J. Mos, R. van Oorschot and S. Swidde

Department of Pharmacology, Solvay Duphar b.v., Weesp, The Netherlands

Although we use a variety of different techniques to detect and describe behavioural effects of new drugs, we have not yet used automated video analysis extensively to analyze drug effects on locomotion parameters. The present version of EthoVision offers a number of features with which path parameters can be described. With path parameters those characteristics are meant that adequately describe the way in which an animal moves through a test arena. In particular, the velocity of the animal, the average distance moved between successive samples, turn angles, meander or sinuosity can be quantified, apart from the more obvious measures such as total distance moved. Although the latter measure is still quite often used in behavioural pharmacology, a more elaborate analysis might be useful to characterize unknown compounds.

In this presentation we will present our data on two dopamine agonists, the non-selective dopamine agonist quinpirole and the indirect agonist amphetamine. These drugs were tested in mice and rats and the optimal path parameters were analyzed to describe drug induced changes in locomotion. We measured the behaviour during one hour after administration of the drug and analyzed the data in blocks of fifteen minutes.

Quinpirole (0.03, 0.3 and 3 mg/kg ip) reduced locomotion in rats specifically at low doses, but slightly increased locomotion at the higher doses. Whether a large or a small observation cage was used made no difference. Quinpirole reduced the speed of locomotion at all doses tested and decreased the absolute angular velocity. Whether this is a side effect of effects on the velocity can not easily be established. Although the plot analysis of the path suggested a more meandering gait', this could not be quantified in the meander and sinuosity parameters. A newly developed measure for the average length of a straight pathway, however, revealed statistically significant effects. In mice the hypolocomotion was more evident at the higher doses and was of shorter duration than in rats. Similarly the effects on the velocity were less marked and at low doses even absent. Again the path parameters only showed moderate effects on the angular velocity and the meander. Although a careful analysis of the minimal-distance-moved criterion is necessary (in our hands 1.5 cm with two samples/s), the obtained optimal values did not lead to greatly improved statistical significances.

Amphetamine (0.3, 1 and 3 mg/kg ip) dose-dependently increased locomotion, i.e. the total distance moved increased as well as the velocity. This was irrespective of the size of the observation cage. The visual observation that the animals showed more straightforward locomotion could not be verified with the meander and sinuosity measures, but the average length of a straight pathway was increased. By and large the same conclusions were drawn in the experiments using mice.

In summary, the EthoVision system could be used to quantify velocity and distance moved quite reliably: the expected drug effects were quantified in a reliable way. The other path parameters were less informative with regard to drug effects. The major advantage of the present system is the possibility to determine time of onset and duration of action of a drug and the capacity to test more than one animal at the same time. Currently we are testing some improved algorithms to verify whether the human observations on path parameter changes can be reliably expressed in other ways.

Paper presented at Measuring Behavior '96, International Workshop on Methods and Techniques in Behavioral Research, 16-18 October 1996, Utrecht, The Netherlands