Behavioral and brain mechanisms mediating conditioned flight behavior in rats

Abstract

Environmental contexts can inform animals of potential threats, though it is currently unknown how context biases the selection of defensive behavior. Here we investigated context-dependent flight responses with a Pavlovian serial-compound stimulus (SCS) paradigm that evokes freeze-to-flight transitions. Similar to previous work in mice, we show that male and female rats display context-dependent flight-like behavior in the SCS paradigm. Flight behavior was dependent on contextual fear insofar as it was only evoked in a shock-associated context and was reduced in the conditioning context after context extinction. Flight behavior was only expressed to white noise regardless of temporal order within the compound. Nonetheless, rats that received unpaired SCS trials did not show flight-like behavior to the SCS, indicating it is associative. Finally, we show that pharmacological inactivation of two brain regions critical to the expression of contextual fear, the central nucleus of the amygdala (CeA) and bed nucleus of the stria terminalis (BNST), attenuates both contextual fear and flight responses. All of these effects were similar in male and female rats. This work demonstrates that contextual fear can summate with cued and innate fear to drive a high fear state and transition from post-encounter to circa-strike defensive modes.

Figure 1

A serial compound stimulus (SCS) elicits context-dependent flight-like behavior. (A) Behavioral design. (B) Schematic representation of the SCS. (C) Average freezing data for tone and noise stimuli during each SCS presentation during Habituation, Conditioning, and Retrieval. Rats showed lower freezing to the noise on the second and third day of conditioning. (D) Percentages of rats that showed at least one jump during an SCS for each respective day of behavioral testing. Of the rats that showed at least one jump, jumps were exclusive to noise stimuli (C3). (E) Average flight ratio in which positive numbers represent increased movement to the noise relative to tone, whereas negative numbers represent decrease activity relative to tone. Rats displayed flight like behavior when tested in the conditioning, but not habituation context. (F) Average freezing data during Retrieval shows that rats tested in the conditioning context showed high freezing during baseline and the first tone presentation but decrease to the noise, whereas rats tested in the habituation context showed low freezing to baseline and the tone which increase to the noise. (G) Averaged motor activity data from 10 s before SCS onset to 10 s after SCS offset in both the Habituation (Ctx A) and Conditioning (Ctx B) contexts. All data are represented as mean ± SEM pooled across sex [within-subjects design (n = 8)]; *, **, and *** denote p < 0.05, p < 0.01, and p < 0.001, respectively.

Figure 2

Flight-like behavior depends on context-US associations. (A) Behavioral design. (B) Average freezing data shows rats that received unsignaled footshocks (Shock) froze at high levels, whereas rats that were merely exposed to the context (No-Shock) froze at low levels. For Retrieval testing, Shock animals showed higher baseline freezing and a decrease in freezing upon white noise onset whereas No-Shock animals showed low baseline levels and remained freezing at moderate levels throughout the SCS. Average flight ratio shows that rats that Shock animals showed a bigger flight response than No-Shock animals. (C) Averaged activity data during the first trial of Retrieval for Shock and No-Shock animals. All data are represented as mean ± SEM pooled across sex [Shock (n = 8); No-Shock (n = 8)]; *, **, and *** denote p < 0.05, p < 0.01, and p < 0.001, respectively.

Figure 3

Extinguishing contextual fear reduces flight-like behavior. (A) Behavioral design. (B) Freezing data across the first extinction (Ext) or no-extinction (No-Ext) session comparing males and females averaged into 5-min blocks. Rats that underwent context extinction froze at high levels at the beginning of extinction which reduced by the end of extinction. Rats that did not undergo extinction did not show a reduction in freezing from the first to last block of context exposure. In addition, female Ext rats froze less to the conditioning context at late time points, whereas male No Ext rats showed greater generalization in the neutral context. (C) Ext animals did not show a significant reduction in baseline freezing, but did show a reduced flight response as shown by the reduced flight ratio. (D) Averaged activity data showing that Ext animals showed reduced activity during the white noise stimulus compared to No-Ext animals. All data are represented as mean ± SEM pooled across sex [EXT (n = 15; 7 male & 8 female); No-EXT (n = 14; 6 male and 8 female)]; *, **, and *** denote p < 0.05, p < 0.01, and p < 0.001, respectively.

Figure 4

Flight-like responses in rats are specific to white noise and not due to sensitization. (A) Behavioral design. (B) Average freezing and activity data showing that flight-like behavior is specific to the white noise stimulus regardless of the temporal of order of the SCS. In a Reversed order SCS, rats show a decrease in freezing and corresponding increase in activity to the first stimuli (noise) rather than the second (tone). The data additionally show that unpairing the SCS and US with a 60-s gap (Unpaired) prevents flight-like behavior compared to Standard SCS-US controls. This all further shown by averaged activity time across time (C). All data are represented as mean ± SEM pooled across sex [Standard (n = 7); Reverse (n = 8); Unpaired (n = 8)]; *, **, and *** denote p < 0.05, p < 0.01, and p < 0.001, respectively.

Figure 5

An unconditioned SCS fails to evoke flight behavior in a threatening context. (A) Behavioral design. (B) Averaged freezing and activity data showing that rats that received US-alone trials throughout conditioning (US) and tested in a US-associated context (Shock) had a reduced flight response in comparison to rats that received SCS-US pairings. This is shown by increased freezing and decreased activity to the white noise stimulus, and a reduced flight ratio. Rats that were tested in a neutral context (No-Shock) also showed reduced flight responses compared to SCS-US/Shock animals. This is further shown by the average activity trace of each group (C). All data are represented as mean ± SEM pooled across sex [SCS-US/Shock (n = 7); SCS-US/No-Shock (n = 7); US-alone/Shock (n = 7); US-alone/No-Shock (n = 7)]; *, **, and *** denote p < 0.05, p < 0.01, and p < 0.001, respectively.

Figure 6

Pharmacological inactivation of either the BNST or CeA disrupts flight-like behavior. (A) Behavioral design. Histological summary of CeA (B) and BNST (C) cannula placements with representative thionin-stained sections and drug spread with fluorescent muscimol. Labeled anterior–posterior coordinates are relative to bregma. (D) averaged freezing data showing that CeA and BNST animals both showed reduced baseline freezing. BNST animals increased freezing to tone presentation and remained at higher freezing levels during white noise. CeA animals remained at low levels of freezing during the SCS. (E,F) averaged activity levels during the first SCS presentation showing that BNST animals showed a reduced flight response and CeA animals did not increase activity from tone to noise at all despite a lack of freezing behavior. This is reflected in the flight ratio (G). All data are represented as mean ± SEM pooled across sex [CeA (n = 14); BNST (n = 16); SAL (n = 16)]; *, **, and *** denote p < 0.05, p < 0.01, and p < 0.001, respectively.