and body weight measurement.

Mice were given their assigned treatment for 3 d (days 7 to 9), approximately 24 h apart starting around 0900. The capromorelin dose was extrapolated from a study in rats³⁶ and the dosing frequency from a study in rabbits¹¹ and label indications for the capromorelin formulation.^13,14 Capromorelin was dosed 10 mg/kg with a 100-µL variable volume pipette, and the dose was mixed into the distal end of the condensed milk to ensure the mouse received the entire dose and none was lost in the gavage tip hub. All mice were offered their treatment for voluntary consumption through the gavage tip, and only if not consumed voluntarily was the dose administered via oral gavage. Each mouse’s route of daily treatment administration, voluntary compared with gavage, was documented. The personnel administering the treatments were blinded. Mice were given their first treatment on day 7 after receiving their assigned interventions and fully recovering from anesthesia, as applicable.

Body weights were obtained at approximately 0730 daily ± 30 min: before interventions on day 7, before the administration of treatment on days 8 and 9, and approximately 24 h (day 10), 48 h (day 11), and 120 h (day 14) after the last treatment was given on day 9. The personnel administering the treatments and measuring weights were blinded to the treatment allocations but not the intervention groups. The mice were observed by study personnel daily on days 7 to 11 and day 14, and their appearance and behavior were scored using a standardized chart to ensure welfare and identify adverse outcomes. Each mouse’s appearance and behavior were scored from 0 to 3, with 0 being normal and 3 being hunched with piloerection and immobile. Mice that received surgery also had their surgery sites inspected for evidence of infection or dehiscence. Measurements and treatments were conducted in the same order each day, starting with the treated group N females and finishing with untreated control group S males.

Statistical analysis.

The study design had the following 4 factors: capromorelin treatment, buprenorphine XR, anesthesia + meloxicam, and surgery + bupivacaine. A full factorial design was considered to allow for the examination of multiple 2- and 3-way interaction effects with a reduced sample size. However, many of the possible factor combinations would have been unethical to perform. The main treatment effect comparing capromorelin with condensed milk only (control) was deemed the most important factor for investigation. Four ethically permissible combinations of the other 3 factors were identified and together were defined as the intervention factor. The intervention factor had 4 levels: 1) no intervention (control); 2) buprenorphine XR alone; 3) buprenorphine XR + anesthesia + meloxicam; and 4) buprenorphine XR + anesthesia + meloxicam + surgery + bupivacaine. Interactions between the treatment effect and the ethically permissible interventions were examined using a 2 × 4 factorial design, with specific pairwise intervention comparisons of interest as follows: 1) no intervention compared with buprenorphine XR alone; 2) buprenorphine XR alone compared with buprenorphine XR + anesthesia + meloxicam; and 3) buprenorphine XR + anesthesia + meloxicam compared with buprenorphine XR + anesthesia + meloxicam + surgery + bupivacaine.

A power analysis of the primary outcome measure of weight was conducted using PASS 14 (NCSS, Kaysville UT) before the start of the study for the specific pairwise comparisons of interest. A factorial design with 2 factors, each with 2 levels, resulting in 4 possible combinations, was used to test whether there are differences among the levels of the factors. Terms were to be tested using a linear model analysis of variance F test with a type I error rate (α) of 0.05. With a sample size of 8 subjects per treatment combination, and a total number of subjects is 16, this design achieved greater than or equal to 78% power to detect an effect size of 0.5 for each factor and an effect size of 1.0 for the interaction effect.

Mixed model ANOVA was performed using the GLIMMIX procedure of SAS Version 9.4 (SAS Institute, Cary, NC) to examine the effects of treatment, intervention, and interaction thereof on daily posttreatment animal weights as the primary outcome measure. The model included the random effects of mouse and day, as well as the covariant of baseline weight, which was defined as the animal’s average weight over days 0 to 4 and day 7 (before initiation of treatment). Because the time points (days) between posttreatment weight observations were not uniform, a power covariance structure was used. Orthogonal contrasts were used to test the effects of treatment within interventions, as well as between interventions as previously specified in the discussion of study design. The Kenward and Roger method²⁶ was used to calculate degrees of freedom. Sex and use of gavage were examined as possible covariates but were found not to contribute to the model. Model assumptions were checked using the Shapiro-Wilk normality test⁴⁰ and by visual inspection of residual and fitted value plots. Descriptive statistics are presented as means, SDs, and minimum and maximum observed values.

Kaplan-Meier failure analysis²⁵ was performed using the LIFETEST procedure of SAS to examine the effects of treatment, intervention, and interaction thereof on the time in days required to return to or exceed (RoE) baseline weight after initiation of treatment. The event of interest was the recording of weight equal to or greater than the animal’s average weight over days 0 to 4 and day 7 (before initiation of treatment). Animals failing to RoE baseline weight were right censored. Log-rank tests³⁵ were used to compare the cumulative failure distributions. Descriptive statistics are presented as median time to RoE baseline weight, means and standard errors of time to RoE baseline weight, number and percentage of animals RoE baseline weight, and number and percentage of animals failing to RoE baseline weight (censored).

The level of significance was set at P < 0.05. All tests were 2 tailed. No adjustments for multiple comparisons were made. Missing values were treated as missing at random, and values were not imputed. The total number of mice used for analysis was 61. The statistician performing the analysis was not blinded.

Results

Study population.

The weights obtained during habituation (days 0 to 4) and the day of interventions (day 7), before intervention or treatment, were used as the baseline weights by which to evaluate treatment outcome (Figure 1). Mice gained a small amount of weight (less than 10%) over the habituation period (P < 0.001); however, the rate of change over the period did not differ by intervention (P = 0.3321) nor treatment assignment (P = 0.5509). There were no statistically significant differences in baseline weight between treatment assignments (P = 0.9386), intervention groups (P = 0.8943), or any intervention-by-treatment group combinations (P = 0.8385).

Figure 1.

Intervention group body weight, in grams, over the baseline period (days 0 to 4 and day 7) by treatment allocation. Boxplots represent the median and IQR, the marker represents the mean, and the range is denoted by the whiskers. There were no statistically significant differences in baseline weight between treatment assignments (P = 0.9386), intervention groups (P = 0.8943), or any intervention-by-treatment group combinations (P = 0.8385).

Changes in weight.

There was a significant difference between the percentage of capromorelin-treated mice compared with controls that required gavage (P = 0.0416) but not in the percentage of gavage-events for capromorelin-treated mice compared with controls (P = 0.0546). When gavage was entered into the model for weight change, the method of treatment administration was not a significant factor that affected the change seen in weight (P = 0.9078). The effect of sex was not significant (P = 0.0615) when entered into the model for weight change; females have lower weight than males, but sex did not have a significant interaction with treatment or intervention.

Treatment effect.

A comparison between treatments inclusive of all intervention groups N through S showed a significant increase in body weight in mice given capromorelin as compared with controls, over the period starting one day (day 8) after initiation of capromorelin administration and continuing until the study concluded 7 d postintervention (day 14) (P_{days 8 to 14} < 0.0001). In addition, the treatment effect was significant on each of the individual days when each day was analyzed separately (Figure 2; P_{day 8} = 0.0070; P_{day 9} < 0.0001; P_{day 10} < 0.0001, P_day11 = 0.0128, P_{day 14} = 0.0404).

Figure 2.

Treatment group body weight, in grams, by day from baseline (days 0 to 4 and day 7) and after interventions (days 8 to 14) for all intervention groups combined (groups N-S). Boxplots represent the median and IQR, the marker represents the mean, and the range is denoted by the whiskers. Treatment effect (capromorelin compared with control) was significant (*) on each of the individual days when each day was analyzed separately (P_{day 8} = 0.0070; P_{day 9} < 0.0001; P_{day 10} < 0.0001; P_{day 11} = 0.0128; P_{day 14} = 0.0404) and when analyzed over the postintervention period days 8 to 14 (P < 0.0001). The interventions, regardless of treatment (combined capromorelin and control), resulted in a significant (†) decrease in group weights 3 d (P_{day 10} = 0.0313) and 4 d (P_{day 11} = 0.0157) after interventions but not over the period days 8 to 14 (P = 0.0586).

The effect of treatment was significant in 3 of the 4 interventions: groups N, I, and S. Mice in intervention groups N^t, I^t, and S^t gained more weight than untreated control mice over days 8 to 14 (P_N = 0.0348, P_I = 0.0032; P_S = 0.0020). When each day was analyzed separately, treatment effect was significant for group N days 9 to 10 (P_{day 9} = 0.0116; P_{day 10} = 0.0384), group I days 8 to 10 (P_{day 8} = 0.0423; P_{day 9} = 0.0023; P_{day 10} = 0.0032), and group S days 9 to 11 (P_{day 9} = 0.0088; P_{day 10} = 0.0050; P_{day 11} = 0.0013) (Figure 3). No significant differences in weight change were seen on any day between treated and nontreated mice that were given only buprenorphine XR (group B).

Figure 3.

Mean percent change in body weight from baseline weight for intervention groups N through S by treatment allocation (capromorelin or control) and for all treatment groups combined (capromorelin and control mice). Groups N^t, I^t, and S^t gained more weight than controls over days 8 to 14 (P_N = 0.0348; P_I = 0.0032; P_S = 0.0020). When each day was analyzed separately, treatment effect was significant (*) for group N days 9 to 10 (P_{day 9} = 0.0116; P_{day 10} = 0.0384), group I days 8 to 10 (P_{day 8} = 0.0423; P_{day 9} = 0.0023; P_{day 10} = 0.0032), and group S days 9 to 11 (P_{day 9} = 0.0088; P_{day 10} = 0.0050; P_{day 11} = 0.0013).

Intervention effect.

A comparison between interventions inclusive of both treatment groups (combined capromorelin and control) showed an effect that was suggestive but failed to meet the threshold for statistically significant change in body weight over the period from day 8 to day 14 (P = 0.0586). When each day was analyzed separately, the intervention effect resulted in a significant decrease in group weights 3 (P_{day 10} = 0.0313) and 4 d (P_{day 11} = 0.0157) after the interventions were conducted (Figure 2).

Pairwise comparisons of specific interventions inclusive of both treatment groups showed that the intervention effect was most pronounced between mice given buprenorphine XR alone (group B^{o + t}) and mice with no intervention (group N^{o + t}). Mice in group B^{o + t} gained significantly less weight than mice with no intervention (group N^{o + t}) over the period from day 8 to day 14 (P = 0.0414), as well as 3 (day 10) and 4 (day 11) days after buprenorphine XR was given (P_{day 10} = 0.0032; P_{day 11} = 0.0241). The addition of interventions in group I^{o + t} did not have a significant effect on mice weights compared with group B^{o + t} nor did addition of interventions in group S^{o + t} have a significant effect on mice weights compared with group I^{o + t}.

Like what was seen when both treatment groups were analyzed together, a comparison between interventions within only the treated mice showed that the intervention effect was most pronounced between mice given buprenorphine XR alone (group B^t) and mice with no intervention (group N^t). Mice in group B^t gained significantly less weight than mice with no intervention (group N^t) over the period from day 8 to day 14 (P = 0.0383), as well as 3 (day 10) and 4 (day 11) days after buprenorphine XR was given (P_{day 10} = 0.0048; P_{day 11} = 0.0086). Interestingly, treated mice that underwent anesthesia and were given meloxicam in addition to buprenorphine XR (group I^t) gained more weight 3 d after the interventions relative to mice given buprenorphine XR alone (group B^t) (P_{day 10} = 0.0298). There was no significant intervention effect noted for group S^t mice. In contrast to the changes seen in treated animals, pairwise comparisons of specific interventions in control animals did not demonstrate any significant changes in body weight between the interventions over the period from day 8 to day 14 or on any of the individual days.

Time to return to or exceed baseline weight.

Treatment effect.

A comparison between treatments inclusive of all intervention groups N through S showed a significantly shorter time required to RoE baseline weight in mice given capromorelin as compared with unmedicated controls (P < 0.0001). The median time to RoE baseline weight for the treated group was one day compared with the untreated (control) group’s median time of 4 d (Figure 4). Further, 31.3% (10 of 32) of untreated (control) animals never RoE their baseline weight by 7 d postintervention, compared with 3.4% (1 of 29) of capromorelin-treated animals that never RoE baseline weight.

Figure 4.

Kaplan-Meier graph demonstrating the time, in days, to RoE baseline weight for control (solid lines) and capromorelin-treated mice (dashed lines). Inclusive of all interventions, the median time to RoE baseline weight for the treated group was 1 d compared with the untreated (control) group’s median time of 4 d. The median time to RoE baseline weight for group N^t was 1 d compared with 1.5 d for group N^o (P = 0.0253) with 100% N^t compared with 87.5% N^o RoE baseline weight by 7 d postintervention. The median time to RoE baseline weight for group I^t was 1 d compared with 7 d for I^o(P = 0.0015) with 100% I^t compared with 75% I^o animals RoE baseline weight by 7 d postintervention. A comparison between treated and control animals within groups B and S trended toward significance (P_B = 0.0698; P_D = 0.1003).

A more granular analysis of treatment groups by intervention group revealed significant differences in time required to RoE baseline between treated and control groups for intervention groups N (P = 0.0253) and I (P = 0.0015). Within group N, 100% of treated animals RoE their baseline weight within 1 d postintervention, while control animals RoE their baseline weight at a median of 1.5 d postintervention, with 12.5% (1 of 8) mice never RoE baseline weight by 7 d postintervention (Figure 4). Similarly, within group I, 100% of treated animals RoE their baseline weight within 2 d postintervention, with a median of 1 d, compared with control animals that had a median time to RoE baseline of 7 d, with 25.0% (2 of 8) of I^o mice never RoE baseline weight. A comparison between treated and control animals within groups B and S trended toward significance, but no significant difference in time to RoE baseline weights was found (P_B = 0.0698; P_D = 0.1003).

Intervention effect.

A comparison between interventions inclusive of both treatment groups showed no significant difference (P = 0.1345) in time to RoE baseline weight. Pairwise comparisons of specific interventions within the capromorelin-treated mice showed significant differences in time required to RoE baseline weight between mice given buprenorphine XR alone (group B^t) and mice with no intervention (group N^t) (P = 0.0455). Although the median time to RoE baseline weight was the same for both groups (1 d), 100% of group N^t mice RoE baseline weight within 1 d postintervention, whereas only 50% of group B^t mice RoE baseline weight within 1 d postintervention (Figure 4). There was no significant difference in time to RoE baseline weight between group B^t and I^t, nor between groups I^t and S^t. Pairwise comparisons of specific interventions within the control-treated mice showed no significant differences in time required to RoE baseline weight between group B^o and group N^o (P = 0.2127), between group N^o and group I^o (P = 0.9939), or between group I^o and group S^o (P = 0.5335).

Discussion

This study investigated the effect of enterally administered capromorelin 10 mg/kg for 3 d in mice and demonstrated a positive effect on body weight. It was demonstrated that mice who received capromorelin gained more weight than those in the control group in 3 of 4 intervention groups: no intervention (group N), anesthesia plus analgesia (group I), and surgery under anesthesia plus analgesia (group S). However, treating mice with capromorelin did not significantly mitigate body weight loss in mice administered one dose of buprenorphine XR alone (group B). The increase in body weight seen in the capromorelin-treated group resulted in a return of more mice to their baseline weight (96.6%) more rapidly than mice that did not receive the capromorelin (68.8%). In mice with no intervention (group N) and anesthetized mice given analgesia (group I), more capromorelin-treated mice returned to their baseline weight by 7 d (100% N^t compared with 87.5% N^o; 100% I^t compared with 75% I^o). Similarly, in mice administered buprenorphine XR alone (group B) and mice that underwent surgery with anesthesia and analgesia (group S), more capromorelin-treated mice returned to their baseline weight by 7 d (100% B^tcompared with 62.5% B^o; 87.5% S^tcompared with 50% S^o). Although the time to RoE baseline weights was not statistically significant between treated and control group B or group S mice, capromorelin may be worth attempting clinically to mitigate opioid-related or surgical-related weight loss for individual mice.

To the authors’ knowledge, this is the first time that administration of capromorelin enterally has been demonstrated to result in increased weight in mice. The results from this study are consistent with research on enterally or parenterally administered GhrRAs in mice and rats demonstrating decreased gastrointestinal transit time, increased food consumption, and increased fecal output.^10,28,29,36 Specifically, others have demonstrated a GhrRA’s ability to mitigate the effect of opioid administration on gastrointestinal motility in mice.¹⁰ This present study showed capromorelin was not able to counter buprenorphine XR-associated weight loss, which is suggestive that either the weight loss associated with buprenorphine XR may not be related to gastrointestinal motility or the GhrRA capromorelin was not able to confer a change in motility to an appreciable change in body weight. This present study did not specifically analyze gastrointestinal motility; therefore, the correlation of mechanisms is theoretical. The positive effect of capromorelin and other GhrRAs on body weight has also been demonstrated in dogs, cats, rabbits, chickens, nonhuman primates, and humans.^{4,9,11,38,41,42,48,49,51,52} This present study establishes a clinical application for capromorelin administration to help mitigate weight loss in mice that endure stressors in a research environment. Supporting food consumption and body weight is important when body weight could confound research outcomes and to ensure animal welfare.

Capromorelin marketed under the brand name Elura is labeled for oral administration to cats that have been fasted with food offered 30 min after the Elura dose.¹³ One author’s (EMP) conversations with Elanco elucidated that Elura is often given to cats in a high-fat food to encourage consumption of the medication and support body condition. Elanco studied the pharmacokinetics of Elura by measuring both serum capromorelin levels and insulin-like growth factor 1 (IGF-1) as a proxy for the bioactivity of capromorelin.^6,31,41,47 The presence of food significantly reduced both the peak concentration and overall systemic exposure (area under the curve) of serum capromorelin levels in cats, despite having no impact on their serum IGF-1 levels. Therefore, Elanco recommends fasting to ensure maximum drug exposure but does not expect a significant decline in effectiveness in fed animals.⁴⁷ Comparable to the bioactivity of Elura, the oral bioavailability of capromorelin has been reported to be up to 65% in fasted rats.⁶ In this present study, the Elura solution was administered to mice that were allowed free access to rodent chow, and the dose was administered enterally in condensed milk. Administration of capromorelin in fed mice, therefore, may have dampened the response that may have been seen in fasted mice. However, fasting the mice would not accurately represent how capromorelin may be used in the clinical care of research mice; thus, fasting the mice in this study would have obscured the interpretation of how capromorelin affects body weight in research mice.

This study aimed to unravel the individual and combined impacts of analgesia, anesthesia, surgery, and capromorelin on body weight in mice. It was demonstrated that these specific research-related interventions are correlated with a decrease in body weight 3 and 4 d after the interventions were administered. The most pronounced intervention effect was in mice administered buprenorphine XR; mice administered the opioid lost more weight than mice that had not received the opioid. This suggests that buprenorphine XR administration leads to weight loss even in the absence of pain, which is consistent with the label indications for Ethiqa XR¹⁵ and other reports of side effects of opioid administration on rodents.^1,2,33,39 Contrary to what was expected, the addition of meloxicam analgesia, isoflurane anesthesia, or a surgical sham procedure did not magnify the weight loss seen from buprenorphine XR dosing. It may be possible to elucidate a difference through repeated investigation with a larger sample size. Furthermore, it could be that the analgesic regime was sufficient to counter weight loss that may have otherwise followed from postsurgical pain and inflammation.

Each factor investigated in the study provides the opportunity for additional exploration, such as modification of medications selected or medication dosages. For example, the mice in this study were anesthetized with isoflurane, which has been demonstrated to not affect body weight in mice.²² A diverse array of sedatives and anesthetics are employed in mouse research protocols, tailored to the specific needs of each experiment. Other agents, such as dissociatives (e.g., ketamine), α₂-agonists (e.g., xylazine), or sedatives (e.g., acepromazine), may affect gastrointestinal motility, food consumption, or body weight. Capromorelin was dosed at 10 mg/kg, which is the same dose administered to rats via orogastric gavage³⁶ but is higher than doses used in other species.^{4,9,11,38,49,51,52} Additional studies should be conducted in mice to evaluate other doses, dosage frequencies, or variable dosing as might occur if the mice were dosed using a capromorelin-infused gel or food. Capromorelin was administered to mice during the light period of the light:dark cycle. However, mice are nocturnal and known to consume most of their food in the dark period.²⁴ Capromorelin has a relatively short duration of action with a 1-h T_max after oral dosing and plasma elimination half-life of 1.3 h in rats.⁶ Thus, administration of capromorelin later in the day or during the dark phase may increase food consumption in sync with the mouse’s natural feeding schedule, resulting in an increase in body weight beyond what was seen in this study. To avoid disruption of the mice, capromorelin could be administered as a food supplement in the cage.

Reported adverse events from capromorelin administration, principally diarrhea, hypersalivation, and lethargy,^4,14,47 were not observed or investigated (i.e., serum biochemistry changes or specific histologic changes¹⁴) in this study. The 3 mice that died or were euthanized during the intervention phase were excluded from statistical analysis so as to not confound a change in weight due to clinical decline with a change in weight due to intervention or treatment. These 3 mice were all from the capromorelin treatment group, so it cannot decisively be declared that clinical decline was not related to treatment. To determine the cause of death, all mice that died or were euthanized during any point in this study were sent for anatomic pathology analysis, with no definitive or suggestive cause of death identified. The assigned pathologists were not blinded to any interventions or treatments the animals received before death. Specifically, the pathologist was asked to critically examine the gastrointestinal tract of the mice that died during the intervention phase due to the mechanism of action of capromorelin on this system, but the pathologists did not find any abnormalities. It was suspected that the 2 mice that died in intervention group I may have succumbed due to concurrent meloxicam and Ethiqa XR administration. Meloxicam was administered concurrently with buprenorphine XR to investigate the effect of capromorelin on body weight when mice were administered multimodal analgesia.¹⁹ This present study selected a relatively high meloxicam dosage compared with what may be administered in other species based on evidence suggesting higher doses of NSAIDs are more effective at alleviating pain in mice.^27,32,50 In the time since Ethiqa XR was released, concurrent administration of NSAIDs and Ethiqa XR to mice has been reported to Fidelis Animal Health, as causing several adverse events, including death. The warning against concurrent NSAID and Ethiqa XR administration was first published on the Ethiqa XR Frequently Asked Questions website¹⁶ in August 2023, after this study had been completed.

Mice in research may undergo a variety of surgical procedures depending on the research means and objectives. This study used a laparotomy sham procedure to replicate the nonspecific physiologic responses triggered by tissue disruption in surgery, while deliberately avoiding internal organs. However, manipulation of visceral organs may cause physical or physiologic alterations that could affect surgical recovery, gastrointestinal motility, and/or appetite. In models in which surgery is conducted on mice as preprotocol support, it is unlikely that administration of capromorelin could itself become a confounding variable because of the short duration of action of capromorelin and because a minimum one-week recovery period would often be required for tissue healing before any further intervention. However, investigators should consider each model individually to determine if capromorelin may affect parameters within their specific studies.³⁷

Finally, the effect of capromorelin or interventions on body weight may be different with different ages or strains of mice. The BALB/c mice in this study were procured at the oldest age available from the vendor, and they were allowed to acclimate for several weeks at the institution to reach a more adult body weight to avoid any influence of natural growth on the results. Young mice or very old mice may have a different body weight response to capromorelin due to differences in growth rate and hormonal influences throughout a mouse’s life.

In summary, mice administered 10 mg/kg capromorelin enterally once daily for 3 d increased in weight more than mice administered a control solution, even when they received interventions like analgesics, anesthesia, and/or surgery. Route of administration, oral consumption or orogastric gavage, did not change the efficacy of capromorelin. The interventions also resulted in a change in body weights, which was mitigated by capromorelin administration in all interventions except for the buprenorphine XR-only group. Mice that received capromorelin were more likely to have returned to a baseline weight by 7 d postintervention, and the median time to RoE the baseline weight was shorter for mice that received capromorelin. This is the first study to evaluate the effect of enterally administered capromorelin on body weight in mice and its ability to counteract weight loss arising from research interventions. Capromorelin may be a reasonable auxiliary treatment to mitigate research-induced weight loss in mice.

Conflict of Interest

The authors have no conflicts of interest to declare.

Funding

This work was internally funded.