The results of the present study demonstrate developmental alterations in various measures of fear conditioning, anxiety-like behavior and depression-related behavior across the juvenile-to-adult period in C57BL/6J mice.
Four-week old, early adolescent mice showed more rapid and more robust acquisition of a conditioned association between an auditory stimulus (conditioned stimulus, CS) and footshock as compared to older mice. These young mice showed higher freezing than adults by the second presentation of the CS, demonstrative of augmented one-trial learning.
Enhanced fear conditioning in early adolescent mice was further evidenced by higher freezing during the last of four CS-shock presentations. Importantly, these differences did not appear to be due to increased sensitivity to the pain of the footshocks in the early adolescents, as reactions to footshock was actually lesser in these mice as compared to the older groups. Enhanced fear acquisition during conditioning was paralleled by relatively greater recall and expression of the fear response when early adolescent mice were again presented with the CS twenty-four hours later in a novel environment. Moreover, while mice of all age groups showed a significant reduction in freezing to the CS with repeated tone presentations (i.e., within-session extinction), freezing was generally higher throughout the session in the early adolescents. Thus, age-related differences in fear conditioning were not associated with differences in rates of within-session extinction learning: the magnitude of the reduction in freezing from start to the end of the 50-trial extinction session was no different between the adolescents and adults. Although peri-adolescent mice exhibited normal freezing during conditioning, overall freezing during the recall and extinction session was modestly but statistically higher in this age-group than in adults. Taken together, these data suggest that adolescent mice, particularly during early adolescence, have a greater propensity to acquire and/or express conditioned fear responses than adults, and extend previous work showing that various conditioned fear responses, including freezing, develop in early adolescence in rats (reviewed in [6
]). More generally, these findings potentially speak to view that human adolescence is a period of vulnerability to anxiety disorders characterized by abnormalities in emotional memory, such as posttraumatic memory (PTSD) [4
]. Although beyond the scope of the current study, elucidating the neural substrates of enhanced fear conditioning in adolescent mice could provide insight into the pathophysiology of adolescent PTSD [8
]. For example, examining the development of the amygdala and prefrontal cortex across mouse adolescence would be of significant interest [26
Interestingly, ontogenic shifts in emotional learning and memory were dissociated from changes in unconditioned anxiety-like behaviors across adolescence. Exploration of the aversive, open arms of the elevated plus-maze was similar in adolescents and adults; while exploration of the aversive center of the novel open field was lower in peri- but not early adolescents than adults. These data concur with the finding that 4-week old Swiss Webster mice were no different from 8-week olds on the elevated plus-maze at baseline [48
] (see also [21
]), but not with another report that 7-week, but not 5-week old CD-1 mice showed lesser anxiety-like behavior in this test than adults [34
], or with data variously reporting either relatively increased or decreased anxiety-like behavior in adolescents rats in both the elevated plus-maze, light/dark exploration test and the social interaction test [9
]. Such differences may be due to methodological differences between studies. Another potentially critical variable influencing these behaviors and other is genetic background of mice tested. Mouse strains are well known to differ on various measures anxiety- and depression-related behavior and stress reactivity (for recent review, see [22
]). Recent findings also demonstrate that adolescent mice of different strains may vary in their anxiety-related responses to certain environmental manipulations. For example, A/J mice housed with C57BL/6J mice during adolescence subsequently exhibited lower levels of anxiety-like behavior in adulthood as compared to A/J mice housed with conspecifics from the same strain while, in contrast, anxiety-like behaviors in C57BL/6J mice were unaffected by social housing conditions [22
]. These data suggest that the ontogeny of anxiety- and possibly other ‘emotion-related’ behaviors may vary across genetic backgrounds, and that a comparison of adolescents versus adults in different mouse strains, both inbred and outbred, may yield valuable insights.
While indices of anxiety-like behavior were largely similar across age-groups in our study, early adolescent mice did show lower overall levels of locomotor activity than adults in both the elevated plus-maze and novel open field, and peri-adolescents exhibited lower locomotion in the novel open field only, as compared to adults. Further analysis indicated that these differences were explained by the young mice traveling a shorter distance within the safe regions of the apparatus (i.e., closed arms of the elevated plus-maze, periphery of novel open field). Similarly, relatively lesser activity in the open field and light/dark exploration tests has been reported in, respectively, 5-week old CD-1 mice [2
] and 4-week old Swiss Webster mice [21
], as compared to adult counterparts. However, other studies have found relatively greater novel environment locomotor exploration in adolescent CD-1 mice [1
], while studies in rats report both normal and greater open field locomotion in adolescents relative to adults [14
]. Some authors have proposed that differences in locomotor exploration between adolescents and adults are related to the extent that novelty drives behavior in a given test situation and this factor may account for the apparent discrepancies across studies [30
]. Notwithstanding, levels of locomotor activity in adolescent mice in the current study are unlikely to account for the relatively higher levels of active swimming behavior exhibited by adolescent mice in the forced swim test for depression-related behavior. Thus, early adolescents showed significantly less immobility than adults across three daily exposures to the test. In contrast to the step-wise age-dependent differences in fear extinction and locomotor activity (with peri-adolescents being intermediate), differences in the forced swim test were restricted to the early adolescent mice. Relatively lesser forced swim test immobility in early adolescents mimics the effects of antidepressant treatment typically seen in adult rats and mice [11
] and as such can be interpreted as an reduced depression-like profile.
To our knowledge there are no other studies assessing depression-related behaviors in adolescent rodents. However, pertinent to present findings are data describing altered behavioral, neural and neuroendocrine responses to stress in adolescent rodents. For example, using c-fos immunoreactivity as an index of neuronal activation, Kellogg and colleagues showed that restraint stress produced less activation of certain brain regions mediating stress-reactivity, including parts of the amygdala and medial PFC, in 4-week old rats as compared to adults [27
]. An attenuated stress-response in adolescent animals is consistent with the finding that corticosterone responses to forced novelty exposure were lesser in 5-week old CD-1 mice than adults [2
], but not with another showing that adolescent Swiss Webster mice or various rat strains were more sensitive to social and restraint stress as measured by neuroendocrine release, body weight reductions and/or increased elevated plus-maze, light/dark exploration test or open field anxiety-like behavior [10
]. Again these data emphasize that task, species and strain factors likely play a major role in determining anxiety- and stress-related behavior in adolescent mice and highlight the need for further studies in this area of research.
In summary, the results of the present study demonstrate that the developmental transition from juvenility through adulthood in mice sees by pronounced changes in fear- and stress-related behaviors in the C57BL/6J inbred strain. Mice at the early adolescent stage acquired and expressed conditioned fear responses to a greater degree as compared to adults. In contrast, unconditioned anxiety-related behaviors in the elevated plus-maze and novel open field did not differ between adolescents and adults, while depression-related responses in the forced swim test were lower in early adolescent mice than adults. Thus, differences between adolescents and adults appear complex and specific to certain behavioral domains, at least in this strain of mice, and cannot be explained by a general ‘deficiency’ in fear/anxiety regulation and stress-reactivity in the younger animals. Rather, the distinct behavioral profile of adolescent mice likely reflects evolutionary shaped strategies for meeting the unique environmental demands of this phase of life. Elucidating the neural basis of these behaviors could shed light on the pathogenesis and improved treatment of neuropsychiatric disorders such as anxiety and depression that often have their origins in adolescence.