Effect of Home Cage Bedding in the Induction Chamber on Serum Cortisol and Corticosterone Levels in Response to Isoflurane-induced Anesthesia in C57BL/6J Mice

Abstract

Mice are routinely anesthetized with isoflurane in an induction chamber. The AVMA Guidelines for the Euthanasia of Animals states that distress should be minimized during euthanasia but does not address this point in regard to induction of anesthesia. Here we evaluated the potential for familiar surroundings to reduce the adrenocortical response of mice during anesthesia induction with isoflurane. However, adding bedding from the animals’ home cage to the induction chamber failed to significantly reduce serum cortisol or corticosterone levels in male and female C57BL/6J mice. These results indicate that familiar surroundings do not appear sufficient to reduce the adrenocortical response of mice during anesthesia induction with isoflurane.

The AVMA Guidelines for the Euthanasia of Animals (2013 edition) states, “for virtually all animals, being placed in a novel environment is stressful; therefore, a euthanasia approach that can be applied in familiar surroundings may help reduce stress.” Both the Guide and PHS Policy require euthanasia be performed in accordance with the 2013 AVMA euthanasia guidelines.^1,6,9 Although the reduction of stress is emphasized when euthanizing animals, neither the Guide nor PHS Policy address reducing stress when anesthetizing animals, despite the fact that the only difference between the 2 situations is whether the anesthetic event is terminal.^6,9

To allow for the completion of various procedures, laboratory rodents are routinely anesthetized, often by using inhalant anesthesia introduced into an induction chamber. Handling an animal to transfer it into an unfamiliar induction chamber or the odor from the anesthetic induce changes in the concentrations of various hormones, including corticosterone and other glucocorticoids.¹¹ As a result of stress, which activates the hypothalamic–pituitary–adrenal axis, the release of the glucocorticoids corticosterone and cortisol from the adrenal cortex is increased.^3,5,14 Glucocorticoids aid in maintaining homeostasis after stress by mobilizing energy stores. Because these glucocorticoids also affect behavior and regulate daily activity patterns, any variation in their levels may be a potential confounder for research.¹⁴ Historically, corticosterone has been shown to be the main glucocorticoid involved in the regulation of the rodent stress response.^2,5,8 However, recent studies have revealed increased cortisol levels in plasma and adrenal glands in response to stress, and many studies have used cortisol as the index for stress activation in mice.^2,5,8,15 One group has even shown that cortisol is a quicker responder than corticosterone during severe acute stress.⁵ Because the onset of anesthesia in an induction chamber is an acute event, we analyzed both serum cortisol and corticosterone levels in C57BL/6J mice in the current study.

Our goal was to develop a refinement technique to decrease the level of stress that mice may experience when anesthetized in an induction chamber. The study evaluated a simple practice that could easily be replicated and repeated in subsequent studies. We hypothesized that mice that were placed in an anesthesia induction chamber that contained bedding from their home cage would be less stressed, as indicated by lower levels of serum cortisol or corticosterone (or both), than control mice anesthetized in a bedding-free chamber.

Materials and Methods

Animals.

Adult female (n = 20) and male (n = 20) C57BL/6J mice (Mus musculus) were obtained from The Jackson Laboratory (Bar Harbor, ME). Mice were housed in groups of 5 in static polycarbonate shoebox-type cages with filter tops (Ancare, Bellmore, NY) on rodent hardwood bedding (item number 7090M, Laboratory-grade Maple SaniChips, Envigo, Indianapolis, IN), and each cage had shredable nesting material (Cotton squares, Ancare) and huts (Mouse Igloo, BioServ, Flemington, NJ) for environmental enrichment. Mice had unrestricted access to pelleted rodent food (item number 2018, Tekland Global 18% Protein Rodent Diet, Envigo) and filtered (15 to 50 µm) tap water. The room was maintained on a 12:12-h light:dark cycle (lights on, 0600; lights off, 1800), temperature was 68 to 79 °F (20.0 to 26.1 °C), and relative humidity was between 30% and 70%. All procedures were approved by the IACUC of the Uniformed Services University of the Health Sciences, an AAALAC-accredited facility.

Sentinel mice were used to monitor the health status of the experimental animals. Every week, female CD1 mice were directly exposed to dirty bedding from all colony animals. Every quarter, the sentinel mice were euthanized by using CO₂, and blood samples, fur swabs, and fecal pellets were collected. Serology and PCR analyses were performed (Mouse Assessment Profile, Fur Mite Panel, and Pinworm PCR, IDEXX BioResearch, Columbia, MO). Tissues were submitted quarterly for histopathology (Joint Pathology Center, Silver Spring, MD). In addition, all sentinel mice were examined inhouse for fur mites and pinworms by using hair pluck and tape tests. Selected mice (25% of the colony; chosen at random) also were tested for fur mites and pinworms by using fur swabs and fecal pellets (Fur Mite Panel and Pinworm PCR, IDEXX BioResearch). At the time of this study, all mice housed in this room were negative for Mycoplasma pulmonis, ectromelia virus, mouse rotavirus, K virus, lymphocytic choriomeningitis virus, mouse adenovirus types 1 and 2, mouse hepatitis virus, mouse norovirus, mouse parvovirus, minute virus of mice, polyomavirus, pneumonia virus of mice, reovirus 3, Sendai virus, Theiler murine encephalomyelitis virus, fur mites, and pinworms. Mice were not tested for Helicobacter spp. in this room.

Induction of isoflurane anesthesia.

Half of each population of mice by sex (n = 10) were induced under isoflurane (Baxter Healthcare, Deerfield, IL) anesthesia in an empty 1-L induction chamber (interior dimensions, 3.75 in. × 4.50 in. × 3.75 in.; VetEquip, Pleasanton, CA). The remaining mice (n = 10 of each sex) were induced under isoflurane anesthesia in the same size induction chamber into which approximately 60 mL of bedding from their home cage was placed prior to transferring the mouse. On a single day, 2 cages of mice of the same sex were anesthetized and blood collected. Five mice from one cage were placed in the experimental group, which received bedding in the induction chamber, whereas the other 5 mice had no bedding. A representative sample of 60 mL of bedding was collected from the home cage by using a 60-mL syringe. The bedding sample was placed into the induction chamber, mimicking the approximate depth of bedding in the home cage, just prior to transferring the first mouse (Figure 1). The induction chamber was filled with 4% isoflurane vaporized in oxygen at a flow rate of 0.5 L/min until the righting reflex was absent. Two different chambers were used on a single day, to prevent the transfer of pheromones. The chamber containing bedding was emptied, and both chambers were disinfected (Micro-Kill+ wipes, Medline Industries, Mundelein, IL) daily.

Figure 1.

Induction chamber with 60 mls of bedding.

Blood collection and analysis.

As soon as the righting reflex was lost, while the mouse was under anesthesia, approximately 300 µL of blood was collected from the facial vein by using a 5-mm animal lancet (Medipoint, Mineola, NY) to make a stab incision into the cheek caudodorsal to the cheek skin gland. Drops of blood were collected in a collection tube, and bleeding was controlled by using sterile gauze to apply light compression to the cheek. If the required blood volume was not collected before the animal regained consciousness, the mouse was returned to the induction chamber, anesthetized until the righting reflex was lost, and blood collection was performed on the contralateral side. Studies in mice have shown that the glucocorticoids cortisol and corticosterone have a diurnal release pattern, with lower concentrations in the morning.^{4,5,7,8,12,15} To minimize the effect time of day has on glucocorticoid levels, mice were anesthetized between 0930 and 1015 on subsequent mornings.

One mouse per cage was anesthetized at a time and marked with a permanent marker on its tail prior to being returned to its home cage, to indicate that it had been used. Blood was centrifuged within 1 h of collection. Serum was separated and frozen prior to shipment. Serum was shipped on dry ice to Charles River Research Animal Diagnostic Services (Wilmington, MA) and submitted for cortisol and corticosterone testing.

Statistical analyses.

Mean serum cortisol and corticosterone levels were compared separately by using a t test for independent samples. Analysis of covariance was used to adjust for number of stab incisions and number of trips to the induction chamber, and factorial ANOVA was used to test for potential interactions. A sample size of 10 mice per group had 80% power with a 5% 2-sided significance level to detect differences of 1.3 SD, corresponding to a difference in cortisol concentration of roughly 0.2 ng/mL for both male and female mice and in corticosterone concentrations of 19 and 46 ng/mL in female and male mice, respectively, according to estimated standard deviations observed after adjusting for bedding, number of stab incisions, and trips to the induction chamber. Pearson correlation was used to measure the linear correlation between cortisol and corticosterone. Statistical analyses were conducted by using SPSS (version 22, IBM, Armonk, NY).

Results

The presence of bedding from the home cage in the induction chamber while male or female mice were anesthetized failed to have significant effects on serum cortisol or corticosterone levels (Table 1). Subjectively, mice anesthetized with bedding appeared less agitated during induction, as indicated by less digging in the corners of the induction chamber, than observed with control mice. Some mice required multiple stab incisions with the lancet to allow the collection of the entire 300 µL of blood to be collected. The overall range in the number of stab incisions was 1 to 8 per mouse, with a mean of 2 incisions per mouse (Figure 2). In addition, some mice required multiple trips to the induction chamber to achieve adequate anesthesia for blood collection, with a maximum of 5 trips and an overall average of 1.7 trips (Figure 3). Neither serum cortisol nor corticosterone levels differed significantly between mice requiring more than one stab incision (Table 2) or anesthetic event (Table 3) and those with single stab incisions or events. After combining data from male and female mice and adjusting for sex, number of stab incisions, and number of trips to the chamber, the estimated cortisol concentration (mean ± SEM) was 0.65 ± 0.04 ng/mL in the bedding group compared with 0.61 ± 0.04 ng/mL in the no-bedding group (P = 0.457) and the estimated corticosterone level was 93.45 ± 8.03 ng/mL in the bedding group compared with 100.89 ± 8.03 ng/mL in the no-bedding group (P = 0.521). Additional ANOVA models were used to test whether the effect of bedding differed in mice with a single stab incision and trip to the chamber compared with other mice, but no significant interactions were found.

Table 1.

Serum cortisol and corticosterone levels of male and female mice with and without bedding in the induction chamber

		Cortisol (ng/mL)		Corticosterone (ng/mL)
	Bedding	Mean ± 1 SD	Range	Mean ± 1 SD	Range
Male	Yes	0.65 ± 0.18a	0.47–1.04	101 ± 42b	77–211
Male	No	0.59 ± 0.11	0.40–0.79	124 ± 51	52–199
Female	Yes	0.65 ± 0.22c	0.32–1.10	84 ± 22d	36–111
Female	No	0.63 ± 0.17	0.38–0.90	80 ± 20	56–123

^aP = 0.343 compared with value for no-bedding male mice.

^bP = 0.285 compared with value for no-bedding male mice.

^cP = 0.827 compared with value for no-bedding female mice.

^dP = 0.663 compared with value for no-bedding female mice.

Figure 2.

Number of stab incisions with the lancet for blood collection.

Figure 3.

Number of times the mouse was anesthetized in the induction chamber for blood collection.

Table 2.

Serum cortisol and corticosterone levels of male and female mice according to the number of stab incisions made

		Cortisol (ng/mL)		Corticosterone (ng/mL)
	No. of stab incisions	Mean ± 1 SD	Range	Mean ± 1 SD	Range
Male	1 (n = 11)	0.61 ± 0.10	0.48–0.84	124 ± 47	60–211
Male	>1 (n = 9)	0.64 ± 0.19a	0.40–1.04	98 ± 46b	52–211
Female	1 (n = 5)	0.63 ± 0.20	0.32–0.90	73 ± 11	56–85
Female	>1 (n = 15)	0.64 ± 0.20c	0.38–1.09	84 ± 22d	36–123

^aP = 0.722 between means for 1 and >1 stab incisions (t test for equality of means) in male mice.

^bP = 0.223 between means for 1 and >1 stab incisions (t test for equality of means) in male mice.

^cP = 0.975 between means for 1 and >1 stab incisions (t test for equality of means) in female mice.

^dP = 0.288 between means for 1 and >1 stab incisions (t test for equality of means) in female mice.

Table 3.

Serum cortisol and corticosterone levels of male and female mice according to the number of sessions in the induction chamber

		Cortisol (ng/mL)		Corticosterone (ng/mL)
	No. of sessions in induction chamber	Mean ± 1 SD	Range	Mean ± 1 SD	Range
Male	1 (n = 10)	0.62 ± 0.10	0.32–0.85	124 ± 47	60–211
Male	>1 (n = 10)	0.63 ± 0.19a	0.44–1.10	98 ± 46b	52–211
Female	1 (n = 11)	0.60 ± 0.18	0.48–0.84	81 ± 15	56–111
Female	>1 (n = 9)	0.67 ± 0.20c	0.40–1.04	83 ± 25d	36–123

^aP = 0.860 between mean for 1 and >1 session in the induction chamber (t test for equality of means) in male mice.

^bP = 0.229 between mean for 1 and >1 session in the induction chamber (t test for equality of means) in male mice.

^cP = 0.444 between mean for 1 and >1 session in the induction chamber (t test for equality of means) in female mice.

^dP = 0.848 between mean for 1 and >1 session in the induction chamber (t test for equality of means) in female mice.

Discussion

We measured both corticosterone and cortisol in this study, because a review of the literature revealed cortisol might be a better indicator of acute stress in mice.^2,5,8,15 In addition, no studies to date have evaluated potential methods to reduce stress in small rodents during induction with inhalant anesthesia. Therefore, this study took a novel approach to evaluating stress in C57BL/6J mice, with the objective of reducing stress during the time spent in the induction chamber.

The aim of this study was to test the hypothesis that having bedding from the home cage in the induction chamber would reduce adrenocortical activation in mice during induction of anesthesia with isoflurane. However, serum cortisol and corticosterone levels were not significantly different between mice having bedding from their home cage in the induction chamber compared with mice in an empty chamber. Perhaps no difference emerged due to the small sample size or variability in study conditions (for example, number of stab incisions and sessions in the induction chamber). Therefore, lack of significant differences between the groups indicates that bedding in the induction chamber does not appear to influence cortisol or corticosterone levels in mice undergoing isoflurane anesthesia.

As stated in the Guide “loss of consciousness occurs at a light plane of anesthesia, before antinociception, and is sufficient for purposes of restraint, but painful stimuli can induce a return to consciousness.” As we observed in this study, some mice regained consciousness prior to completion of blood collection and consequently were returned to the induction chamber to be reanesthetized. Therefore, a potential limitation of this study is that blood collection was not always completed during a single anesthetic event. However, glucocorticoid levels did not differ between mice requiring repeated placement in the induction chamber and those placed in the chamber only once. Future studies may include placement of the entire home cage or multiple mice from the home cage into the induction chamber. These changes, however, might limit the procedures to be performed to those that are relatively quick, because mice would be anesthetized simultaneously. Alternatively the addition of a hut or cotton square from the home cage into the induction chamber might influence the adrenocortical response. Developing an ethogram and videotaping the induction, thus quantifying behaviors exhibited, such as digging, may reveal differing results. In addition, because heart rate, an indicator of stress, might vary among these mice, the use of an implanted telemetry device in future studies would allow for collection of this information. Furthermore, several studies^10,13 have shown interstrain differences in response to stress, and reproducing this study in other strains of mice may reveal differing results.