Differences between adult and adolescent male mice in approach/avoidance and expression of hippocampal NPY in response to acute footshock

Anxiety disorders are the most common neuropsychiatric disorders diagnosed in adolescence and adulthood. Stress can lead to an increase in anxiety-related behaviors, although the consequences of stress in rodents are typically investigated only in adults. The levels of Neuropeptide Y (NPY), a mediator of stress resilience, are reduced in adult patients with Post-Traumatic Stress Disorder. For rodents, footshock is a physical stressor that increases anxiety-like behavior and reduces NPY in adults, however, the effects in adolescents are unknown. Here we used a 30-min unpredictable footshock protocol to investigate the differences in behavior and stress-relevant molecules between adolescent (6 weeks) and adult (3 months) male C57Bl6/J mice. The protocol resulted in fear expression in both ages as observed by enhanced freezing during footshock and elevation in plasma corticosterone and NPY shortly after exposure. However, effects on approach/avoidance behavior were different between the two ages. One week after footshock exposure, adult mice showed reduced open arm time and entries on elevated plus maze (EPM), whereas adolescent mice showed no effect. Footshock mice in both age groups displayed reduced activity levels in EPM and open field. The hypolocomotion did not relate to motor deficits, as there were no differences between footshock and control groups using rotarod. Surprisingly, we found that the adolescent mice had elevated NPY peptide expression in hippocampus, whereas adults had reduced expression one week after footshock exposure. Together, these results demonstrate that stress differentially affects both behavior and the important stress resilience factor NPY in adolescents compared to adults.

Keywords: NPY; footshock; stress; mouse; behavior; corticosterone

Anxiety and trauma-related disorders are the most common family of neuropsychiatric disorders in both adolescents and adults (Kessler et al., [18]). Traumatic stress affects adolescents as well as adults, and can result in post-traumatic stress disorder in both ages (Merikangas et al., [27]). However, animal studies investigating the effects of traumatic stress are performed almost exclusively in adult rodents. Even the consequences of traumatic experience in early life are most commonly studied in adult rodents (Brydges et al., [3]; Husum & Mathé, [16]; Jacobson-Pick & Richter-Levin, [17]). Adolescence is a time of rapid physical, emotional, and cognitive changes (Raymond et al., [31]; Spear, [45], [46]) that can affect mental health (Paus et al., [29]; Sturman & Moghaddam, [50]; Romeo, [33]). It is, therefore, essential to understand the impact of traumatic stress on both adolescents and adults.

Neuropeptide Y (NPY) is a key neuromodulator that is regarded as a stress resilience factor (Reichmann & Holzer, [32]). NPY is the most abundant neuropeptide in the mammalian brain (Smith, [42]), which regulates anxiety, cognition, feeding behavior, arousal, and blood pressure (Hökfelt et al., [15]; Medeiros & Turner, [26]; Tasan et al., [52]; Zimanyi et al., [55]). Adult patients with post-traumatic stress disorder (PTSD) have lower NPY levels in cerebrospinal fluid (CSF) (Sah et al., [37], [36]; Sandberg et al., [38]). Importantly, a clinical trial concluded that intranasal NPY may be associated with anxiolytic effects in adults with PTSD (Sayed et al., [40]), and a follow up trial is ongoing. Whether NPY is affected in adolescent patients with PTSD is unknown. Interestingly, self-reported childhood trauma correlates with altered NPY in adults (Sher et al., [41]; Soleimani et al., [43]) indicating that early life trauma has lasting consequences on NPY. The normal functions of NPY during adolescence are only beginning to be discovered, and its therapeutic potential remains unexplored. It is, therefore, critical to determine the consequences of traumatic experience during adolescence on the NPY system.

A rodent traumatic stress protocol, predator scent stress (PSS), has been administered and the effects studied in both adolescent and adult rodents (Cohen et al., [7]; Dengler et al., [12]; Li et al., [23]; Post et al., [30]). In both age groups, PSS induces a robust increase in avoidance behavior on the elevated plus maze (EPM) (Cohen et al., [7]; Dengler et al., [12]; Li et al., [23]; Post et al., [30]), which is typically interpreted as increased anxiety-like behavior (Kraeuter et al., [21]). Additionally, increased avoidance behavior has been characterized as a core symptom of PTSD in humans (Substance Abuse and Mental Health Services Administration n.d.), which makes studying approach-avoidance behavior in rodents an ideal model to study aspects of the disorder. PSS also causes lasting effects on hippocampal NPY in adult and adolescent rodents (Cohen et al., [7]; Li et al., [23]; Post et al., [30]). It is not yet known whether these effects are specific to PSS, which relies on activating an innate fear of predators in rodents, or are generalizable to other traumatic stress protocols. Footshock is a physical stressor that increases avoidance behavior on the EPM in adult rodents (Louvart et al., [24]), as well as causing increased immobility on EPM and open field (OF) (Anderson et al., [1]; Chalmers et al., [4]; Louvart et al., [24]). Footshock applied during adolescence also increases avoidance behavior and immobility in adulthood, however, the effects on behavior and NPY at the adolescent age are undetermined.

Here we investigated the effects of footshock as an acute physical stressor on behavior and NPY in adolescent and adult male C57bl/6J mice. We determined that a robust footshock protocol heightens avoidance behavior on the EPM in adult mice, but not in adolescent mice. Footshock also causes a lasting hypoactive phenotype in both age groups. Additionally, adult mice display a lasting reduction in hippocampal NPY after footshock, whereas adolescent mice have an increase. Overall, these findings highlight that both avoidance behavior and hippocampal NPY are differentially affected by traumatic experience depending on age.

All experiments were conducted in accordance to the Guide for the Care and Use of Laboratory Animals adopted by the National Institutes of Health. All experimental protocols conducted were approved by the Institutional Animal Care and Use Committee at the University of Alabama at Birmingham (APN 20119).

C57BL/6J Mice (Stock no. 000664) from Jackson Laboratories and housed in groups of 5 until 2 days prior to footshock administration when they were individually housed. All animals were handled for a minimum of 3 days prior to the behavioral tasks. Mouse colonies were maintained at 21 ± 2 °C with food and water ad libitum on a 12 h light/dark cycle (6:00 am–6:00 pm). All mice used for this study were males that were 5–6 weeks old (adolescent) at the date of footshock or between 3 and 4 months of age (adults). Adult mice were directly purchased from Jackson lab and allowed to habituate to the housing room for 1 week prior to individually housing. Adolescent mice were bred in house from purchased C57Bl/6 J breeders in order to be able to accurately monitor age.

Footshock exposure occurred between the hours of 8 am and 12 pm, during the animal's light cycle. Mice were placed into a 30.5 cm × 24.1 cm × 21 cm box with a clear plexiglass front and an electrifiable grid floor. Subjects were allowed to freely explore for 2 min. Test subjects were then subjected to 15 footshocks (1 mA, 2 s duration) over the course of 26 min. Footshock times were determined using a random number generator (106.1 ± 85.4 s, minimum: 6 s, maximum: 255 s). Mice were then allowed a 2-min rest period before being removed from the box. Control mice were allowed to freely explore the box for 30 min. Time spent freezing was recorded throughout the exposure. Freezing behavior was compared between habituation period (2 min before start of footshock) and recovery period (2 min after the end of the footshocks). Six total cohorts with at least two animals per condition were used to obtain the adult data, with a total of 41 mice receiving footshock and 40 mice as controls. Two adult mice were excluded from the adult footshock freezing behavior data due to loss of raw video files. Eight total cohorts were used with at least 2 mice per condition in each cohort for the adolescent data, with a total of 35 mice receiving footshock and 34 mice as controls. Three adolescent mice were excluded from the freezing behavior calculations due to loss of raw video files.

Approach/avoidance behavior was assessed using the EPM 8 days after footshock exposure. EPM uses a cross plexiglass platform raised one meter above the floor with two closed arms, two open arms, and a central hub. Experiments were conducted in low light (∼9 lux) and white noise (∼64.8 dB) were maintained throughout the test. At the beginning of each experiment, the mouse was placed in the central hub, facing an open arm. Mice were allowed to explore the maze for 5 min with various parameters (including time spent and entrances into each arm) tracked via automated software (Med Associates, St. Albans, VT). Any animal that stayed in any of the arms without any movement for more than 30% of the time was excluded. Five adult footshock mice were excluded based on this criteria. Four adolescent footshock mice and one control adolescent mice were excluded based on this criterion. One adolescent footshock mouse jumped off the EPM and was therefore excluded. Additional criteria for exclusion were environmental conditions, such as cage flooding (one control adult animal and one control adolescent animal excluded) or impairments with the automated tracking software (one control adult animal and one control adolescent animal excluded). EPM exposure occurred during the animal's light phase, between the hours of 8 am and 12 pm.

The OF test (OFT) was used as to measure locomotor activity 8 days after footshock exposure, and 1 h after completion of EPM. The OFT apparatus is a square (27.9 cm3) box with plexiglass sides consisting of 48 infrared beams and tracking software (Med Associates, St. Albans, VT). All OFT experiments were conducted in a lighted room (∼140 lux) with a white noise machine (∼73 dB). At the beginning of each experiments, each mouse was placed in the same corner of the box. Total ambulatory time and distance were measured in the box for 5 min, and percent time in the center region was calculated. Exclusion criteria for the OF were cage flooding and impairments of the automated tracking software. Two adult control animals were excluded due to these criteria. No animals in the adolescent group were excluded in the OF. OF exposure occurred during the animals' light phase, between the hours of 11 am and 2 pm.

The rotarod task can be used to evaluate motor coordination and motor function in the mice. Here we used rotorod to see if there was motor impairment due to possible injury to the paws caused by the footshock protocol. If this had occurred, it would likely be evident at the 24 h time point. The rotarod apparatus consists of a walking wheel with variable speed settings. The protocol used was set to increase in speed progressively from 4 rotations per minute (rpm) to a maximum of 40 rpm. Animals were placed on the rotarod and the time at which they fall off the apparatus was recorded as well as the speed at which the animal falls. A separate cohort of adolescent animals (8 per group) was used only to gather rotorod data 24 h after footshock. A subset of adult animals was used to gather rotorod data 24 h after footshock and subsequently used for EPM and OF (8 of 12 controls and 9 of 13 footshock). Rotorod experiments were performed during the animals' light phase between the hours of 8 am and 12 pm.

Neuropeptide Y and corticosterone levels were measured using commercially available ELISA kits (EMD Millipore, Darmstadt, Germany, EZHNPY-25K and Enzo, ADI-901-097, respectively). Separate plates were run for adult and adolescent samples. Multiple cohorts were included for each age, each cohort contained at least two samples per condition. ELISA was run in duplicate for each sample, standard, and quality control. Plasma NPY was measured 30 min after footshock, and tissue NPY was measured one week after footshock. In separate cohorts, we obtained blood samples from adult and adolescent animals immediately after the onset of the 30 min footshock protocol, or an hour after the end of the protocol (90 min post onset of exposure), in order to determine if there were increases in HPA response. For plasma collection, mice were anesthetized with isoflurane and euthanized by decapitation and trunk blood was collected and placed into tubes containing 13 µL heparin (Fisher, Waltham, MA, Cat#28-12100), 10 µL aprotinin (Millipore-Sigma, Burlington, MA, Cat#616370), and 10 µL DPP-IV inhibitor (Millipore-Sigma, Burlington, MA, Cat#DPP4-010) for 500 µL of blood. Tubes were prepared for plasma measurement of NPY levels or in tubes containing only 13 µL of heparin for corticosterone levels. Blood samples were spun at 3000 rpm for 15 min at 4 °C in a microcentrifuge. The plasma was collected and frozen with dry ice. The plasma was diluted 1:10 with ELISA assay buffer before ELISA measurements. For NPY tissue levels, hippocampus, amygdala, and prefrontal cortex were collected one week after footshock and flash frozen with dry ice. Samples were homogenized in NP-40 homogenization buffer (10 mM Tris pH 7.5,10 mM NaCl, 3 mM MgCl2·6H2O, 1.28 mM EDTA, 0.05% NP-40) and centrifuged at 16,000 ×g for 10 min at 4 °C. Supernatant was collected and stored at −80 °C. A Bradford assay was conducted to ensure uniform total protein level loading (1 μg/1 μL) into the ELISA assay. Samples were stored at −20 °C.

Data are reported as mean ± SEM. Data were tested for normal distribution and variance to determine whether to use parametric or nonparametric statistical tests. Data sets in which the means between two groups were compared, either T-test or a Mann–Whitney U test were applied. Grubb's test was used to determine statistical outliers. If the data did not exhibit equal variance, a Welch correction was used. Data sets in which means across four groups were compared, such as comparison between epoch and condition, a two-way repeated measures ANOVA was used with a Bonferroni post hoc test to identify significantly different groups.

Increased avoidance of the open arms of the EPM is traditionally interpreted as an increase in anxiety-like behavior, or avoidance of high-risk situations (Kraeuter et al., [21]; Sorregotti et al., [44]). Here we exposed adult male mice to a traumatic stress protocol consisting of a single 30-min session of fifteen variably timed 1 mA footshocks. To confirm that this protocol causes lasting behavioral changes, we tested their response on the EPM 8 days later. Adult mice exposed to footshock exhibited increased avoidance in the EPM compared to control (naïve) mice, including reduced open arm time (Figure 1(A)) and a lower percentage of entrances into the open arms (Figure 1(B)). The footshock protocol also caused a reduction in total entrances in to the four arms (Figure 1(C)), which may indicate a potential decrease in total locomotor activity in treated mice. Additionally, we found that the footshock group excreted significantly more fecal boli while on the EPM relative to the control adult group (Figure 1(D)).

PHOTO (COLOR): Figure 1. Adults display increase in avoidance behavior in EPM one week after footshock protocol. (A) The adult footshock group spends less time in the open arms compared to the control group (control, n = 10; footshock, n = 8; Mann–Whitney, *p = 0.027). (B) The adult footshock group has a lower percentage of open arm entrances compared to the control group (control, n = 10; footshock, n = 8; T-test, *p = 0.024). (C) The adult footshock group shows reduced total entrances compared to the control group (control, n = 10; footshock, n = 8; T-test, *p = 0.021). (D) The adult footshock group excreted more fecal boli on EPM than the control group (control, n = 10; footshock, n = 8; Mann–Whitney, *p = 0.017).

The decrease in total entries may indicate an impairment of locomotion or exploratory drive. To further test for changes in locomotor function, we used the OF task. We found that footshocked mice exhibited a significant decrease in locomotor activity as indicated by a decrease in the total distance traveled (Figure 2(A)) and increase in total resting time (Figure 2(B)). We found no significant difference in percent time spent in center between to two conditions (Figure 2(C)). To test whether the footshock protocol caused injury to the paws that could potentially cause motor impairment, we tested a subset of mice on the accelerating rotorod task 24 h after footshock exposure. However, there was no difference in the latency to fall off the rotorod as it accelerated (Figure 2(D)), suggestive of no paw injury or gross motor impairment.

PHOTO (COLOR): Figure 2. Adults display reduced activity after footshock protocol. (A) Total distance traveled in the open field is significantly lower in the footshock group compared to the control group (control, n = 10; footshock, n = 13; T-test, *p < 0.01). (B) Total resting time is higher in the footshock group compared to the control group (control, n = 10; footshock, n = 13; T-test, *p = 0.013). (C) Percent time spent in the center relative to periphery is not different between groups (control, n = 10; footshock, n = 13; T-test, p > 0.05). (D) Latency to fall off the rotarod is not significantly different between groups (control, n = 10; footshock, n = 9; T-test, p = 0.27).

We next exposed adolescent (6 weeks) male mice to the same footshock protocol and tested them 8 days later in the EPM. In contrast to adult mice, adolescent mice had no change in percent open arm time (Figure 3(A)) or in the percentage of entrances into the open arms (Figure 3(B)). There was a decrease in total arm entrances on the EPM (Figure 3(C)). Additionally, we found that the footshock group excreted significantly more fecal boli while on the EPM relative to the control adolescent group (Figure 3(D)). These results indicate that this robust footshock protocol causes a different behavioral profile in adolescent mice than was observed in adult mice.

PHOTO (COLOR): Figure 3. Adolescent mice show reduced exploratory behavior on EPM one week after footshock. (A) No significant difference between the adolescent footshock group and the control group in time spent in the open arms (control, n = 20; footshock, n = 14; T-test, p = 0.29). (B) No significant difference between adolescent footshock and control groups on percentage of open arm entrances out of total entries (control, n = 20; footshock, n = 14; T-test, p = 0.71). (C) Adolescent footshock shows reduced total entrances compared to control groups (control, n = 20; footshock, n = 14; T-test, *p < 0.001). (D) The adolescent footshock group excreted more fecal boli on EPM than the control group (control, n = 20; footshock, n = 14; Mann–Whitney, *p = 0.02).

We further examined the locomotor activity of treated adolescent mice using the OF, tested 8 days after footshock exposure. Adolescent mice exposed to footshock had significantly lower measures of total distance traveled (Figure 4(A)), and increased total resting time (Figure 4(B)). We found no significant differences between control and footshock group in the percent time spent in the center of the OF relative to the periphery (Figure 4(C)). This change in locomotor activity was not accompanied by a significant change in performance on the rotarod (measured 24 h after footshock), as animals exposed to the footshock protocol did not show a difference in the latency to fall off the accelerating rotarod (Figure 5(D)).

PHOTO (COLOR): Figure 4. Adolescent mice show reduced exploratory behavior after one-week after footshock. (A) Total distance traveled is significantly lower in the footshock group compared to the control group (control, n = 20; footshock, n = 13; Mann–Whitney, *p < 0.001). (B) Total resting time is higher in the footshock group compared to the control group (control, n = 20; footshock, n = 13; Mann–Whitney, *p < 0.001). (C) Percent time in center of open field is not different between control and footshock groups (control, n = 20; footshock, n = 13; T-test, p = 0.33). (D) Latency to fall off the rotarod is not significantly different between groups (control, n = 8; footshock, n = 8; T-test, p = 0.097).

PHOTO (COLOR): Figure 5. Footshock causes fear expression in adult and adolescent mice. (A) Percent of time freezing during habituation (1st epoch) and recovery (2nd epoch) (control, n = 40; footshock, n = 41). Two-way repeated measures ANOVA shows significant interaction between condition and epoch (F(1,77)=738.67 *p < 0.001), significant effect of condition (F(1,77)=674.87, *p < 0.001), and significant effect of epoch (F(1,77)=911.32, *p < 0.001. Bonferroni post hoc testing showed FS recovery group had more freezing than other groups (# p < 0.001). (E) Number of fecal pellets is not different between groups after footshock exposure (control, n = 40; footshock, n = 41; Mann–Whitney p = 0.59). (B) Percent of time freezing during habituation (1st epoch) and recovery (2nd epoch) (control, n = 34; footshock, n = 35). Two-way repeated measures ANOVA shows significant interaction between condition and epoch (F(1,67)=481.63, *p < 0.001), significant effect on condition (F(1,67)=470.80, *p < 0.001), and significant effect on epoch (F(1,67)=711.38, *p < 0.001). Bonferroni post-hoc testing showed that the FS recovery group had more freezing than other groups (#p < 0.001). (C) Number of fecal pellets is increased after footshock exposure (control, n = 34; footshock, n = 35; Mann–Whitney, *p < 0.001).

The footshock protocol was designed to study the effect of a traumatic experience in mice. However, we observed a reduction in approach (increased avoidance) behavior on the EPM only in adult mice but not in adolescent mice. We wanted to assess if the adolescent mice failed to have an acute aversive response to the footshocks. Alternatively, both age groups found the footshock protocol stressful, but the behavioral expression is different. To test this, we compared the freezing time measured during the 2-min habituation period at the beginning of the exposure and the 2-min recovery period at the end. Using two-way ANOVA, we observed an effect of epoch (recovery versus habituation), treatment (footshock versus control), and a significant interaction between epoch and treatment in adult mice (Figure 5(A)). Specifically, there was no difference in freezing during the habituation period, however, adult footshocked mice had a significantly higher freezing percentage at the recovery period than control mice (no shock) (Figure 5(A)), consistent with expression of fear. In adolescent mice, two-way ANOVA also showed an effect of epoch (recovery versus habituation), treatment (footshock versus control), and a significant interaction between epoch and treatment (Figure 5(C)). Freezing during recovery was greater than during habituation in both control and footshock mice. In addition, freezing during recovery was significantly greater in footshock mice than in control mice, indicating that the footshock caused fear expression (Figure 5(C)). These results show that the footshock protocol was aversive in both adolescent and adult mice.

A positive correlation between emotional reactivity and defecation in rodents such as increased fecal count on the OF with enhanced fear behavior can be observed (Denenberg, [11]). As a complimentary measure of fear expression (Denenberg, [11]; Leitermann et al., [22]), we counted the fecal pellets that were excreted by the end of the 30-min protocol. We did not observe a significant difference between the adult naïve or footshock groups (Figure 5(B)), indicating that fecal boli at this age may not be a robust indicator of fear expression. In contrast to adults, the adolescent footshocked animals had higher fecal pellets compared to their controls (Figure 5(D)). This finding is suggestive that fecal boli may serve as a secondary indicator of fear in adolescent mice, although it is not a robust feature.

Footshock exposure has been shown to activate the HPA axis in rodents, with elevations of corticosterone (CORT) at 30 min after the stressor in adults (Romeo et al., [35]), and delayed peak of CORT levels in juveniles (Eiland & Romeo, [14]; Romeo et al., [35]). We found that plasma CORT is significantly elevated after footshock in both age groups, with a significant increase at 30 min after the onset in adult footshocked animals (Figure 6(A)), and at 90 min after the onset in adolescent mice (Figure 6(D)). Together, these results demonstrate that the footshock protocol is aversive and produces activation of the HPA axis in both age groups, indicating a lasting stress response.

PHOTO (COLOR): Figure 6. Footshock exposure in results in enhanced CORT in plasma, with adolescent mice having delayed peak levels. (A) Adult mice that had experienced the footshock protocol had significantly higher levels of CORT in their plasma when measured immediately after the 30 min footshock exposure (control, n = 6; footshock, n = 6; T-test, *p < 0.01). (B) Adult mice that had experienced footshock did not have significantly higher levels of CORT compared to controls when measured 90 min after the onset of the footshock protocol (control, n = 5; footshock, n = 5; Mann–Whitney, p = 0.095). (C) Adolescent mice that had experienced the footshock protocol did not have a significant increase in plasma CORT levels when compared against controls 30 min after the onset of the footshock protocol (control, n = 7; footshock, n = 7; T-test, p = 0.26). (D) Adult mice that had experienced the footshock protocol had significantly higher levels of CORT in their plasma when measured 90 min after the onset of footshock exposure (control, n = 11; footshock, n = 11; Mann–Whitney, *p < 0.01).

Traumatic stress has been shown to acutely increase plasma NPY levels in humans (Morgan et al., [28]). We next tested the effects of footshock exposure on plasma NPY levels from blood collected immediately after the 30 min footshock exposure in adult and adolescent mice. We found that plasma NPY levels were significantly higher in footshocked mice than their respective controls in both adults and adolescents (Figure 7(A,B)). This is in agreement with other studies that have shown increased plasma NPY in adult rodents in response to footshock (Blizard et al., [2]; Zukowska-Grojec et al., [56]) as well as other acute stress paradigms (Sapolsky et al., [39]). NPY is released into plasma after sympathetic nervous system activation in rodent models (Lundberg et al., [25]); our results show that this is a feature that is also present in our stress paradigm in both adult and adolescent mice.

PHOTO (COLOR): Figure 7. Adult and adolescent mice have elevated NPY in plasma after footshock exposure. (A) Adult mice that had experienced the footshock protocol had significantly higher levels of NPY in their plasma when measured immediately after the 30 min footshock exposure (control, n = 6; footshock, n = 6; Mann–Whitney, *p < 0.01). (B) Adolescent mice that had experienced the footshock protocol had significantly higher levels of NPY in their plasma when measured immediately after the 30 min footshock exposure (control, n = 4; footshock, n = 6; T-test, *p = 0.038).

It has previously been shown in adult rodents that traumatic stress induced by predator odor causes a reduction in NPY in multiple brain regions including limbic areas (Cohen et al., [7]). We next tested whether footshock exposure also causes lasting reductions in NPY protein levels in hippocampus, prefrontal cortex, or amygdala that may be contributing the behavioral effects measured at the one week time point. Brain tissue from adult and adolescent mice was collected one week after footshock, and NPY protein levels were quantified using ELISA. In both adult and adolescent mice, alterations in NPY protein levels were only observed in hippocampus, with no change in prefrontal cortex or amygdala (Figure 8). Surprisingly, the direction of the effect was different between the two age groups. Adult footshock mice showed a significant reduction in hippocampal NPY (Figure 8(A)), whereas adolescent footshock mice showed a significant increase in hippocampal NPY (Figure 8(D)). In summary, while footshock triggers plasma NPY release in both adult and adolescent mice, the lasting consequences on hippocampal NPY levels are opposite.

PHOTO (COLOR): Figure 8. Adult and adolescent mice have lasting, opposing consequences on hippocampal NPY, but not amygdala or PFC. (A) One week after footshock exposure adult mice have in a reduction in hippocampal NPY compared to controls (control, n = 5; footshock, n = 6; T-test, *p < 0.01). (B) Adult mice do not have different levels of NPY in PFC compared to controls one week after footshock exposure (control, n = 5; footshock, n = 6; T-test, p = 0.39). (C) Adult mice do not have different levels of NPY in amygdala compared to controls one week after footshock exposure (control, n = 5; footshock, n = 6; T-test, p = 0.43). (D) One week after footshock exposure adolescent mice have increased hippocampal NPY relative to naïve mice (control, n = 8; footshock, n = 8; T-test with Welch correction, *p = 0.044). (E) One week after footshock exposure adolescent mice do not show a change of NPY in PFC relative to naïve mice (control, n = 5; footshock, n = 5; T-test, p = 0.72). (F) One week after footshock exposure adolescent mice do not show a change in amygdalar NPY relative to naïve mice (control, n = 7; footshock, n = 7; T-test, p = 0.51).

Here we found that a 30-min protocol of unpredictable footshocks causes lasting fear in adult and adolescent male mice, with a different pattern of behavioral expression. Whereas both age groups find the exposure immediately aversive, and exhibited a lasting robust hypoactive phenotype, only adult mice displayed increased avoidance behavior one week after footshock. Additionally, while both age groups had increased CORT as a consequence of footshock exposure, the increase in CORT occurred later in adolescent animals than adults, as previously reported (Eiland & Romeo, [14]; Romeo, [34]). Intriguingly, although both age groups display a heightened elevation of plasma NPY immediately after footshock exposure, the lasting effects on hippocampal tissue are opposite in adult and adolescent mice. While adult mice have reduced hippocampal NPY one week after footshock exposure, adolescent mice instead have increased hippocampal NPY.

The hippocampus is a key region involved with negative feedback of the HPA response (Kim et al., [19]; Sapolsky et al., [39]), therefore, it is striking that it is the only region tested to have alterations in NPY levels after footshock in both age groups. The footshock-induced reduction in hippocampal NPY levels in adults is consistent with reduced hippocampal NPY in adult rats after PSS (Cohen et al., [7]). Interestingly, reduced amygdalar and cortical NPY was also seen in that study (Cohen et al., [7]), whereas we observed no change in amygdalar or prefrontal cortex NPY after footshock. Because reduced hippocampal NPY occurs after footshock and exposure to predator odor, it may be a generalized response to trauma experienced in adulthood. Surprisingly, we found the opposite effect in adolescent mice, with increased NPY peptide in hippocampus one week after footshock. Although effects of footshock on adolescent NPY levels have been unexplored, our lab has previously shown that PSS in adolescent mice impaired NPY release in hippocampal CA1-Temporoammonic synapses (Li et al., [23]). The impaired NPY release following PSS could indicate reduced NPY levels in hippocampus, although alterations in the mechanisms governing NPY release could also contribute to these findings. Further work would be necessary to determine if abolished NPY release is also evident in adolescent footshock mice, despite having higher levels of NPY than naive mice. Together, these results show that hippocampus is a key region where NPY is altered by traumatic stress, in response to multiple stress protocols and in both adolescent and adult rodents.

Avoidance of situational reminders of traumatic experience is a core feature of PTSD (Substance Abuse and Mental Health Services Administration n.d.). Importantly, while adult footshock mice showed both increased avoidance on the EPM and hypolocomotion, adolescent footshock mice showed only hypolocomotion. The hypolocomotion was a robust effect of footshock in both age groups, as it was observed on both the EPM and the OF. Our results are in agreement with others that have demonstrated that footshock exposure causes immobility in young adult and adult rodents (Stam, [47]). Since we did not identify a deficit in motor function on the rotarod (Curzon et al., [9]), we interpret this hypoactive behavior as a generalized freezing response due to the traumatic nature of the footshock protocol. However, it is possible that impairments would be evident at the one week time point, contributing to the observed hypoactivity. While trauma-induced immobility occurs in both age groups, our results show that heightened avoidance behavior is a feature observed only in the adult group, indicating that this effect is age dependent. Together, these findings support the idea that in adult rodents, acute traumatic experiences result in lasting defensive behaviors, as previously observed (Li et al., [23]; Louvart et al., [24]).

It is tempting to speculate that the increased NPY in adolescent footshock mice is protective against the negative behavioral consequences that result from traumatic exposure. In adult rodents, both footshock and PSS caused reduced hippocampal NPY and increased avoidance behavior. Similarly, in adolescent mice, predator odor exposure causes increased avoidance behavior and reduced hippocampal NPY release (Li et al., [23]). Here we find footshock adolescent mice have no alterations in avoidance behavior and no reduction in hippocampal NPY; instead these mice actually have enhanced NPY. While the effects NPY overexpression on traumatic exposure have not yet been tested, previous studies have shown that viral NPY overexpression in hippocampus reduces baseline avoidance behavior on the EPM (Christiansen et al., [6]). However, previous work from our lab has shown that transgenic overexpression of NPY results in a mild avoidance behavior and hypoactivity compared to wild-type mice (Corder et al., [8]). Physiology experiments demonstrated impaired NPY receptor function in hippocampal slices from the NPY overexpression mice (Corder et al., [8]), possibly due to chronic overexpression. Future studies could measure the time course of enhanced hippocampal NPY in adolescent footshock mice, and determine if NPY receptor signaling is altered. It is possible that NPY was also reduced in adolescents at an earlier time point, and that the measured increase in NPY at the one week timepoint was a compensatory response. Nevertheless, our findings demonstrate the importance of understanding differential NPY regulatory mechanisms as a consequence of traumatic exposure in the hippocampus across the lifespan.

We observed increased plasma CORT expression in both age groups after footshock, indicating that this traumatic stressor results in activation of the HPA system. We found that the timing of the CORT increase after footshock was age-dependent, as previously observed (Eiland & Romeo, [14]).The later peak of CORT in adolescents is thought to be caused by immature glucocorticoid receptor expression at a younger age, which prolongs CORT elevation (Eiland & Romeo, [14]). Elevation of CORT is an acute response to stress, as footshock exposure in rats does not result in prolonged differences in plasma CORT levels (Stam, [47]). However, glucocorticoid receptors are elevated 2 week after footshock in adult rodents (van Dijken et al., [54]), indicating a lasting effect on the HPA system. Indeed, PTSD patients are reported to have hyperactive GR function without having a lasting difference in circulating cortisol levels (Dunlop & Wong, [13]; Klaassens et al., [20]), suggesting that footshock in adult rodents is an ideal model for studying aspects of the disorder.

Additionally, adolescent mice, but not adult mice, excreted more fecal pellets in response to footshock, suggesting that this may be an age-dependent secondary indicator of sympathetic activation (Denenberg, [11]; Leitermann et al., [22]; Stam et al., [48], [49]). It was stated in Denenberg ([11]) that an "emotional animal is one that will not move and will defecate." Therefore, it is interesting that there was no difference in fecal count among adults during the footshock administration, but there was a significant difference in the adolescent groups. These findings suggest that the adolescent age group may be more inclined to express negative affect through physiological processes than adult mice immediately after trauma. Additionally, the finding that both groups excreted more fecal boli on the EPM one week after footshock exposure may be an indirect indicator that both age groups have heightened sympathetic activation even at a week after trauma exposure. Previous work in adult rats has shown that NPY in the plasma is acutely increased after footshock, which is in line with findings of NPY being co-released from sympathetic terminals upon stress exposure (Zukowska-Grojec et al., [56]). Importantly, we show that footshock also causes elevated plasma NPY in both age groups when tested at the end of the 30 min protocol, despite differences in the timing of CORT elevation. Elevation of plasma NPY, in addition to the elevated fecal excretion, may be demonstrative of sympathetic co-release of noradrenaline and NPY, as reported by other labs (Lundberg et al., [25]). Further work is necessary to determine if there is a relationship between sympathetically released NPY and the regulatory mechanisms underlying alterations in hippocampal NPY observed in both age groups.

One limitation of our study is that the study design precludes making direct comparisons between adults and adolescents. However, it provided valuable new information about the effects of traumatic stressor on each age group independently, which revealed key differences. It would be of interest to compare between the two age groups in order to determine if there are underlying differences in fear memory acquisition, recall, or extinction. Additionally, it would be of interest to determine the role of the elevated hippocampal NPY in the adolescent mice and if it is a contributing or protective factor to the behavioral manifestations observed after footshock. Adult footshock mice did not differ from controls in the percent time spent in the center of the OF, which can serve as an additional measure of avoidance behavior. However, it is likely that our 5 minute OF trials were too short to properly assess for avoidance behavior, as it has been reported that even control mice tend to avoid the center regions for the first 10 min of the OF exposure (Choleris et al., [5]). Therefore, the lack of difference in the time spent in the center between footshock and control groups may be due to a floor effect. With EPM, altered locomotion is potential confound for interpreting the results (Dawson et al., [10]). However, it is well established that immobility is a rodent response to fear, and that footshock as a traumatic stressor induces immobility, particularly when measured in the EPM and OF (Stam, [47]). Overall, the changes we observed are consistent with lasting behavioral and biochemical consequences of the traumatic stressor in both age groups.

Alterations in NPY have been observed in multiple studies of adult patients with psychiatric disorders (Tural & Iosifescu, [53]), although results vary based on the type of disorder. The reduction in hippocampal NPY we observed in adult rodents is well in agreement with what has been seen in human studies of PTSD patients, who consistently have demonstrated lower levels of NPY both in the plasma and CSF compared to healthy controls (Tural & Iosifescu, [53]). Although there are no human studies to date that have measured NPY levels in adolescent patients, previous studies have identified childhood trauma to be positively correlated with bipolar depression (Soleimani et al., [43]) and suicidal behavior (Sher et al., [41]). Together, these data suggest that trauma in early life could have effects on NPY levels that consequentially impact adult behaviors and coping mechanisms.

Overall, our data indicate that traumatic experience results in a lasting impact on hippocampal NPY in both adolescent and adult male mice, although the effects are opposite. We also find differences in physiological functions and fear behavior between adults and adolescents in response to footshock as a traumatic stressor. Further work is needed to determine the underlying mechanisms by which stress alters NPY and affects the underlying circuits that drive fear-related behaviors at both ages. The observed reduction in hippocampal NPY in response to traumatic footshock supports the idea that adult PTSD patients may benefit from current investigational strategies to elevate NPY. However, there is a critical need to better understand the effects of stress on NPY in adolescents. In particular, determining if the increased hippocampal NPY in adolescents is protective or maladaptive could aid in the development of appropriate NPY-based treatments for PTSD patients in this age group.

No potential conflict of interest was reported by the author(s).