Automated tests for measuring the effects of antidepressants in mice

https://doi.org/10.1016/j.pbb.2004.03.014Get rights and content

Abstract

The forced swim test (FST) and the tail suspension test (TST) are used widely for measuring the pharmacological effects of antidepressant drugs or changes in stress-evoked behavior in mice. However, inconsistent scoring techniques and poor reproducibility may result from their reliance on subjective ratings by observers to score behavioral changes. In this paper, automated versions of the mouse FST and TST were characterized and validated against observer ratings. For the FST, a commercially available video tracking system (SMART II; San Diego Instruments) measured the duration that mice swam in water-filled cylinders at a set velocity. For the TST, a commercially available automated device (Med Associates, St. Albans, VT) measured input from a strain gauge to detect movements of mice suspended from an elevated bar. Dose-dependent effects of the antidepressant desipramine on FST and TST immobility were measured in CD-1 mice using both automated devices and manual scoring from videotapes. Similar dose–response curves were obtained using both methods. However, a wide range of correlations for raters in the FST indicated that scoring criteria varied for individual raters despite similar instructions. Automated versions of the mouse FST and TST are now available and provide several advantages, including an opportunity to standardize methods across laboratories.

Introduction

The forced swim test (FST) and the tail suspension test (TST) are used widely for measuring the pharmacological effects of antidepressant drugs or changes in stress-evoked behavior in mice (Cryan et al., 2002). In the mouse FST Porsolt et al., 1977a, Porsolt et al., 1977b, a mouse is placed in a cylinder of water from which there is no escape. Although procedures can vary, behavior is most commonly measured in a single session lasting for 6 min. Mice usually display escape-oriented behaviors when immediately placed in the cylinder, consisting primarily of swimming across the water surface, although climbing behavior could be directed at the cylinder walls. After several minutes, the behavior of mice consist predominantly of bouts of immobility and passive floating, and pretreatment with antidepressant drugs reduce the amount of time spent in the latter two behaviors. The FST is sensitive to all major classes of antidepressant drugs (Borsini and Meli, 1988), including tricyclics, selective norepinephrine and serotonin reuptake inhibitors, monoamine oxidase inhibitors, and atypical antidepressants Koe et al., 1983, Kulkarni and Mehta, 1985, Cesana et al., 1993, Nixon et al., 1994, Bourin et al., 1996, Redrobe et al., 1996, Da-Rocha et al., 1997, Sanchez and Meier, 1997. The FST is also performed with rats (Porsolt et al., 1977b), usually using different procedures, and modifications of it allow one to distinguish serotonergic and noradrenergic antidepressants Detke et al., 1995, Lucki, 1997.

In the TST (Steru et al., 1985), mice are suspended by the tail from an elevated bar for several minutes. Typically, they immediately engage in several escape-oriented behaviors, such as leg kicks and body jerks, followed temporally by increasing bouts of immobility. The frequency of immobility is reduced by antidepressant treatments. The TST has been shown to be sensitive to an array of antidepressants, including tricyclics, selective serotonin reuptake inhibitors, monoamine oxidase inhibitors, atypical antidepressants, and electroconvulsive therapy Steru et al., 1985, Steru et al., 1987, Perrault et al., 1992, Teste et al., 1990, Teste et al., 1993. Tests of locomotor activity are often used to distinguish the effects of antidepressants from general psychomotor stimulants, because most antidepressants do not increase activity at doses that reduce FST and TST immobility.

Although used most often in drug discovery research, the mouse FST and TST are used by divergent laboratories in the field of psychopharmacology and genetics research (Cryan et al., 2002). Significant procedural drawbacks to the use of these models include their lack of consistency in reference experimental conditions and reliance on subjective ratings by observers to score behavioral changes. Individual raters and different laboratories apparently use distinct criteria for judging immobility, based on the range of immobility scores reported in publications. For example, baseline immobility values for DBA/2 mice on the FST range from 99 (Lucki et al., 2001) to 234 (David et al., 2003) s out of a possible 240 s (final 4 min of a 6 min test). Although some laboratories use videotaped sessions to archive reference standards for training, subjective rating methods can hamper attempts at replicating findings across laboratories and the maintenance of standardized scoring methods over time in the same laboratory. Manual scoring is also tedious and time consuming, especially when a large number of animals are being tested. Extra personnel may also be required so that raters are unaware of the experimental conditions for individual animals. Automated versions of tests would be extremely desirable for several reasons: (1) assurance of consistent immobility ratings, (2) an opportunity to standardize methods across different laboratories, and (3) greatly increased throughput. However, any proposed automated system needs to be validated against manual scoring techniques and for the ability to detect antidepressant drug effects. In this article, automated versions of the mouse FST and TST are described and validated.

Section snippets

Animals

Male CD-1 mice 10 to 14 weeks old were purchased from Charles River (Wilmington, MA). Subjects were housed in groups of four per cage (cage size: 28.5×17.5×13.0 cm) in a temperature-controlled environment (22±1 °C) under a 12-h light–dark cycle, with lights turned on at 0700 h, for at least 1 week prior to testing. Food and water were freely available. All behavioral testing was performed between 1000 and 1800 h. All procedures are in compliance with the National Institutes of Health Guide for

Results

Fig. 1 shows the effect of desipramine on FST immobility values determined by the automated FST device and three experienced raters (all authors; identified by numbers). Two-way ANOVA indicated that both the dose of desipramine [F(3,268)=32.6, P<.0001] and the identity of the rater [F(3,268)=15.5, P<.0001] had a significant effect on immobility values, but no significant interaction between dose of desipramine and rater identity was detected [F(9,268)=0.68, P=.72]. Follow-up tests indicated

Discussion

The present study validated automated versions for measuring the FST and TST in the mouse by examining dose–response curves for the antidepressant drug desipramine and by comparing the similarity of scores from experienced human raters with the automated devices. Although this article is primarily methodological, the findings have practical utility for labs that use the mouse FST or TST. The SMART II system from San Diego Instruments is utilized as a behavioral tracking device for locomotor

Acknowledgments

This research was supported by USPHS grants MH 01465 and MH 48152. The authors are grateful to Dr. Richard Butcher and to San Diego Instruments for providing a prototype SMART II video tracking system for supporting this research.

References (27)

  • M.A Da-Rocha et al.

    Influence of anxiolytic drugs on the effects of specific serotonin uptake

    J. Psychopharmacol.

    (1997)
  • D.J.P David et al.

    Antidepressant-like effects in various mice strains in the forced swimming test

    Psychopharmacology

    (2003)
  • M.J Detke et al.

    Active behaviors in the rat forced swimming test differentially produced by serotonergic and noradrenergic antidepressants

    Psychopharmacology

    (1995)
  • Cited by (97)

    View all citing articles on Scopus
    View full text