Object recognition test in mice - Nature

protocol

Object recognition test in mice Marianne Leger1,3, Anne Quiedeville1,3, Valentine Bouet1, Benoît Haelewyn2, Michel Boulouard1, Pascale Schumann-Bard1 & Thomas Freret1,2 1Groupe Mémoire et Plasticité comportementale (GMPc), Université de Caen Basse-Normandie, Caen, France. 2Centre Universitaire de Ressources Biologiques –

Behavioral Research Platform (CURB - BRP), Université de Caen Basse-Normandie, Caen, France. 3These authors contributed equally to this work. Correspondence should be addressed to T.F. ([email protected]).

© 2013 Nature America, Inc. All rights reserved.

Published online 21 November 2013; doi:10.1038/nprot.2013.155

The object recognition test is now among the most commonly used behavioral tests for mice. A mouse is presented with two similar objects during the first session, and then one of the two objects is replaced by a new object during a second session. The amount of time taken to explore the new object provides an index of recognition memory. As more groups have used the protocol, the variability of the procedures used in the object recognition test has increased steadily. This protocol provides a necessary standardization of the procedure. This protocol reduces inter-individual variability with the use of a selection criterion based on a minimal time of exploration for both objects during each session. In this protocol, we describe the three most commonly used variants, containing long (3 d), short (1 d) or no habituation phases. Thus, with a short intersession interval (e.g., 6 h), this procedure can be performed in 4, 2 or 1 d, respectively, according to the duration of the habituation phase. This protocol should allow for the comparison of results from different studies, while permitting adaption of the protocol to the constraints of the experimenter.

INTRODUCTION Access to novelty (e.g., an object or an environment) can elicit approach behaviors in rats. Starting from this observation, notably made by Berlyne in 1950 (ref. 1), a new behavioral test was developed in the late 1980s: the so-called object recognition test2,3. Typically, the test itself is carried out in two sessions in the same context, divided by an intersession interval (ISI). During the first session (familiarization session), the animal is free to explore two similar objects, and during the second session (test session), one of the objects is replaced by a novel, unfamiliar object. As defined by its authors, this test is really a one-trial task, as it does not involve learning of rules. In addition, the test does not require reinforcers and is purely based on the innate preference of the rodent to explore the novel object rather than the familiar one. Thus, a rodent that remembers the familiar object will spend more time exploring the novel object. As this type of test was derived from the ‘visual pairedcomparison paradigm’ widely used both in humans and in nonhuman primates, in which a familiar stimulus and a new one are paired, and in which recognition is inferred from preferential looking to the novel target4, it also allows for interspecies comparisons. Such qualities have made this novel object preference test very popular in experimental research on memory5,6, for example, after neurotransmission system alterations by pharmacological challenges7, after brain lesions or models of brain diseases8 and in development or aging studies9. After advances in the transgenic and knockout mice technology, the object recognition test was adapted from rats to mice10,11 and was used to investigate several transgenic models of neurodegenerative disorders12. However, mice are not little rats and thus may neither behave in a similar way nor use similar strategies13. In addition, many variations on the original test have been reported in terms of object and arena features, animal strain and sex, exploration time and other procedural aspects. Unfortunately, specific behavioral variables that may modulate novel-object preference have not yet been thoroughly examined, and alternative explanations of memory deficit may explain a failure to observe novel-object preference14.

Variations in object recognition protocols Object exploration duration. One major variant in protocols is the object exploration duration. In the original protocol for rats3, a minimal exploration time for both objects during the test phase (~20 s) was used. Such a criterion ensures a similar exploration time of the two objects and between animals, independently of their individual exploratory activity. Thus, this criterion affords the advantage of decreasing the inter-individual variability and helps to strengthen the accuracy of the results obtained. Even though the time for exploration was extended in mice15–17, the measure of the absolute time spent exploring each object during each session (a fixed duration of classically 3 or 5 min) remains the most commonly used way to assess recognition memory performances18–20. Habituation. An additional variation is whether or not a habituation phase is performed. Some protocols preface the test phase with a habituation phase to the apparatus and to the procedure to reduce stress and to avoid a potential neophobic response11, and consequently to promote the exploratory activity of rodents toward the objects. In the original study, the rats were allowed to explore the box for 2 min the day before the test3. Since then, a multitude of protocols have emerged with various durations of habituation procedures ranging from a few minutes 16 to several days21. Among such protocols, the use of a 3-d habituation phase has been the most commonly used in mice20. It consists of exposure to an open-field apparatus twice a day for 3 d before the test (3 min for each session). A shorter habituation phase, consisting of a single exposure to the apparatus, has also been proposed16,22 but its usefulness remains a matter of debate. Although rodents explore novel stimuli in a familiar environment as opposed to a new one23, overall object exploration time depends on several factors, including the state of the animal21, the strain24 and the environment25. ISI. The object recognition test has been used in several fields of neuroscience research to investigate brain networks underlying recognition memory and their alteration during aging or nature protocols | VOL.8 NO.12 | 2013 | 2531

protocol


different neurodegenerative disorders. Differences appear according to the protocol used that may lead to different interpretations of the results. For example, the maximal ISI duration for observing significant performances between different strains of mice was limited to 4 h in the protocol of Sik et al.20, whereas it was 24 h for Frick and Gresack16. Whether this memory trace extension was related to the use of a criterion (total time spent exploring the objects limited to 20 s) or to the reduction of the habituation phase has not been determined. Unfortunately, the use of different methodological approaches does not allow comparison of the results between laboratories and may influence the choice of the protocol depending on the anticipated results. Experimental design To allow possible comparison, we adapted the three most commonly used variations performed in mice in an open field15,16,20 to establish this protocol. In the protocol of Sik et al.20, the habituation phase consists of two 5-min exposures to the apparatus per day separated by 6 h, during 3 d before the test phase (Step 1A). In the protocol of Frick and Gresack16, the habituation phase is reduced to a single 5-min exposure to the open field 24 h before the test phase (Step 1B). In the original protocol, there was also a 1-min habituation phase performed just before the familiarization session. In an effort to standardize the conditions of the test session, however, this 1-min habituation phase is not included in the present protocol. The third variant (Step 1C) presented here is adapted from the protocol of Arque et al.15, in which a criterion of 20 s is used for the two sessions to facilitate the analysis and the comparison of the results, with a maximal time of 10 min to reach the criterion. The three variants are referred to here as long (Step 1A), short (Step 1B) and no habituation (Step 1C), and memory performances were tested with each protocol after an ISI of 6, 24 , 48 and 72 h. The protocol described herein will greatly decrease interindividual variability and thus improve the reliability of the results obtained with this test in the strain of mice used (NMRI). Nevertheless, one has to keep in mind that there can always be some small interlab or interstrain differences. We thoroughly advise authors to add their own additional control groups when testing pharmacological treatments or transgenic animals (for example, a control group receiving a vehicle solution or a wild-type group). Equipment. A great variability exists within the open fields available, in terms of size, shape (square, circular or rectangular) and color. All the open fields sold commercially are square or rectangular, and their size is between 27.3 × 27.3 × 20.3 cm (Noldus or Med Associates) and 50 × 50 × 50 cm (Cleversys) for mice. To our knowledge, there is no paper that directly compares object recognition performances of mice according to the shape of the testing arena, i.e., either square or circular. Nonetheless, Kalueff et al.26 have demonstrated that mice behave similarly in an open field (vertical, unsupported vertical and horizontal activity, grooming, defecation, urination and wall-leaning activity) regardless of its shape. Here we chose to use a black wooden box (33 × 33 × 20 cm). Concerning the objects, the experimenter should observe the features described by Ennaceur27 with respect to the brightness, shape, color, odor and texture of the objects . The two objects may have the same size and odor (no odor seems ideal). 2532 | VOL.8 NO.12 | 2013 | nature protocols

An asymmetric object should always be presented the same way. As rodents have difficulty in discriminating colors, the experimenter should focus on the brightness and texture of the objects. The goal is to maximize the difference between the objects without inadvertent induced preference for one of the objects. In addition, when choosing the objects, the experimenter must pay attention to the recommendations of Heyser and Chemero13. Allowing mice the possibility to climb over the objects provides for a sufficient level of exploration without habituation to the objects. Of course, when using a video-tracking package, one may choose objects that cannot be climbed onto, as these systems will always score climbing as exploration. Finally, the weight of the objects must be such that mice cannot move them. If heavy objects are not possible, double-sided adhesive tape or Patafix may be used to hold the objects stuck on the floor. In the experiments reported here, we used common objects that differ in shape and texture; both of these objects can be climbed onto by mice, and both stand out against the background: towers of Lego bricks (8-cm high and 3.2-cm wide, built in white, blue, yellow, red and green bricks) and Falcon tissue culture flasks filled with sand (9.5 cm high, 2.5 cm deep and 5.5 cm wide, transparent plastic with a blue bottle cap) (Fig. 1). These two objects can be easily obtained in laboratories. Timing of the test. Exploratory behavior is influenced by the animal’s state, which may vary according to the circadian rhythms and light cycle28. In experiments using the protocol described herein, mice were housed in a reversed light-dark cycle and were always tested in the dark phase (active phase between 08:00 and 20:00). The familiarization session was always carried out during the morning (from 09:00). Thus, similarly to the anxiety-related behavioral tests29, the timing of the test is crucial and must be taken into consideration when setting up experiments, as it is a factor that can explain differences in durations to reach the criterion. Handling of animals before testing. Because it has been demon strated that prior experience can alter behavioral responses of rodents in the object recognition test7, animals were consistently accustomed to being handled by experimenters. It is important to ensure that handling before the experiment is consistent among animals and treatment groups. Ideally, the experimenter should handle the mice twice a week for 1 min each session for at least 1 week before the beginning of the experiments so as to avoid novelty-induced stress. Again, time duration and frequency of handling must be adapted according to the strain used and,

1 cm

1 cm

Figure 1 | A photograph of two typical objects (tower of Lego bricks and Falcon tissue culture flask filled with sand) to use in an object recognition protocol. These specific objects were used in the example results presented in the ANTICIPATED RESULTS.

protocol otably, its level of anxiety. In addition, mice should be placed in n the testing room at least 30 min before testing.


Exploratory behavior toward objects. The most efficient, preferred method for assessing exploratory behavior is still a matter of debate13 and no consensus has yet been reached. In the original paper3, exploration of an object was defined as follows: ‘directing the nose’ toward the object ‘at a distance less than or equal to 2 cm.’ However, a minimal proximity of 4 cm has also been reported30. A point of agreement for the majority of investigators is that climbing over or leaning on an object is not considered to be an explorative behavior, unless that action was accompanied with a nose-directing behavior toward the object14,31. Here we chose to score object exploration whenever the mouse sniffed the object or touched the object while looking at it (i.e., when the distance between the nose and the object was less than 2 cm). Climbing onto the object (unless the mouse sniffs the object it has climbed on) or chewing the object does not qualify as exploration. Data collection. Exploratory behavior can be assessed either by an experimenter or by an automated video-tracking system. Because of several inefficiencies of manual scoring (time-consuming factors, necessity of extensive training and interobserver variability), automatic systems have emerged to track animal behavior. Such videotracking systems automatically score time spent in a defined zone

around the object. As long as the acquisition parameters are not changed, these systems will provide reliable results for any animals in all of the different groups or even different studies. However, such systems may also flaw results in ways that must be taken into consideration. For example, in addition to the narrowing of choices of objects, the results obtained are strongly related to the definition of the targeted zone and may thus differ between laboratories. An additional potential variability may also appear in the frame of the video camera chosen or in between systems (with respect to the animal body part that will be tracked). Thus, a system that counts exploration time whenever any animal body part is in the target zone will diminish the experimental efficiency. In other words, detection time in the zone will increase, even though the animal is merely climbing onto the object or simply grooming in the zone but not exploring the object. For these reasons, acknowledging that rodents explore primarily by sniffing, video-tracking systems based on nose-point detection are more reliable than packages based on the center point of the animal32. At any rate, when acquiring such a system, it is important to compare results obtained at least once with a series of experiments carried out by an expert observer of the laboratory. The comparison via a Pearson’s correlation suggested by Benice and Raber32 is a good way to ensure that the automated scoring is reliable. Finally, even though the price of such software may appear relatively high—at least for the one supplied by Noldus—it has been proven to be efficient32.

MATERIALS REAGENTS • Mice (see Reagent Setup) ! CAUTION Experiments involving rodents must conform to all relevant institutional and governmental regulations. • Use an appropriate solution (odorless soap, isopropyl or ethanol solution) to clean the objects and experimental apparatus. A solution of 70% (vol/vol) ethanol was used in our experiments. EQUIPMENT • Open-field apparatus: a black wooden box measuring 33 × 33 × 20 cm constitutes our open field • Two different objects in triplicate: the two first copies of each object are randomly used for the familiarization session and the third copies are used for the test session.  CRITICAL Further information about the features of the objects can be found in the INTRODUCTION. • Use paper tissue to clean the objects and the apparatus in between uses • Light meter (Luxmeter HI97500; Hanna Instruments) REAGENT SETUP Mice Variables related to animal subjects that need to be considered for the object recognition test are the following: strain21,24,33 (exploratory behaviors between strains vary and must be considered, particularly the background strain when using transgenic models), sex16 and estrous cycle24. In the experiments performed to obtain the data shown in the ANTICIPATED RESULTS, NMRI male mice (8–10 weeks of age, n = 144) were obtained from the Centre d’Elevage Roger Janvier (France) 1 week before the beginning of the experiments to ensure acclimation to the housing room of the animal facility. They were housed in standard cages (n = 12 per cage) containing a sheet of paper tissue for nesting behavior and a cardboard tube. The housing room was maintained at a constant temperature (22 ± 1 °C) and humidity (55%) and was under a 12-h reversed light-dark cycle (light 20:00–8:00). Water and food were available ad libitum. ! CAUTION All experiments must be carried out in accordance with the relevant guidelines and regulations regarding the care and the use of animals for the experimental procedures. EQUIPMENT SETUP Experiment room Place spatial cues (i.e., posters with contrasting colors) on the walls of the room. Provide dim light, for example, by using four wall lamps

directed to the ceiling. Ideally, you should have a panel control to allow fine adjustment of luminosity. Place the open field in the center of the room. Light meter Adjust the luminosity of the room to obtain 15 lux in the center of the open field.  CRITICAL To avoid any stress from bright lights, lighting should be diffused throughout the experimental room and adjusted to ensure the well-being of the mice, and it should provide sufficient illumination for observation by the experimenter. If albino animals are used, light intensity must be adapted to their high sensitivity to light. Moreover, according to the time of the day, the light-dark cycle of the animal facility has to be taken into account. Light intensity adjustments can be difficult for some video-tracking systems. Data collection system Several different video-tracking systems that include nose-point detection are now available (for instance, Noldus, Viewpoint, TSE Systems). These software packages require a computer interface and a video camera located vertically to the arena (most often sold with the necessary software). The other way to collect data is manually, by an experimenter who should be blinded to the treatment group. This can be done either directly during the experiment (better through a mirror placed above the open field to minimize interference with animal behavior) or afterward, if a videorecording has been made by a video camera placed above the arena. It is worth noting that a video recording allows for video analysis by more than one experimenter. Other equipment needed includes a one-channel and a two-channel timer (one channel for each object).  CRITICAL The experimenter should be at least 1 m from the open field so that the mouse cannot see him or her directly. The mirror must be placed above the open field to allow the experimenter to collect data without disturbing the mouse. In a similar way, for all the phases of the experiment, the experimenter should always be the same. Manual analysis of object recognition performance is highly vulnerable to interobserver differences. To obtain valid and reproducible data, the experimenter should have experience in identifying an actual exploration of an object by a mouse. If a single experimenter cannot do the whole experiment, it is important to ensure that all experimenters identify an exploratory behavior according to the same criteria.

nature protocols | VOL.8 NO.12 | 2013 | 2533


protocol PROCEDURE Habituation ● TIMING ~5 min 1| The major difference between the three procedural variations presented here is in the amount of habituation. If you are performing long habituation, follow option A. If you are performing short habituation, follow option B. If you are performing no ‘habituation’ (option C), then proceed directly to Step 2. (A) Long habituation (repeat on 3 consecutive days) (i) Place the mouse in the empty open field, facing the wall that is nearest to the experimenter, and allow it to explore the open field for 5 min. (ii) Return the mouse to its home cage. (iii) Clean the open field with 70% (vol/vol) ethanol to minimize olfactory cues. (iv) Repeat Step 1A(i–iii) 6 h later. (B) Short habituation (i) Place the mouse in the empty open field, facing the wall that is nearest to the experimenter, and allow it to explore the open field for 5 min. (ii) Return the mouse to its home cage. (iii) Clean the open field with 70% (vol/vol) ethanol to minimize olfactory cues. (C) No habituation (i) Skip directly to Step 2. Familiarization session ● TIMING