Effects of Scruff Restraint and Handling—Tail and Tunnel—on Plasma Corticosterone in Female BALB/c Mice (Mus musculus)

Abstract

Tunnel handling is a widely recommended, less-aversive method for improving laboratory mouse welfare by reducing handling-induced anxiety. While its effects on behavioral tests and physiologic outcomes have been reported, little is known about its impact on stress responses to brief restraint procedures commonly used in daily health checks and experimental work. This study aimed to investigate whether tunnel or tail handling influences the physiologic stress response of mice following a brief scruff restraint. Female BALB/c mice were assigned to one of 4 groups combining 2 handling methods (tail or tunnel) with or without a 10-second scruff restraint. Both single and repeated restraint procedures were evaluated separately. Voluntary interaction with the handling device was used to assess the suitability of the handling method. Plasma corticosterone (pCORT) concentrations were measured 20 minutes after restraint to evaluate the physiologic stress response. Tunnel-handled mice showed longer voluntary interaction times than tail-handled mice, indicating reduced handling aversion. Scruff restraint increased pCORT concentrations in both tail- and tunnel-handled mice. However, the handling method did not significantly affect pCORT concentrations following either single or repeated restraint. These findings suggest that the handling method has minimal impact on physiologic stress following brief restraint, supporting the use of less-aversive handling techniques without concern for confounding stress effects.

Introduction

When using laboratory mice for research, daily husbandry and care procedures are designed to minimize distress. However, removing mice from their cages is unavoidable during cage changes and experimental procedures. A long-standing method for this purpose is tail handling, in which the mouse is picked up by its tail. Recent studies^1,2 have suggested that tail handling reduces interactions between the mouse and the handler’s hand. In contrast, less-aversive methods, such as tunnel handling and cupping (holding the mouse in the palm), are considered refined handling techniques. These methods are believed to reduce handling-induced anxiety and improve the welfare of mice, and their use is therefore recommended.^3,4

Beyond interactions with handling devices such as tunnels or hands, behavioral studies^1,5 have shown that mice subjected to less-aversive handling exhibit reduced anxiety-like behaviors. Physiologic differences have also been observed, including lower blood glucose levels due to improved glucose tolerance⁶ and increased responsiveness to sucrose rewards.⁷ The effects of handling methods on stress responses are also of interest. Some studies⁸ have examined their impact on plasma corticosterone (pCORT) concentrations, while others^6,9 have evaluated pCORT changes following procedures such as exposure to an elevated plus maze or tube restraint. Despite these findings, the impact of handling methods on stress responses remains unclear, contributing to hesitancy in adopting less-aversive handling techniques.¹⁰

Laboratory procedures such as restraint, blood collection, and routine cage changes can cause stress in animals.¹¹ In mice, scruff restraint is frequently used for health checks and experimental procedures, but it can alter physiologic parameters such as heart rate and core body temperature.¹² pCORT concentrations also increase following scruff restraint.¹³ Scruff-restrained mice show reduced interaction with the handling device compared with unrestrained mice, regardless of whether tail or tunnel handling is used.¹⁴ However, even after scruff restraint, tunnel-handled mice still exhibit greater levels of interaction than tail-handled mice.^1,14,15

The effects of experimental procedures on mice handled using less-aversive methods have been examined. These methods limit the opportunities of the mice for human contact, raising concerns that they may experience significant stress during procedures involving direct human interaction.¹⁰ This concern has been reported as a reason for the continued use of tail handling.¹⁰ Understanding how handling methods affect stress responses during routine procedures is crucial not only from an animal welfare perspective but also for ensuring the reliability of experimental outcomes, as handling methods may influence experimental results.

We aimed to examine how different handling methods affect the stress response associated with scruff restraint in mice. Specifically, the effects of tail compared with tunnel handling, with and without scruff restraint, were investigated. Tail handling remains the commonly used method,^10,16,17 while tunnel handling is well-known and adopted as a less-aversive method.^10,17 As the stress response to restraint is known to depend on its frequency and duration,¹⁸ we evaluated both single and repeated scruff restraints. The stress response was assessed using pCORT concentrations. In addition, the suitability of tail and tunnel handling, as well as changes in anxiety toward the handling device following scruff restraint, were evaluated using voluntary interaction tests.

Ethical Review

All procedures in this study were conducted in accordance with the Japanese Law for the Humane Treatment and Management of Animals (Law Number 105, enacted on October 1, 1973). All animal experiments were reviewed by the Institutional Animal Care and Use Committee of Yakult Central Institute and approved by the Director of the Yakult Central Institute. All animals were housed and used in compliance with a program accredited by AAALAC International.

Materials and Methods

Study design.

This study consisted of 3 experiments, each conducted over 12 days. The timelines for each experiment are shown in the figures along with the corresponding results (Figures 1A, 2A, and 3A). A total of 128 female BALB/c mice were used across all experiments. This sex and strain of mouse were chosen because they were expected to exhibit more pronounced stress responses following scruff restraint, based on known differences in stress sensitivity between sexes^19–21 and strains.^22–26 In each experiment, 4 groups were formed by combining 2 handling methods (tail and tunnel) with or without scruff restraint, as follows: Tail-1, tail-handled and unrestrained; Tail-2, tail-handled and scruff-restrained; Tunnel-1, tunnel-handled and unrestrained; and Tunnel-2, tunnel-handled and scruff-restrained.

Figure 1.

Effect of Handling Method and Single-Scruff Restraint on Voluntary Interaction with the Handling Device. (A) Timeline of procedures. All mice were handled with the assigned methods on days 1–5 and 7–10. On day 11, handling or scruff restraint was carried out, and on day 12, handling was performed again. The VI test was conducted immediately after handling or scruff restraint on days 1, 5, 11, and 12. (B) Time interacting with handling devices is shown as the median for each group, with individual values indicated by circles (days 1 and 5, n = 8 cages per handling method; days 11 and 12, n = 4 cages per group). Significance was tested with the Mann–Whitney U test on days 1 and 5, and the P values and effect sizes (r²) are shown. On days 11 and 12, a 2-way ANOVA was used to test the effects of handling method and restraint treatment, and the P values and effect sizes (${\epsilon }_p^2$) are shown. P values <0.05 are shown in red. Hnd, handling method; Rst, scruff restraint treatment; Tail-1, tail-handled and unrestrained group; Tail-2, tail-handled and scruff restrained group; Tunnel-1, tunnel-handled and unrestrained group; Tunnel-2, tunnel-handled and scruff restrained group; VI test, voluntary interaction test.

Figure 2.

Effect of Handling Method and Single-Scruff Restraint on pCORT. (A) Timeline of procedures. All mice were handled from day 1 to day 11. Scruff restraint was performed once on day 12 in the Tail-2 and Tunnel-2 groups. Blood samples for measurement of pCORT were collected from all mice between 8:00 and 11:00 am on day 12; in the Tail-2 and Tunnel-2 groups, sampling was performed 20 minutes after scruff restraint. (B) pCORT in each group is shown as the median, with individual values indicated by circles (n = 8 mice per group). A 2-way ANOVA was used to test the effects of handling method and restraint treatment. The P values and effect sizes (${\epsilon }_p^2$) are shown. P values <0.05 are indicated in red. CORT, blood sampling for measurement of pCORT; Hnd, handling method; pCORT, plasma corticosterone; Rst, scruff restraint treatment; Tail-1, tail-handled and unrestrained group; Tail-2, tail-handled and scruff restrained group; Tunnel-1, tunnel-handled and unrestrained group; Tunnel-2, tunnel-handled and scruff restrained group; VI test, voluntary interaction test.

Figure 3.

Effect of Handling Method and Repeated Scruff Restraint on Voluntary Interaction with the Handling Device and pCORT. (A) Timeline of procedures. Handling was performed on all mice from day 1 to day 5. Scruff restraint was performed once daily from day 8 to day 12 in the Tail-2 and Tunnel-2 groups, and handling was continued for the other mice. VI tests were performed on days 1, 5, and 11. Blood samples for measuring pCORT were collected from all mice between 8:00 and 11:00 am on day 12; in the Tail-2 and Tunnel-2 groups, sampling was performed 20 minutes after scruff restraint. (B) Time interacting with handling devices on day 11 is shown as the median for each group, with individual values indicated by circles (n = 8 cages per group). Significance was tested with a 2-way ANOVA with handling method and restraint treatment as factors, and the Steel–Dwass test was used for the post hoc test. (C) pCORT on day 12 is shown as the median for each group, with individual values indicated by circles (n = 16 mice per group). A 2-way ANOVA was used to test the effects of handling method and restraint treatment. P values and effect sizes (${\epsilon }_p^2$ for 2-way ANOVA and r² for the Steel–Dwass test) are shown. P values <0.05 are shown in red. CORT, blood sampling for evaluation of pCORT; Hnd, handling method; pCORT, plasma corticosterone; Rst, scruff restraint treatment; Tail-1, tail-handled and unrestrained group; Tail-2, tail-handled and scruff restrained group; Tunnel-1, tunnel-handled and unrestrained group; Tunnel-2, tunnel-handled and scruff restrained group; VI test, voluntary interaction test.

Upon arrival, the mice were transferred from the transport crates into cages in pairs by an animal caretaker who was not involved in any other parts of this study. Each cage was then randomly assigned to a group by lottery. All experimental procedures and data analyses were conducted in an unblinded manner. The cages were changed once per week. The final cage change during the acclimation period was conducted 3 days before the start of handling (day 1). Cage changes during the experimental period were performed after the voluntary interaction test on day 5 to avoid influencing the test results. During acclimation, the mice were transferred between cages using the tail-handling method. From the start of the experiment, each mouse was handled according to its assigned method. All routine care and experimental procedures were conducted by the same handler.

In Experiment 1, the effects of a single scruff restraint and handling methods on voluntary interactions were evaluated. The 30-second handling procedure described below was performed on days 1–5, 8–10, and 12. On day 11, mice in the Tail-2 and Tunnel-2 groups were subjected to a single scruff restraint, while those in the Tail-1 and Tunnel-1 groups were handled as on the previous day. Voluntary interactions were assessed on days 1 and 5 (the first and last days of the initial consecutive handling period), on day 11 (immediately after the scruff restraint), and on day 12 (the day after the scruff restraint).

In Experiment 2, the effects of a single scruff restraint and handling methods on pCORT were investigated. Mice were handled on days 1–5 and 8–11, and voluntary interactions were assessed on days 1, 5, and 11. On day 12, mice were subjected to a single scruff restraint, and blood was collected 20 minutes later for pCORT measurements.

In Experiment 3, the effects of repeated scruff restraints and handling methods on pCORT were evaluated. Mice were handled on days 1–5. From days 8–12, the Tail-2 and Tunnel-2 groups received daily scruff restraint, while the Tail-1 and Tunnel-1 groups were handled as on days 1–5, except on day 12. Voluntary interactions were assessed on days 1, 5, and 11. On day 12, blood was collected 20 minutes after scruff restraint for pCORT measurements.

The experimental unit was the cage for voluntary interaction assessments and the individual mouse for pCORT measurements.

Sample size.

For the sample size calculation,²⁷ the significance level was set at 0.05 and the target power at 0.8. In Experiment 1, the sample size was calculated based on an effect size of 0.8 derived from previous research,¹⁴ with 8 mice (4 cages) per group, totaling 32 mice (16 cages). In Experiment 2, we used the same sample size and handling protocol as in Experiment 1 for pCORT measurement; this allowed detection of an effect size of 0.52 or greater. In Experiment 3, the sample size required to detect a large effect size (f = 0.4) for the interaction effect on pCORT was calculated, accounting for potential exclusions. The required sample size was 16 mice (8 cages) per group, totaling 64 mice (32 cages). Mice were divided into 2 runs to ensure adequate treatment capacity, and results were pooled for analysis.

If a mouse failed to enter the tunnel within 2 minutes during handling, it was to be excluded due to concerns about excessive guidance. However, no mice met this criterion; therefore, no exclusions were made.

Experimental animals.

Specific pathogen-free, 4-week-old female BALB/cAJcl mice (CLEA Japan, Tokyo, Japan) were used. The mice were acclimated for 13 days before the experiment. Housing conditions included a 12-hour light/dark cycle (lights on at 8:00 am), a temperature of 23 ± 3 °C, and a relative humidity of 50% ± 20%. The mice were housed in pairs in flat-bottomed cages (216 mm width × 377 mm depth × 148 mm height; CLEA Japan) with Pulsoft bedding (Oriental Yeast, Tokyo, Japan). Food (solid F-2 feed; Funabashi Farm, Chiba, Japan) and water were provided ad libitum.

Each cage was enriched with 3 sheets of nesting material (Nesting Sheet; Bio-Serv, Flemington, NJ) and a transparent acrylic tunnel (50 mm diameter × 150 mm length; Hikari, Osaka, Japan). The tunnel was placed along the inner side of the cage. Bedding and nesting materials were changed weekly, water bottles were changed twice weekly, and food was replaced every 2 weeks. One mouse in each pair was marked on the back with Animal Marker (Fuchigami Kikai, Kyoto, Japan) for identification.

After completing Experiment 1, the mice were transferred to a different study, which was neither Experiment 2 nor Experiment 3. Mice from Experiments 2 and 3 were euthanized by exsanguination under isoflurane anesthesia (Viatris Pharmaceuticals Japan, Tokyo, Japan) after blood collection from the posterior vena cava for pCORT measurement. Monitoring tests using sentinel animals confirmed that all mice were free of the following pathogens: Pseudomonas aeruginosa, Salmonella spp., Citrobacter rodentium, Rodentibacter pneumotropicus, Helicobacter spp., Clostridium piliforme, Corynebacterium kutscheri, Mycoplasma pulmonis, Ectromelia virus, mouse hepatitis virus, lymphocytic choriomeningitis virus, Sendai virus, intestinal protozoa, pinworms, and ectoparasites.

Handling.

Handling was performed by the same investigator in the morning during the light period. Gloves were changed for each cage to prevent contamination from the scent of mice in other cages. After removing the cage lid, nesting material, and tunnel, the handler stood motionless in front of the cage for 60 seconds. The mouse was then picked up using the assigned method. For tail handling, the mouse was grasped at the base of the tail between the thumb and forefinger and lifted onto the handler’s arm. After being held for 30 seconds, the mouse was returned to its home cage. For tunnel handling, the tunnel placed in each home cage as an enrichment item was used. The handler held the tunnel along the inner side of the cage and guided the mouse into it by creating a path with the other hand. Once the mouse entered the tunnel, it was lifted and held for 30 seconds before being returned to the home cage.

Scruff restraint.

Gloves were changed for each cage to prevent contamination from the scent of mice in other cages. Each mouse was picked up using the assigned method and placed on the cage lid. It was then held by the tail and immobilized by grasping the scruff (loose skin on the back of the neck) between the thumb and forefinger. The mouse was positioned on its abdomen, facing into the cage, for 10 seconds and then released into its home cage.

Voluntary interaction.

Voluntary interactions were measured as previously described.¹ Gloves were changed for each cage to prevent contamination from the scent of mice in other cages. After removing the cage lid, nesting material, and tunnel, the investigator stood motionless in front of the cage for 60 seconds. The handling device, either a gloved hand (tail handling) or a tunnel held by hand (tunnel handling), was placed in the front half of the cage for 60 seconds. Behavior was recorded with a video camera, and the recordings were later analyzed. Interactions were defined as sniffing (nose within approximately 5 mm), touching (including paw contact), climbing on, or chewing the handling device. The total time spent interacting with the handling device was calculated for each mouse within a cage. Interaction time was expressed as the proportion of the 60 seconds spent interacting. A voluntary interaction test was conducted immediately before and after handling or scruff restraint, as described in previous studies.^1,2 However, because this study focused on responses after restraint, only post-handling/restraint interaction data were analyzed.

Plasma corticosterone.

Blood collection began after 8:15 am and was completed by 11:30 am at the latest. To avoid skewing the timing of blood collection by group, 4 cages (one cage per group) were selected as one set and subjected to consecutive blood collection. A maximum of 4 sets were processed per day. The 2 mice in each cage were bled consecutively at 5-minute intervals. More detailed procedures for blood collection are described in the Supplementary Methods.

Under anesthesia, a 23-gauge needle was inserted into the posterior vena cava, and approximately 0.3 mL of blood was collected. The samples were anticoagulated with sodium heparin and placed in tubes on ice. Subsequently, the samples were centrifuged at 3,000 × g for 10 minutes at 4 °C to obtain plasma. The plasma samples were stored at −80 °C until analysis. Samples were diluted 10-fold, and pCORT concentrations were measured using an ELISA kit (Assay Max CORT ELISA Kit; Assaypro, St. Charles, MO) according to the manufacturer’s instructions (Version 12.3). All samples were assayed in duplicate, and the absorbance at 450 nm was read using a microplate reader (SpectraMax 190; Molecular Devices, San Jose, CA). The standard curve was generated using a 4-parameter fit model.

Statistical methods.

Data are presented as medians with individual values. Voluntary interaction data were averaged for each cage and expressed as acceptance values. Interactions before the restraint period were compared based solely on handling methods. During and after the restraint period, analyses were conducted to examine the effects of 2 or 3 factors: handling method, presence or absence of restraint, and time (before compared with after the treatment period). The Mann–Whitney U test was used for comparisons between 2 groups. Two-way ANOVA or 2- or 3-way repeated-measures ANOVA was used to evaluate the effects of multiple factors. When a 3-way interaction was significant, a simple interaction effect analysis was performed as a post hoc test. For significant 2-way interactions, the Steel–Dwass test and the Wilcoxon signed-rank test with Bonferroni correction were used for multiple comparisons of unpaired and paired data, respectively. Data were transformed using the aligned rank transform²⁸ for ANOVA. Effect sizes, estimating the magnitude of factor effects, were calculated as partial eta-squared values (${\epsilon }_p^2$) for ANOVA and r-squared values (r²) for other tests.^29,30 A significance level of 5% was used for all tests. Statistical analyses were performed using Bell Curve for Excel, version 4.08 (Social Survey Research Information, Tokyo, Japan).

Results

Voluntary interaction with the handling device using each handling method.

First, the effects of handling methods and single scruff restraint on voluntary interactions were examined (Figure 1A). On days 1 and 5, before scruff restraint, the handling method significantly affected voluntary interaction: mice in the Tunnel-1 and Tunnel-2 groups spent more time interacting with the handling device than mice in the Tail-1 and Tail-2 groups (P < 0.001; r² = 0.69 on both days; Figure 1B).

No significant interaction was observed between the handling method and scruff-restraint treatment on voluntary interaction, either immediately after scruff restraint (day 11) or 24 hours later (day 12) (P ≥ 0.141; ${\epsilon }_p^2$ ≤ 0.10; Figure 1B). The restraint treatment itself also had no significant effect on voluntary interaction (P ≥ 0.091; ${\epsilon }_p^2$ ≤ 0.15; Figure 1B). In contrast, the handling method significantly affected voluntary interaction: mice in the Tunnel-1 and Tunnel-2 groups continued to interact more (P < 0.001; ${\epsilon }_p^2$ ≥ 0.74; Figure 1B).

Effect of handling method and single-scruff restraint on plasma corticosterone concentrations.

In Experiment 2, the stress response following a single scruff restraint was examined (Figure 2A). Voluntary interactions on days 1–11 in Experiment 2 were similar to those observed during the prerestraint period in Experiment 1 (Figure S1).

The interaction between the effects of handling method and single scruff restraint was not significant (P = 0.654; ${\epsilon }_p^2$ < 0.01; Figure 2B). The handling method also had no significant effect on pCORT concentrations (P = 0.346; ${\epsilon }_p^2$ < 0.01; Figure 2B). In contrast, restraint significantly affected pCORT levels: mice in the Tail-2 and Tunnel-2 groups had higher pCORT concentrations than mice in the Tail-1 and Tunnel-1 groups (P = 0.049; ${\epsilon }_p^2$ = 0.10; Figure 2B).

Effect of handling method and repeated-scruff restraint on voluntary interaction and plasma corticosterone concentrations.

To assess the impact of repeated restraint, Experiment 3 examined the effects of handling method and repeated scruff restraint on voluntary interaction and stress response (Figure 3A).

As mentioned above, voluntary interaction on days 1 and 5 (pre-restraint) was higher in tunnel-handled mice than in tail-handled mice (Figure S2A). The effects of handling method and restraint treatment on voluntary interaction on day 11, following repeated restraint, were then assessed. A significant interaction between these factors was observed (P < 0.001; ${\epsilon }_p^2$ = 0.52; Figure 3B). A post hoc test showed that mice in the Tunnel-1 group spent more time interacting with the handling device than those in the other groups (P ≤ 0.009; r² ≥ 0.62; Figure 3B). In addition, mice in the Tunnel-2 group spent more time interacting than mice in the Tail-1 and Tail-2 groups (P ≤ 0.032, r² ≥ 0.47; Figure 3B). In contrast, no significant difference was observed between mice in the Tail-1 and Tail-2 groups (P = 0.152; r² = 0.28; Figure 3B). Furthermore, when examining the effects of restraint and the changes between the pre- and post-restraint periods within each handling method, only the mice in the Tunnel-2 group showed a decrease in voluntary interaction from day 5 to day 11 (P = 0.047; r² = 0.79; Figure S2B).

Regarding the stress response associated with repeated scruff restraint, no significant interaction between the effects of handling method and restraint treatment was observed (P = 0.732; ${\epsilon }_p^2$ < 0.01; Figure 3C). The handling method also had no significant effect on pCORT levels (P = 0.460; ${\epsilon }_p^2$ < 0.01; Figure 3C). In contrast, repeated restraint significantly affected pCORT: mice in the Tail-2 and Tunnel-2 groups showed higher pCORT concentrations than mice in the Tail-1 and Tunnel-1 groups (P = 0.002; ${\epsilon }_p^2$ = 0.13; Figure 3C).

Discussion

Understanding the effects of handling on the stress response associated with routine procedures in mice is important for animal welfare and the accurate interpretation of experimental results. In this study, we investigated how handling methods affect the stress response to scruff restraint, a routine procedure, in female BALB/c mice. The results showed that pCORT concentrations were higher in restrained mice than in unrestrained mice, regardless of whether the restraint was applied once or repeatedly. In contrast, the handling method had no significant effect on pCORT levels following either single or repeated scruff restraint.

Before restraint, tunnel-handled mice spent more time interacting with the handling device than tail-handled mice. This difference in voluntary interaction was consistent with previous reports,¹ confirming that handling was performed in accordance with established protocols. In other words, we ensured that the mice exhibited the expected handling-specific behavioral responses.

Despite being handled by their tails during the pre-experimental period, the tunnel-handled mice interacted more with the handling device (the tunnel) on day 1. A previous study¹ also reported high levels of interaction with a tunnel on the first day of a voluntary interaction test in mice that had been handled by their tails before the experiment. In contrast, mice that were handled with the tunnel before the experiment and then presented with a human hand in the voluntary interaction test did not interact more on the first day.¹⁴ These findings suggest that, on the first day of the experiment, mice readily approach a tunnel but do not readily approach a human hand, regardless of whether they were previously handled by their tail or by a tunnel. One possible reason mice prefer the tunnel is that they are already familiar with it, and they may also use it as a refuge. Furthermore, mice in the Tunnel-1 group exhibited more interaction than those in the Tunnel-2 group in Experiment 3, suggesting that repeated tunnel handling may cause mice to exhibit reduced wariness or increased habituation.

Our study did not investigate how mice perceive the tunnel. To interpret the emotions underlying mouse behavior observed in voluntary interaction tests, a multifaceted evaluation is required, including validated behavioral assessments. Understanding how mice prefer to interact with handlers and handling devices would contribute to the development of more appropriate handling methods.

Scruff restraint increased pCORT concentrations in both tail- and tunnel-handled mice, with no significant relation between handling method and restraint. Several confounding factors may have influenced pCORT levels in this study. Mice subjected to scruff restraint were moved between the rack and the table up to 3 times more than the non-restrained group on the blood collection day, and this additional cage movement could have increased pCORT.¹¹ Therefore, the increase in pCORT observed in the Tail-2 and Tunnel-2 groups may have been due to a series of procedures associated with scruff restraint, including cage movement.

Isoflurane anesthesia is also known to influence pCORT.³¹ A time lag of 3–5 minutes exists between exposure to a stressor and the subsequent rise in circulating corticosterone levels.³² In this study, blood collection was completed within 4 minutes of initiating isoflurane anesthesia to minimize its influence. However, a higher concentration of plasma corticosterone was observed in female rats immediately following 5 minutes of isoflurane exposure.³³ In female NMRI mice, an increase in corticosterone concentration attributed to a blood collection procedure performed as a stressor 20 minutes prior has been detected after exposure to 4.5% isoflurane for 2.4 ± 2.2 minutes (average ± SD).³⁴ In addition, differences in corticosterone concentrations between blood collection methods have also been detected.³⁴ These findings suggest that although brief exposure to isoflurane is a confounding factor, an increase in pCORT levels due to the experimental intervention can be detected to a certain magnitude. Therefore, the results of this study suggest that scruff restraint elevates pCORT levels, while handling history has little effect on them.

Scruff restraint is known to cause changes in physiologic parameters such as heart rate and core body temperature in mice¹² and to increase pCORT.¹³ The degree of this increase varies with the type of restraint; for example, scruff restraint induces lower pCORT levels than restraint using a resin tube restrainer (Perspex restrainer).¹³ In this study, pCORT concentrations following scruff restraint were not extremely elevated, indicating that the stress induced was relatively mild.

Previous studies^6,9 have assessed the effects of handling methods on pCORT either in isolation⁸ or in combination with experimental procedures. One study⁸ reported higher pCORT in tunnel-handled mice, possibly due to the time required for the mouse to enter the tunnel. In our study, mice were guided into the tunnel without delay, and pCORT did not increase notably with tunnel handling. Studies using the elevated plus maze and tube restraint have also reported no significant differences in pCORT between tail handling and less-aversive handling methods.^6,9 Although our study was not designed to draw a definitive conclusion, the handling methods appear to have little or no impact on pCORT under mild stress.

These results may not meet the expectations of those who anticipated that tunnel handling would suppress pCORT, but they also provide reassurance: tunnel-handled mice did not exhibit a greater stress response than tail-handled mice. Some researchers have expressed concern that mice handled without direct contact (as in tunnel handling) may react more strongly during experimental procedures that require physical contact.¹⁰ However, tunnel handling is widely regarded as a refinement that improves animal welfare due to higher voluntary interaction.¹⁰ Our findings indicate that tunnel-handled mice do not have a heightened pCORT response, addressing these concerns and supporting the broader implementation of tunnel handling.

Voluntary interaction was reduced in tunnel-handled mice immediately after repeated scruff restraint compared with unrestrained mice. Although the effect of a single scruff restraint was not statistically significant, effect size analysis indicated that even a single scruff restraint may reduce interaction, particularly immediately afterward. Comparing effect sizes across conditions, repeated restraint had the strongest effect, followed by immediate and 24-hour post-single restraint. This suggests that repeated scruff restraint could reduce interaction with the handling device.

Reduced voluntary interaction after repeated treatment has also been reported following scruff restraint and oral administration.^14,35 One possible explanation is that mice associate being picked up with restraint, leading to tunnel avoidance. Notably, the low level of voluntary interaction in tunnel-handled and scruff-restrained mice remained higher than that in tail-handled and unrestrained mice, suggesting that the benefits of tunnel handling remained.¹⁴ However, reduced interaction can increase the time required to guide mice into the tunnel, potentially raising pCORT⁸ and prolonging care and experimental procedures. To address this, refined tunnel-guiding techniques¹ and alternative strategies, such as using a different, less-aversive method (for example, cupping) during experimental procedures, may help prevent mice from associating the tunnel with aversive events. Further research is needed to explore this possibility.

Stress sensitivity in mice varies with strain and sex.^19–26 Therefore, determining whether these findings in female BALB/c mice, known for their high stress sensitivity, are generalizable to other strains remains an important question. Although pCORT, evaluated in this study, is a typical stress marker, it does not capture all facets of the complex stress response. Furthermore, the stress response differs depending on the method of stress loading.¹³ Studying mice with lower stress sensitivity, such as C57BL/6, using stress markers other than pCORT, and applying different methods of stress loading is necessary to achieve a more comprehensive understanding of how handling methods affect stress responses in mice.

In this study, we found no significant effect of handling method on pCORT following either single or repeated scruff restraint. This finding suggests that the handling method has little to no effect on pCORT after scruff restraint. In addition, tunnel-handled mice, which have less frequent contact with human hands, did not appear more sensitive to experimental stress than tail-handled mice. The observed decrease in voluntary interaction in tunnel-handled mice after scruff restraint highlights the need for further refinement in the effective implementation of tunnel handling.

These results may assist investigators who have been reluctant to adopt tunnel handling due to concerns about the consistency of data obtained using tail handling, helping them determine whether differences in handling methods constitute a confounding factor across experiments. The findings also help temper excessive expectations or concerns regarding the influence of less-aversive handling methods on experimental outcomes, supporting their use based on an informed understanding.

Supplementary Materials

Supplementary Methods. ^{(90.2KB, pdf)} Procedures for treating mice for blood collection.

Figure S1. ^{(169.9KB, pdf)} Voluntary interaction with the handling device before single-scruff restraint in Experiment 2.

Figure S2. ^{(152KB, pdf)} Effect of handling method and repeated-scruff restraint on voluntary interaction with the handling device in Experiment 3.

Conflict of Interest

The authors have no conflicts of interest to declare.

Funding

This work was internally funded.

Protocol registration

This protocol was not registered with any external resources.

Data availability

All data collected and analyzed in this study are presented in the manuscript and supplementary figures.