Social and anxiety-like behaviors contribute to nicotine self-administration in adolescent outbred rats

Abstract

Both emotional and social traits interact with genetic factors to influence smoking behavior. We previously established a socially acquired nicotine intravenous self-administration model where social learning of a nicotine-associated odor cue reversed conditioned flavor aversion and promoted nicotine intake. In this study, we first phenotyped ~800 adolescent heterogeneous stock rats in open field, novel object interaction, social interaction, elevated plus maze, and marble burying behaviors. These rats were then phenotyped on socially acquired nicotine self-administration. We found 243 significant correlations between different behavioral tests. Principal component regression analysis found that ~10-20% of the variance in nicotine-related measures, such as intake during the first or the last three fixed-ratio sessions, the progressive ratio session, and reinstatement behavior, can be explained by variations in behavioral traits. Factors corresponding to social behavior and anxiety were among the strongest predictors of nicotine intake and reinstatement of nicotine-seeking behavior. We also found many sex differences in behavioral measures. These data indicated that the genetic diversity of this population, in combination with social behaviour and anxiety, are significant contributors to the divergent nicotine self-administration behavior and indicated a high probability of discovering sex-specific genetic mechanisms for nicotine intake in future genome-wide association studies.

Figure 1

Correlations between behavioral tests. A total of 35 measures were obtained from the five behavioral tests conducted before the commencement of nicotine self-administration. Pearson correlation coefficients between these measures were plotted on the upper (left) triangle of the heatmap. The Holm-adjusted p values (converted using -log10) for these correlations were shown on the lower (right) triangle of the heatmap. The top dendrogram was obtained using hierarchical clustering. OFT: open-field test, NOIT: novel object interaction test, SIT: social interaction test, EPM: elevated plus maze, MBT: marble bury test.

Figure 2

Selected scatter plots of behavioral correlations. EPM total travel distance is significantly correlated with (A) EPM close arm duration (p < 2.26 × 10⁻¹⁶, r = −0.62) and (B) NOIT object zone frequency (p < 0.5 × 10⁻¹³, r = 0.26). (C) EPM central duration is correlated NOIT distance to object zone (p < 1.80 × 10⁻¹³, r = −0.26). NOIT: novel object interaction test, EPM: elevated plus maze. The black line represents the result of linear regression.

Figure 3

Socially acquired nicotine self-administration, fixed ratio 10. Rats were split into high vs low nicotine intake groups for each sex based on the median of nicotine infusion during the 10 fixed-ratio sessions. The high intake groups licked significantly more on the active spouts compared to the inactive spout (A), while the reverse was true for the low intake group (B). Error bars in these panels represent standard errors. The mean nicotine intake (log transformed) during the 10 fixed-ratio sessions is shown as a box plot (C), where the thick black line in the middle of the boxes represent the median, the height of the boxes represent the interquartile range (IQR), the top and bottom hinges represent the ±1.5 IQR, and the grey circles represent individual data points that are outliers. The total nicotine infusion during the 10 sessions for females and males (log transformed) are shown as histograms in (D).

Figure 4

Sex differences in socially acquired nicotine self-administration. As shown in the box plots, females obtained significantly more nicotine infusions during the first three (A) and the last three (B) fixed-ratio sessions and during the progressive ratio session (C) than males. Both females and males licked significantly more on the active spout than on the inactive spout during the contextual cue-induced reinstatement session. Females also showed stronger reinstatement than males (D). In each plot, the thick black line in the middle of the box represents the median, the height of the box represents the interquartile range (IRQ), the top and bottom hinges represent the ±1.5 IRQ, the grey circles are individual data points that are outliers. ***p < 0.001, compared to females. ^###p < 0.001 compared to the corresponding active spout.

Figure 5

Variable loading for principal component analysis. Principal component (PC) analysis followed by a varimax rotation procedure were conducted on behavioral measures obtained from the five non-nicotine-related behavioral tests. The loading of each PC was plotted as a heatmap.

Figure 6

Lack of correlation between nicotine metabolism and nicotine self-administration. Rats received 5 intravenously nicotine injections (30 μg/kg) one day after the progressive ratio test. Blood samples were withdrawn 3 h after the last injection. Plasma cotinine levels were measured using a rat ELISA assay.