Trpm1 −/− mice show significantly high daily locomotor activity.
We performed a battery of more than 20 behavioral tests (Table 1). There was almost no significant difference for general physical characteristics, such as body weight, body temperature, grip strength, and motor coordination between Trpm1−/− and wild type mice (Fig. S1A-K). Trpm1−/− did not show depression-like behaviors in Porsolt forced swim test and tail suspension test (Fig. S1L-N). Intriguingly, Trpm1−/− mice showed significantly high daily locomotor activity (Fig. 1A) in contrast to mGluR6−/− mice [19]. It has been reported that the absence of vision in human and animals enhances auditory, haptic, and pain sensitivities [20–29], and causes structural changes in the visually deprived cortex and in other areas [23, 30, 31]. We examined sensory responses in Trpm1−/− mice, but we did not find any significance difference between Trpm1−/− mice and wild type for the hot plate test, the acoustic startle response, and prepulse inhibition (Fig. 1B-D). Although Trpm1−/− mice lack a functional ON visual pathway and thus have limited light responses, Trpm1−/− mice showed hyper- locomotor activity but did not show sensory hypersensitivities that was reported in other visually impaired animals.
Hyperactivity and reduction of anxiety-like behavior in Trpm1−/− mice
To assess anxiety-like behavior, we performed the light/dark transition test, the open-field test, and the elevated plus maze test (Fig. 2). In the light/dark transition test, distance traveled in the light and dark chamber were significantly increased in Trpm1−/− mice suggesting a reduction in anxiety-like behavior (Fig. 2A). However, the defect in the ON visual pathway may have affected the longer stay time in light and increased transition time and shorter latency to light for tests started at dark (Fig. 2B-D). In the open field test, which measures voluntary locomotor activity in a novel environment, Trpm1−/− mice exhibited a significant increase in total distance, vertical activity, center time and stereotypic behavior relative to WT mice (Fig. 2E-H), suggesting strong hyperactivity, which also explains longer distance traveled in the light/dark transition test. To implicate the hyperactivity in Trpm1−/− mice with ADHD, we performed the open field test after administration of MPH, which is often used as a first choice for treatment of ADHD in human (Fig. 2I) [32]. 120 min after administration of MPH, both WT and Trpm1−/− mice showed prominent hyperactivity, especially in mice which were injected at 10 mg/kg. As a result, MPH administration did not support the idea that the ADHD behavior displayed in Trpm1−/− mice can be reduced by MPH administration [33].
Additionally, in the elevated plus maze test, Trpm1−/− mice exhibited significantly increased numbers of entries and longer traveled distance, which are also explained by hyperactivity (Fig. 2J, M). Although visually impaired, Trpm1−/− mice did not show differences for entries to open arms, but stayed longer time in open arms suggesting a reduction of anxiety-like behavior (Fig. 2K, L).
To examine what causes a reduction in anxiety-like behavior in Trpm1−/− mice, we examined serum corticosterone levels in Trpm1−/− mice by ELISA [34, 35]. The idea is that an authentic reduction in anxiety should be correlated with an decrease in serum corticosterone levels, while a reduction in anxiety-like behavior in the absence of a decrease may have some other cause. The serum levels of corticosterone were not significantly different between Trpm1−/− mice and WT mice (Fig. S1O). This observations suggests that the hyperactivity of Trpm1−/− mice simulates a reduction of anxiety in our tests.
Abnormal social interaction in Trpm1−/− mice
Four kinds of social interaction tests (novel environment, sociability, novelty preference, and home cage test) were performed to evaluate social behaviors in the Trpm1−/− mice (Fig. 3). The novel environment test revealed significant differences between Trpm1−/− and WT mice, including a shorter duration per contact, increased contact number, and total traveled distances, which may be explained by the hyperactivity in Trpm1−/− mice (Fig. 3A, D, E). Although total duration of contact was shortened, but not significantly, active contacts were longer in Trpm1−/− mice (Fig. 3B, C). Both Crawley’s sociability and social novelty preference test and the test in the home cage did not reveal significant differences between WT and mutant mice (Fig. 3F-M, S1P). These results suggest that social interactions were attenuated in Trpm1−/− mice, which may be partially explained by hyperlocomotion behavior.
Attenuation of fear and spatial memories in Trpm1−/− mice
The contextual and cued fear conditioning test is used to assess fear memory (Fig. 4). In the conditioning phase, Trpm1−/− mice showed a lower level of freezing and traveled longer distances during sessions (Fig. 4A, B). The mutant mice traveled longer immediately after foot shock, an index of pain sensitivity (Fig. 4C). 24 hours after conditioning, Trpm1−/− mice showed decreased freezing and increased distance traveled. Similar significant differences were observed in tests 28 days after conditioning (Fig. 4D, E). These data suggest that attenuation of fear memory in Trpm1−/− mice could be related to hyperactivity.
We performed the Barnes maze test to determine whether the deficit in fear memory of Trpm1−/− contributes to hyper-locomotion or results from a deficit of memory. In both training sessions and reversal task tests, the distance to the escape box (Fig. 4F) and the number of errors to reach the escape box were significantly higher in Trpm1−/− mice (Fig. 4G), but latency to first reach the escape box was equivalent or shorter in Trpm1−/− mice than in WT mice (Fig. 4H), which may be related to hyper- locomotion activity. The probe tests were performed 24 hours and 1 month after the final training sessions. In these tests, Trpm1−/− and WT mice exhibited a significant effect of hole location target against the rest holes: 24 hour, WT p < 0.0001, Trpm1−/− p < 0.0001; 1 month, WT p < 0.0001, Trpm1−/− p < 0.0001; one-way ANOVA followed by Dunnett’s multiple comparison test), indicating that both genotypes were able to distinguish the location of the target. Time spent around the correct hole did not differ significantly between both genotypes at 24 hours after training, but was significantly shorter in Trpm1−/− mice 1 month later (Fig. 4I, J). These results suggest that Trpm1−/− mice have a deficit in long-term memory. In the reversal probe test, although both genotypes distinguished the location of the target (WT p < 0.0001, Trpm1−/− p < 0.0001; one-way ANOVA followed by Dunnett’s multiple comparison test), there was no significant difference in time spent around the correct hole between both genotypes (Fig. 4K). This result indicates that there was no deficit in behavioral flexibility in Trpm1−/− mice. We also performed a T-maze test to examine the working memory of Trpm1−/− mice. Although the shorter latency and longer distance traveled in Trpm1−/− mice were both significantly different, the correct responses at each trial were not (Fig. 4L-N). Taken together, Trpm1−/− mice showed attenuation for fear and long term memories, but no obvious deviation for flexibility and working memory.
Abnormal Structural and biochemical changes in the brains of Trpm1−/− mice
We have shown differences in the behavioral phenotype in Trpm1−/− mice relative to WT. However, Trpm1 functions predominantly as a component of the retinal ON bipolar transduction cascade and its expression is quite minor in the brain. To determine whether there are central structural changes, we compared brain regions between Trpm1−/− and WT mice. The cerebral cortex, olfactory bulb, and pons and medulla of Trpm1−/− mice were significantly heavier than in WT mice at 1 month old (Fig. 5A). In addition, the cerebral cortex, hippocampus, midbrain and cerebellum of Trpm1−/− mice were significantly heavier than those of WT mice at 4 months old (Fig. 5B).
We detected a subtle expression of Trpm1 mRNA throughout the WT mouse brain with the exception of the cerebellum (Fig. 5C). We also quantified levels of biogenic monoamines ex vivo, including dopamine (DA), noradrenaline (NAd), serotonin (5-HT), and their major metabolites using HPLC-ECD in several adult brain regions. Levels of DA, NA, and NM were significantly decreased in the cerebellum (Fig. 5D-F). There was no significant change in the levels of the other monoamines and their metabolites in any other brain region (Fig. S2). The decreased levels of monoamines in the cerebellum of Trpm1−/− mice could influence hyper-locomotory activity, locomotion is regulated by the cerebellum. The lack of overlap between the Trpm1 expression pattern and the change of monoamine distribution in the brain is consistent with the idea that Trpm1 expresses in monoaminergic neurons that project to the cerebellum.