High Salt Intake Lowers Behavioral Inhibition

Abstract

Stress-related neuropsychiatric (e.g., anxiety, depression) and cardiovascular diseases are frequently comorbid, though discerning the directionality of their association has been challenging. One of the most controllable risk factors for cardiovascular disease is salt intake. Though high salt intake is implicated in neuropsychiatric diseases, its direct neurobehavioral effects have seldom been explored. We reported that elevated salt intake in mice augments neuroinflammation, particularly after an acute stressor. Here, we explored how high salt consumption affected behavioral responses of mice to mildly arousing environmental and social tests, then assessed levels of the stress-related hormone corticosterone. Unexpectedly, anxiety-related behaviors in the elevated plus maze, open field, and marble burying test were unaffected by increased salt intake. However, nest building was diminished in mice consuming high salt, and voluntary social interaction was elevated, suggesting reduced engagement in ethologically-relevant behaviors that promote survival by attenuating threat exposure. Moreover, we observed significant positive correlations between social preference and subsequent corticosterone only in mice consuming increased salt, as well as negative correlations between open arm exploration in the elevated plus maze and corticosterone selectively in mice in the highest salt condition. Thus, heightened salt consumption reduces behavioral inhibition under relatively low-threat conditions, and enhances circulating corticosterone proportional to specific behavioral shifts. Considering the adverse health consequences of high salt intake, combined with evidence that increased salt consumption impairs inhibition of context-inappropriate behaviors, we suggest that prolonged high salt intake likely promulgates maladaptive behavioral and cardiovascular responses to perceived stressors that may explain some of the prevalent comorbidity between cardiovascular and neuropsychiatric diseases.

Keywords: anxiety; cardiovascular and neuropsychiatric comorbidity; corticosterone; salt loading; social behavior; stress.

FIGURE 1

Timeline of Experiments 1, 2, and 3. Drinking manipulations commenced at lights off (1800 h; black rectangles) on Day 0 (i.e., Salt Start), and continued for up to 7 days. Mice in Experiments 2 and 3 were group housed, whereas mice in Experiment 1 were singly housed; the latter to enable individual measurements of nest building. Single nestlets were introduced into the cages of singly housed mice in Experiment 2 at 1315 h during the fourth day, and nest building was scored in the colony room at 1, 20, and 48 h time points thereafter (vertical dotted lines). Mice in Experiments 1 and 3 underwent open field testing (10 min) followed immediately by social interaction testing (10 min) between 1115 and 1415 h on the seventh day, then blood was collected 2 h after completion of social interaction testing (approximately 1340–1640 h; red arrow for Experiments 1 and 3). For Experiment 2, mice were tested in the elevated plus maze (5 min) between 1600 and 1800 h at the end of the sixth day, then immediately tested for marble burying (30 min). On the seventh day, 18 h after completion of the marble burying test, blood was collected from mice in Experiment 2 (approximately 1040–1310 h; red arrow for Experiment 2).

FIGURE 2

Behavior in elevated plus maze and open field test. Group housed mice from Experiment 2 were tested in the elevated plus maze (A). No differences were observed the percent of arm time spent in the open arm (A, left half) or latency to first entry in the open arm (A, right half). Single and group housed mice in Experiments 1 and 3, respectively, were tested in the open field (B,C). Neither salt nor housing condition significantly affected overall locomotor activity, measured as distance traveled (B), though a non-significant trend for a main effect of housing was noted (p = 0.085). There was a significant main effect of housing on the percent of total distance traveled in the center (C, ^∗p = 0.013). Individual data points are shown, with means indicated by horizontal lines, and 95% CI indicated with vertical lines.

FIGURE 3

Marble burying, nest building, and social behaviors. Group housed Experiment 2 mice exhibited no significant differences in marble burying behavior (A). Nest building behavior was scored for singly housed mice in Experiment 1 at 1, 20, and 48 h after introduction of a nestlet (B). After 1 h, nest building was significantly impaired in both 2% (^∗∗p = 0.0087) and 4% (^∗∗∗p < 0.0001) salt-consuming mice. Nest building by mice consuming 4% salt remained inhibited at the 20 (^∗∗∗p < 0.0001) and 48 h (^∗∗∗p < 0.0001) time points. Preference for voluntary social interaction (C) was significantly increased in group housed mice in Experiment 3 (^∗p = 0.015), and a non-significant trend for increased social interaction in singly housed Experiment 1 mice was noted (p = 0.086). For (A,C), individual data points are shown, with means indicated by horizontal lines, and 95% CI indicated with vertical lines. Means (N = 8) are indicated with symbols in (B), and vertical lines indicate 95% CI.

FIGURE 4

Log-transformed serum corticosterone levels, serum sodium levels, and serum osmolarity. Corticosterone levels in Experiment 3 (group housed) mice consuming 0 and 2% salt, and in Experiment 1 (singly housed) mice consuming 4% salt, were non-normally distributed, so all corticosterone levels were log-transformed to generate normally distributed data for ANOVA analyses. Log-transformed corticosterone levels consistently revealed elevated corticosterone in mice consuming 4% salt in Experiment 3 [group housed, ^∗p = 0.026, (A) right side], Experiment 1 [single housed, ^∗p = 0.016, (A) left side], and Experiment 2 [group housed, ^∗∗∗p = 0.0004 (B)]. Serum sodium (Na⁺) levels were significantly higher in mice consuming 4% salt in both Experiment 3 [group housed, ^∗∗∗p < 0.0005, (C) right side] and Experiment 1 [singly housed, ^∗p = 0.011, (C) left side]. Serum osmolarity was also significantly elevated in both group [Experiment 3; ^∗∗∗p < 0.0001, (D) right side] and single [Experiment 1; ^∗∗p = 0.004, (D) left side] housed mice consuming 4% salt relative to control mice, as indicated by Dunnett’s post hoc testing. Individual data points are shown, with means indicated by horizontal lines, and 95% CI indicated with vertical lines.

FIGURE 5

Pearson correlations between log-transformed corticosterone levels and behavior in the elevated plus maze and social interaction test. Pearson correlations between individual log-transformed serum corticosterone levels and percent arm time spent in the open arms of the elevated plus maze (A) and latency to first open arm entry in the elevated plus maze (B) were determined for group housed mice in Experiment 2. Pearson correlations between individual log-transformed serum corticosterone levels and social preference in the social interaction test were determined after collapsing across housing conditions for mice in Experiments 1 and 3 (C). Pearson r-values are reported on the right side of each corresponding graph, along with the associated p-value. Bolded and asterisked p-values indicate values < 0.05.