Keywords Locomotor activity pattern; distress; Freund's complete adjuvant; mice. Current systems of assessing pain and distress in experimental animals are ...
.
.
Change in locomotor activity pattern In mice: a model for recognition of distress? C. Jansen van't Land & C. F. M. Hendriksen Central Animal Laboratory, National Institute of Public Health and Environmental Protection, P.O. Box 1, 3720 BA Bilthoven, The Netherlands
Summary Recognition and assessment of pain and distress is made by observing common clinical and behavioural signs. Observation usually occurs during a limited period of time and results can be biased by interpretation of an individual observer. To improve objective assessment of distress we studied the locomotor activity pattern of mice during a 24-h interval. As a reference compound, Freund's complete adjuvant [FCA) was used. Mice were injected intraperitoneally with different doses FCA (0, 0.1, 0.2 and 0.5 ml) and observed for 5 to 7 days. Animals did not appear to be in pain and seemed to have a normal activity and behaviour pattern at first sight, however FCA induced a dose-dependent decrease of body weight. Open field activity (total distance run) measured during a limited period of time was not altered as a result of FCA. However, nocturnal activity was dose dependently decreased during the first 3 to 4 nights after treatment with FCA. The data presented indicate that using locomotor activity patterns over 24 h might be a useful adjunct and an objective approach to assess distress. Keywords Locomotor activity pattern; distress; Freund's complete adjuvant; mice
Current systems of assessing pain and distress in experimental animals are based on observation of common clinical and behavioural signs. The method most frequently used is that of Morton and Griffiths (1985). The outcome of this method might be influenced by the individual observer and depends on experience and personal interpretation of parameters. A more objective representation of the degree of pain or distress can only be obtained as the mean of the results of a large number of observers. Nevertheless, it is obvious that the selection of parameters and the experience and/or attitude of the observer are critically important when assessing the degree of distress in experimental animals (Beynen et al. 1989). Methods which are objective in nature are video-analysis and telemetry but the disadvantages of these methods are that they are rather laborious. Accepted 23 August 1994
In our experiments we studied a fully automatic system based on the locomotor activity of individual mice over 24 h to assess distress. The model is based on the number of interruptions per hour of 2 crosswise placed infrared photocells. The model was originally developed and validated for rats by Van der Laan et ai. [1991, 1994) for the assessment of tolerance and withdrawal symptoms during and after chronic morphine or benzodiazepine treatment. In our model FCA (Difco Laboratories, Detroit Michigan, USA) was used as a reference compound. Immunization with FCA in animals leads to (histo)pathological changes (Brodersen 1989, Claassen et ai. 1992, Rigdon & Schadewald 1972) which are known to be very painful in man (Chapel & August 1976). Therefore, in the Netherlands administration of FCA in animals is classified as 'severe distress'. Laboratory Animals (1995) 29, 286-293
Change in activity: a model for distress?
Nevertheless, signs of pain and distress in mice ILeenaarsj personal communication) and rabbits (Leenaars et a1. 1994) are rarely seen. Toth et a1. (1989) studied the locomotor activity of mice during treatment with FCA, but only during a limited period of time (5 min daytime). Based on these results they concluded that, whilst intraperitoneal injection of FCA is associated with several side effects, it did not cause signs of severe discomfort. Animals, materials and methods Animals Female albino Rivm:N:NIH mice (RIVM, Bilthoven, The Netherlands) bred in a strict barrier maintained SPF colony were used in the experiments. The animals (9-12 weeks of age) were housed in groups of 3. Food and water were available ad libitum and a light-dark schedule of 12h 16.00a.m. lights on) was maintained. Temperature was between 20-23°C. All animals were clinically normal at the beginning of the study and were randomly distributed in test and control groups. All experiments were performed in a soundproof room. Measurement of locomotor activity in open field Activity was measured in an automated open field [40x 40 cm) with 15 infrared lights on each side. The open field was placed in a dark room and illuminated in its centre by a red lamp (60W). On the first day after treatment the animals (injected with different volumes of FCA) were tested at random (between 8.00 and 1l.00a.m.) to avoid any influence of circadian rhythm. The location of the mice in the field was recorded every 0.5 second for 10 min and different parameters were calculated, the most relevant being: the total distance mn, and the difference of the distance mn during the 1st and 2nd periods of 5 min (ie. distance run during the 1st 5 min minus the distance run during the 2nd 5 min). The latter parameter is to estimate progress in explorative activity.
287
Between tests the open field was cleaned using 1% acetic acid to exclude any effect of the previous animal. Measurement of locomotor activity in home cages Animals were placed individually in standard cages (22x 16 x 16 em: 1x b x h) with a small amount of sawdust bedding. Food and water were available ad libitum. Between 08.00 and 09.00 hours the animals were weighed and clinically observed. The activity was measured as the number of interruptions per hour of the light beams of 2 infrared photocells placed crosswise (Sick FR2-122), counted by 2 8-channel counters each consisting of an ARCOM ARC41 single board computer. These 2 counters were coupled to an Olivetti Pc. The data were transmitted to a Micro VAX and analysed. Daytime activity is given as the mean counters per hour from 06.00 till 18.00 h, excluding the period from 08.00 till 10.00 h during which weighing and observation procedures were carried out. Nocturnal activity is given as the mean counts per hour from 18.00 till 06.00 h. Statistical analysis The data were tested on homogeneity of variance between treatment groups. If the variance appeared to be homogeneous an ANOVA was carried out post hoc followed by a t-test and if relevant, a linear regression analysis. If the variances were nonhomogeneous, the nonparametric Mann-Whitney V-test (MWV-testJ was used to test the differences between two means. Design of the experiments On the initial day of the experimental period (day 0), mice (6 animals per group) were injected intraperitoneally (i.p.) with different quantities of Freund's complete adjuvant (0.05, 0.1, 0.25 ml) emulsified with an equal volume of saline [i.e. the total volumes injected were 0.1, 0.2 and 0.5 mIl. These volumes are used to describe groups in Results and Discussion. Control animals were injected intraperitoneally
288
Jansenvan't land & Hendriksen
CHANGE IN BODY WEIGHT
CHANGE IN BODY WEIGHT
%
%
J
2
4
5
3
2
DAY
4
5
DAY
Fig 1 Experiment A. Change in body weight after intraperitoneal injection with different doses of FCA. Mice were weighted daily between 08.00 and 09.00 h. (bars representing SEMare not indicated for clarity). The mean SEMfor FCA0.0.1,0.2 and 0.5 is 0.7, 0.9,0.8 and 1.1 resp.; Fig lA and 1.4, 1.0, 1.2 and 2.7; Fig lB. A: Change in body weight in % from the day before. B: Change in body weight in % from day O. +----+ control treatment (n = 6); 0 ----0 0.1 ml FCA (n = 6); x----x 0.2 ml FCA (n = 6); v ---- v 0.5 ml FCA (n = 5).
with 0.2 ml of saline only. Following FCA administration the mice were weighed every day. For practical reasons, the experiment was divided into 2 sessions (in the 1st and 2nd week 3 animals of each group). For reasons given below the study consisted of two experiments. Experiment A: the general condition of the animals was assessed daily and at the same time the degree of pain and distress was assessed by 2 independent observers using the method of Morton and Griffiths (1985). Bodyweight, appearance, clinical signs, unprovoked behaviour and behavioural responses to external stimuli were scored as described for this method. Open field locomotor activity was measured on the first day after administration. Directly after the open field test the animals were placed in their individual home cages and the
locomotor activity was measured 24 h a day for 4 days. On the morning of the 5th day the animals were killed for macroscopic examination. Experiment B: was essentailly the same as experiment A apart from the observation period which was increased from 5 to 7 days.
Results Experiment A:
After administration of FCA some of the mice showed a 'starey' haircoat (not completely smooth as controls). Two animals of the highest dose group (FCA 0.5) developed mild diarrhoea. The mice did not appear to be in pain, since they did not vocalize or struggle when handled,
289
Change in activity: a model for distress?
NOCTURNAL LOCOMOTOR ACTIVITY
counl/h
DAYTIME LOCOMOTOR ACTIVITY
counl/h
2
A
2
3
B
4
2
3
4
DAY
DAY
Fig 2 Experiment A. Change in locomotor activity after intraperitoneal injection with different doses of FCA. The activity was measured over 24 h. The number of beam interruptions is given as the mean counts per hour. (bars representing SEM are not indicated for clarity). The mean SEM for FCA 0, 0.1, 0.2 and 0.5 is 25,33,14 and 25 resp.; Fig 2A and 15, 10,9 and 12; Fig 2B. A: Nocturnal locomotor activity. B: Daytime locomotor activity. +----+ control treatment (n=6); 0 ----0 0.1 ml FCA (n =6); x----x 0.2 ml FCA (n = 6); v ---- v 0.5 ml FCA (n = 5).
Table 1 Performance in open field (in em) during the first day after treatment Total distance run
Difference in distance run (between 0-5 and 5-10 minutes)
A Control 0.1 FCA 0.2 FCA 0.5 FCA
3174±436 2743±279 3560±280 2749±il90
374±44 252±106 899±88* 391 ± 210
B Control 0.1 FCA 0.2 FCA 0.5 FCA
33B8±204 3206±572 2279±366 3286±607
341±65** 421±91** 473±128** 660±125**
Controls (n=6); 0.1 ml FCA (n=6); 0.2 ml FCA (n=6); 0.5 ml FCA (n = 5). Data are given as the mean ±SEM. Statistical analysis was carried out versus control: * Mann-Whitney U-test P=0.002. ** Linear regression P< 0.05
neither did they show deviant activity or behaviour pattem. The changes noticed were so subtle that the observers concluded that clinical signs, appearance, behavioural responses to external stimuli and unprovoked behaviour should be scored between category a and 1 [according to Morton and Griffiths 1985). The overall assessment for nearly all animals was classified as normal. One of the observers concluded that one animal of the highest dose group (FCA 0.51 suffered mild distress. FCA induced a dose-dependent decrease of body weight on the 1st day (F[1,21] = 22.2j P
