Male ICR mice (8 weeks old) were purchased from Sankyo Labo Service Corporation Inc. (Tokyo, Japan). In the present study we avoided to use female animals because estrous cycles might cause data variability. The animals were housed individually in cages and provided with food and water ad libitum at 24˚C under a 12-h light/12-h dark cycle. After 1 to 2 weeks of acclimation, the animals were subjected to behavioral experiments during the latter half of the 12-h light period. All experimental procedures and animal housing were approved by the Institutional Animal Care and Use Committee of Sophia University and comply with the criteria established by the National Institutes of Health Guide for the Care and Use of Laboratory Animals (1996).

NOR test was conducted in an open-field polypropylene arena (40 cm width × 30 cm depth × 30 cm height). The floor the arena was divided into 20 equal sections by drawing grid lines. The NOR test comprised three phases: habituation, acquisition and test. During the habituation phase, the animals were allowed to freely explore the empty arena for 5 minutes per day for 3 days. In the acquisition phase, mice were subjected to a single 5-min acquisition trial, where they were allowed to freely explore two identical objects which were symmetrically placed in the arena at a distance of 5 cm from the wall and 15 cm from each other. After a predetermined retention interval (inter-trial interval, ITI) during which animals were kept in their home cages, test phase was initiated. In this phase, mice were subjected to a 5-min test trial which was similar to the acquisition trial, with the exception that one of the familiar objects was replaced with a novel object at the same location. Objects were made of ceramic, which were too heavy to be displaced by animals. Object novelty and location were counterbalanced within experimental groups to eliminate potential biases caused by preference for particular objects or locations. Mice were placed in the middle of the two objects facing the wall at the beginning of each acquisition and test trial, and were recorded using a video camera mounted above the arena. After each trial, the objects were thoroughly cleaned with 70% EtOH to minimize the presence of olfactory trails.

The time spent exploring the objects was quantified by a blinded trained observer using the recordings. Exploratory behavior was defined as the animal directing its nose toward the object at a distance of <1 cm. Animals were excluded from data analyses if they met the following criteria: 1) spent 10 seconds or less exploring objects during any acquisition trial; 2) spent 10 seconds or less exploring objects during the test trial; 3) did not explore both objects during the training and test phases. The discrimination index (DI) was calculated as follows: percent of time spent exploring the novel object divided by the total time spent exploring both objects during the test trial. Object recognition was considered as DIs those were significantly above chance performance (50%).

Off-target drug effects were quantified using spontaneous locomotor activity, defined as the number of times mice crossed the grid lines with all four paws (grid-crossings), and the total time spent exploring both objects during each phase.

MEL (Mw: 232.28), RAM (Mw: 259.34), AMK (Mw: 236.27) and luzindole were purchased from Sigma-Aldrich (Tokyo, Japan). MEL, RAM, and AMK were dissolved in dimethyl sulfoxide (DMSO) and then diluted with saline to a final concentration of (1% DMSO) for intraperitoneal administration. The dosage of MEL, RAM, and AMK varied from experiment to experiment. Luzindole was also dissolved in DMSO and then diluted with saline to a final DMSO concentration of 1% for the dosage of 0.1 mg/kg b.w. (Experiment 3) and 10% for the dosage of 1 mg/kg b.w. (Experiment 4). These doses of luzindole were set to more than 1000 times of MEL dose so that the activation of MT1 and MT2 receptors by MEL can be fully antagonized. All drugs were injected intraperitoneally (i.p.) at a volume of 3 ml/kg b.w.

The present study was composed of the following four experiments.

Experiment 1

Untreated mice were tested for their ability to retain STM using NOR task with a 2-h (n = 14) or 3-h ITI (n = 10).

Experiment 2

Mice (n = 7 - 13 per group) were used to investigate the dose-dependent effects of MEL, RAM, and AMK administered immediately after acquisition trial on their STM performance tested after a 3-h ITI.

Experiment 3

Mice (n = 7 - 11 per group) were used to examine if the STM facilitating effects of MEL, RAM, and AMK are inhibited by luzindole. The animals were pretreated with either luzindole or vehicle immediately after acquisition trial and each pretreatment was followed 5 min later by administration of varying doses of either MEL, RAM, or AMK. STM performance of the mice was tested after a 3-h ITI.

Experiment 4

Mice (n = 7 - 13 per group) were used to examine the dose-dependent effects of post training MEL, RAM and AMK on LTM performance, and to examine if LTM formation-promoting effects of these drugs were inhibited by pretreatment with luzindole. The animals were pretreated with either luzindole or vehicle immediately after acquisition trial and each pretreatment was followed 5 min later by administration of varying doses of either MEL, RAM, or AMK. LTM performance of the mice was tested after a 24-h ITI.

For all test trials, one-sample t-tests were used to analyze whether DIs significantly differed from the chance level (50%). Effect size for the one-sample t-tests was given by Cohen’s d. One-way analysis of variance (ANOVA) was used to analyze the differences in the total exploration time and the number of grid-crossings among groups administered with varying doses of either MEL, RAM or AMK, including 0 ng (or μg) /kg (vehicle). Post hoc Dunnett’s tests were applied where appropriate. Effect size estimates for the one-way ANOVA was computed by partial-Eta squared ( η p 2 ). A p value < 0.05 was considered statistically significant.

STM performance in untreated mice

One-sample t-test revealed that the DIs from the untreated mice were significantly greater than the chance level (50%) when tested with a 2-h ITI (t (13) = 2.92, p = 0.0118, d = 1.15), but not with a 3-h ITI (t (9) = 0.19, p = 0.8516, d = 0.09) (Figure 1). These data indicate that the mice were able to retain STM for at least 2 hours, but not for 3 hours or longer in the present experimental conditions.

Dose-dependent effects of a single administration of drugs on STM performance after a 3-h ITI

One-sample t-test revealed that MEL, RAM, and AMK, given at lower doses immediately after the acquisition trial, did not affect DI (p > 0.05) after a 3-h ITI (0 ng/kg (vehicle): t (6) = 0.05, p = 0.9611, d = 0.03; 0.01 and 0.1 ng/kg of MEL: t (9) = 1.25, p = 0.2437, d = 0.59 and t (9) = 0.45, p = 0.6661, d = 0.21, respectively; 0.01 ng/kg of RAM: t (9) = 0.13, p = 0.8987, d = 0.06; 0.01 ng/kg of AMK: t (12) = 0.70, p = 0.5002, d = 0.28). However, higher doses of these drugs all significantly (p < 0.05) increased DI compared to chance level (50%), indicating facilitated STM (Figure 2(a)) (1 and 10 ng/kg of MEL: t (8) = 2.36, p = 0.0458, d = 1.18 and t (7) = 3.66, p = 0.0081, d = 1.96, respectively; 0.1, 1 and 10 ng/kg of

RAM: t (9) = 2.27, p = 0.0497, d = 1.07, t (8) = 3.47, p = 0.0084, d = 1.74, and t (9) = 3.36, p = 0.0084, d = 1.58, respectively; 0.1, 1 and 10 ng/kg of AMK: t (11) = 3.35, p = 0.0065, d = 1.43, t (9) = 2.43, p = 0.0378, d = 1.15, and t (9) = 2.46, p = 0.0364, d = 1.16, respectively). The minimum doses required to facilitate STM were 1 ng /kg for MEL, and 0.1 ng /kg for RAM and AMK.

One-way ANOVA revealed that there were no significant differences (p > 0.05) in either total exploration time (Figure 2(b)) (F (12, 115)  =  1.24, p = 0.2630, η p 2 = 0.11), or number of grid-crossings (Figure 2(c)) (F (12, 115)  =  1.73, p = 0.0696, η p 2 = 0.15) among groups received varying doses of MEL, RAM, or AMK, including 0 ng/kg (vehicle). These observations indicate that at the doses used in this experiment, these drugs did not significantly affect the factors required for exploratory behavior, such as motivation and motor function.

Effect of luzindole pretreatment on STM performance in mice received varying doses of either MEL, RAM, or AMK effective to facilitate STM.

In this experiment, the mice were pretreated with either luzindole (0.1 mg/kg) (n = 7 - 11 per group) or its vehicle (1% DMSO saline) (n = 7 - 10 per group) immediately after the acquisition trial, and each pretreatment was followed 5 min later by a single administration of either MEL, RAM, or AMK at two different doses, 10 and 100 ng/kg. Here, we selected these doses as those that would facilitate STM, based on the results of Experiment 2. As expected, in the vehicle-pretreated control mice (Figure 3(a)), one-sample t-test revealed that all drugs (MEL, RAM and AMK) significantly (p < 0.05) increased DIs compared with chance performance (50%) after a 3-h ITI, indicating facilitated STM (10 and 100 ng/kg of MEL: t (9) = 2.57, p = 0.0301, d = 1.21 and t (9) = 2.36, p = 0.0428, d = 1.11, respectively; 10 and 100 ng/kg of RAM: t (8) = 3.65, p = 0.0065, d = 1.83 and t (8) = 3.01, p = 0.0169, d = 1.50, respectively; 10 and 100 ng/kg of AMK: t (8) = 3.43, p = 0.0090, d = 1.71 and t (7) = 2.92, p = 0.0224, d = 1.56, respectively; cf. 0 ng/kg (vehicle): t (6) = 0.69, p = 0.5167, d = 0.40).

In contrast, as shown in Figure 3(b), pretreatment with luzindole readily suppressed the STM-facilitating effects produced by both 10 ng/kg and 100 ng/kg of RAM (t (10) = 0.78, p = 0.4530, d = 0.35 and t (8) = 0.03, p = 0.974, d = 0.02, respectively), but not those produced by either dose of AMK (10 and 100 ng/kg of AMK: t (9) = 2.36, p = 0.0425, d = 1.11 and t (9) = 2.50, p = 0.0340, d = 1.18, respectively; cf. 0 ng/kg (vehicle): t (6) = 0.15, p = 0.8879, d = 0.08). Luzindole also failed to suppress the STM-facilitating effects produced by 100 ng/kg of MEL (t (7) = 3.33, p = 0.0126, d = 1.78), although it could suppress these effects produced by10 ng/kg of MEL (t (7) = 0.76, p = 0.4781, d = 0.40).

For both vehicle- (Figure 3(c) and Figure 3(e)) and luzindole-pretreated mice (Figure 3(d) and Figure 3(f)), one-way ANOVA revealed that there were no significant differences (p > 0.05) in either total exploration time (Figure 3(c) and 3D) (F (6, 55)  =  0.86, p = 0.5273 η p 2 = 0.09 and F (6, 56)  =  0.25, p = 0.9565, η p 2 = 0.03, respectively) or number of grid-crossings (Figure 3(e)) and Figure 3(f))) (F (6, 55)  =  0.76, p = 0.6074, η p 2 = 0.08 and F (6, 56)  = 0.77, p = 0.5954, η p 2 = 0.10, respectively) among groups received varying doses of MEL, RAM, or AMK, including 0 ng/kg (vehicle).

Effects of pretreatment with luzindole on LTM performance in the mice received

varying doses of either MEL, RAM or AMK.

In this experiment, the mice were pretreated with either luzindole (1 mg/kg) (n = 7 - 13 per group) or its vehicle (10% DMSO saline) (n = 7 - 11 per group) immediately after the acquisition trial, and each pretreatment was followed 5 min later by a single administration of MEL, RAM and AMK. Test trials were performed after a 24-h ITI to assess LTM performance. First, we investigated the doses of MEL, RAM and AMK required to promote LTM formation using the control mice pretreated with vehicle. We used here varying doses of MEL (0.1 or 1 μg/kg), RAM (0.01, 0.1, or 1 μg/kg) and AMK (0.001, 0.01, 0.1 or 1 μg/kg), speculating that promotion of LTM formation requires higher doses than STM facilitation. As shown in Figure 4(a), one-sample t-tests revealed that the minimum doses of MEL, RAM and AMK required to significantly (p < 0.05) promote LTM formation were 1 μg/kg, 0.1 μg/kg, and 0.01 μg/kg, respectively (0.1 and 1 ug/kg of MEL: t (7) = 0.37, p = 0.7228, d = 0.20 and t (7) = 2.79, p = 0.0271, d = 1.49, respectively; 0.01, 0.1 and 1 ug/kg of RAM: t (9) = 1.31, p = 0.2217, d = 0.62, t (10) = 2.51, p = 0.0308, d = 1.12 and t (8) = 2.41, p = 0.0422, d = 1.21, respectively; 0.001, 0.01, 0.1 and 1 μg/kg of AMK: t (9) = 0.76, p = 0.4670, d = 0.36, t (10) = 2.38, p = 0.0383, d = 1.07, t (6) = 2.68, p = 0.0366, d = 1.55 and t (8) = 2.83, p = 0.0221, d = 1.41, respectively; cf. 0 μg/kg (vehicle): t (7) = 0.60, p = 0.5703, d = 0.32).

Based on these results, we then selected the following doses of MEL, RAM, and AMK from a range that included the doses effective in promoting LTM formation: 0.1 and 1 μg/kg for MEL and RAM, and 0.01, 0.1, and 1 μg/kg for AMK. We examined the effects of luzindole pretreatment on LTM performance of the mice received these doses of drugs. As shown in Figure 4(b), one-sample t-tests revealed that pretreatment with luzindole readily suppressed the promoting effects of LTM formation produced by both 0.1 and 1 μg/kg of RAM (t (12) = 1.47, p = 0.1662, d = 0.60 and t (6) = 0.78, p = 0.4642, d = 0.45, respectively), but not those produced by any dose of AMK including 0.01 μg/kg (0.01, 0,1 and 1 ug/kg of AMK: t (10) = 2.32, p = 0.0430, d = 1.04, t (11) = 2.43, p = 0.0332, d = 1.04, and t (11) = 2.37, p = 0.0369, d = 1.01, respectively; cf. 0 ug/kg (vehicle) control: t (11) = 1.09, p = 0.2974, d = 0.47). Luzindole also failed to suppress the LTM formation-promoting effects produced by 1 μg/kg of MEL (t (8) = 2.65, p = 0.0293, d = 1.32).

One-way ANOVA revealed that there were no significant differences (p > 0.05) in either total exploration time or number of grid-crossings among vehicle-pretreated groups received varying doses of MEL, RAM, or AMK, including 0 μg/kg (vehicle) (Figure 4(c) and Figure 4(e)) (F (9, 81)  =  0.67, p = 0.7100, η p 2 = 0.07 and F (9, 81)  =  0.76, p = 0.6536, η p 2 = 0.07, respectively). For luzindole-pretreated mice, however, there were significant differences (p < 0.05) in both total exploration time and the number of grid-crossings (Figure 4(d) and Figure 4(f)) (F (7, 77)  =  2.39, p  = 0.0289, η p 2 = 0.18 and F (7, 77)  =  3.19, p = 0.0051, η p 2 = 0.23, respectively). Post hoc Dunnett tests showed no significant difference (p > 0.05) in both total exploration time and number of grid crossings between the groups received any dose of MEL, RAM, or AMK and the group receiving vehicle (0 μg/kg).