If you don't remember your password, you can reset it by entering your email address and clicking the Reset Password button. You will then receive an email that contains a secure link for resetting your password
If the address matches a valid account an email will be sent to __email__ with instructions for resetting your password
Rosalind Franklin University of Medicine and Science Discipline of Physiology and BiophysicsCenter for Neurobiology of Stress Resilience and Psychiatric Disorders
Context fear memory can be reliably reduced by subsequent pairings of that context with a weaker shock. This procedure shares similarities with extinction learning: both involve extended time in the conditioning chamber following training and reduce context-elicited fear. Unlike extinction, this weak shock exposure has been hypothesized to engage reconsolidation-like processes that weaken the original memory.
Methods
We directly compared the weak-shock procedure to extinction using male and female Long Evans rats.
Results
Both repeated weak shock exposure and extinction result in decreased context freezing relative to animals that received context fear conditioning but no subsequent context exposure. Conditioning with the weak shock is not enough to form a persistent context-shock association on its own, suggesting that the weak-shock procedure does not create a new memory. Weak shock exposure in a new context can still reduce freezing elicited by the training context, suggesting that it reduces responding through a different process than extinction, which does not transcend context. Finally, reduced fear behavior produced through both extinction and weak shock exposure was mirrored by reduced zif268 expression in the basolateral amygdala (BLA). However, only the weak shock procedure resulted in changes in K48 polyubiquitin tagging in the synapse of the BLA, suggesting this procedure produced long-lasting changes in synaptic function within the BLA.
Conclusions
These results suggest that the weak shock procedure does not rely on the creation of a new inhibitory memory, as in extinction, and instead might alter the original representation of the shock to reduce fear responding.
), and fear and anxiety can occur because of similarity between present circumstances and situations that were previously paired with aversive outcomes (
). Extinction-based exposure therapies reduce fear symptoms. Here, patients are repeatedly presented with cues (conditional stimulus, CS) that were previously paired with aversive events (unconditional stimulus, UCS) (
). Reductions in fear regulated by the basolateral amygdala (BLA) following extinction rely on retrieval of inhibitory learning acquired during extinction. Several factors can lead to a return of extinguished responding, including removal from the extinction context, UCS re-exposure, or passage of time (
). This suggests that original learning remains intact throughout extinction and will readily return if circumstances change. ‘Relapse’ phenomena are the most significant hurdle to treatment of neuropsychiatric diseases associated with maladaptive fear responding, including specific phobias and post-traumatic stress disorder (
Behavior can also be reduced by targeting the original memory, taking advantage of a brief time period following memory retrieval during which memory becomes sensitive to disruption known as reconsolidation. Amnesiac events (
), that occur during reconsolidation will weaken the original memory. Recent applications have used behavioral manipulations during reconsolidation to bidirectionally change the content of the memory to be either more (
). This has been confirmed through UCS deflation studies, where a CS is paired with food leading to conditional responding to that CS alone. Later, the food outcome is paired with illness. This decreases CS-elicited responding relative to groups with no UCS-illness pairings (
Cellular imaging of zif268 expression in the hippocampus and amygdala during contextual and cued fear memory retrieval: Selective activation of hippocampal CA1 neurons during the recall of contextual memories.
) mediated by the ubiquitin-proteasome system (UPS). Lysine-48 (K48) polyubiquitin tags specific proteins for degradation. Changes in synaptic K48 are crucial for both memory retrieval and destabilization in the amygdala and hippocampus (
Memory can be bidirectionally modulated by presenting either few or many weak versions of the unconditional stimulus. Animals exposed to a brief updating procedure (two 0.3mA shocks) following contextual fear conditioning (with five 1.0mA shocks) showed increases in fear responding (
). Somewhat paradoxically, 10 0.3mA shocks reduced fear to the conditioning context. Similar behavioral results were obtained using an auditory fear conditioning procedure to reduce fear to a discrete CS (
), suggesting this method is robust across behavioral paradigms. Increased fear following two weak shocks resulted in increased cellular activity and synaptic destabilization within the BLA, but the molecular processes associated with behavior following 10 weak shocks are unclear.
Several possible explanations exist for the behavior-weakening effect produced by 10 weak shocks (
). First, it is possible that this ten-shock procedure could have engaged a reconsolidation-like mechanism, which would reduce fear to the context by updating the context-shock association to now reflect the weak shock. Alternatively, this procedure might engage an extinction-like mechanism, which would reduce fear to the context through prolonged exposure to that context in the absence of the fear-provoking 1.0mA shock, but lower fear would be context-dependent (
) which would reduce behavioral responding by changing the UCS value. The current experiments were designed to directly compare the 10-shock deflation procedure to extinction and examine changes in retrieval-induced molecular processes within the BLA and DH.
Methods
Subjects
Subjects were age-matched male (250-274g) and female (175-199g) Long Evans rats from Envigo (Indianapolis, IN). The colony was maintained on a 12:12 light:dark cycle. Behavior occurred during the light cycle. Animals were acclimated to the colony for 7d prior to experimentation. Animals were handled for 2d prior to behavioral training. Male and female rats were run in separate experimental sessions.
Apparatus
Behavior occurred in Colbourn conditioning chambers each housed in its own sound-attenuating cubicle. Each chamber consisted of plexiglass front, rear, and top walls with brushed stainless steel side walls. The floor consisted of a shock grid with 18 rods (0.4cm diameter) spaced 1cm apart. A fan produced continuous 65dB noise. Chambers were lit with a white LED houselight. Between each animal, chambers were thoroughly cleaned with water. For experiments using two contexts, a second set of chambers was introduced. In these chambers no noise was played, the LED light was red, and they were cleaned with bleach between animals. White- and red-lit chambers were counterbalanced as Contexts A and B.
Training
Animals were assigned to one chamber and placed inside on the first day. After 2min, animals received 5 1-s 1.0mA footshocks with an interstimulus interval of 1min. Animals were removed 2min following the final footshock. In one experiment (Figure 1F), fear conditioning used a 0.3mA footshock.
Figure 1Weak shock presentations following context fear conditioning reduce freezing, but are not enough to create a persistent fear memory on their own. A) Behavioral design. All animals received 5 context-shock pairings during training. The next day animals received either 10 weak shock presentations (weak shock), exposure to the context alone (extinction), or remained in the home cage (no exposure). On the final day all animals were returned to the training context for testing. Group sizes were as follows: Weak Shock, N = 10 (M = 5, F = 5); Extinction, N = 9 (M = 5, F = 4), No Exposure, N = 10 (M = 5, F = 5). B) All groups increased their freezing throughout training. C) Both groups reduced their freezing throughout the day 2 session, although overall responding was higher in the weak shock group. D) Both weak shock and extinction groups showed reduced freezing relative to the no exposure control. E) Both weak shock and extinction groups showed reduced freezing relative to their freezing on day 2, but this was more pronounced in the weak shock group. F) Behavioral design. All animals received 5 context-weak shock pairings during training. The next day animals received either 10 weak shock presentations (weak shock), exposure to the context alone (extinction), or remained in the home cage (no exposure). On the final day all animals were returned to the training context for testing. Group sizes were as follows: Weak Shock, N = 6 (M = 3, F = 3); Extinction, N = 6 (M = 3, F = 3), No Exposure, N = 6 (M = 3, F = 3). G) All groups increased their freezing throughout acquisition. H) Both groups reduced their freezing throughout the day 2 session. I) All groups showed similarly low levels of freezing during the test. J) Both weak shock and extinction groups showed reduced freezing relative to their freezing on day 2. *p < 0.05, **p < 0.01, ***p < 0.001.
Day 2 (Weak Shock or Extinction). 24h later, the animals were split into three experimental conditions. Animals in the weak shock condition were placed in the conditioning chamber and, following 1-min, 10 1-s 0.3mA footshocks were delivered with an interstimulus interval of 1min. Animals were removed from the chamber 15s following the final footshock. Animals in the extinction condition were placed in the chamber for an equivalent amount of time (615s) but no footshocks were delivered. Animals in the no exposure condition remained in their homecages.
Testing
The next day, animals were tested for fear elicited by the conditioning chamber for 10min.
Tissue processing
Animals from the experiment depicted in the top half of Figure 1 were sacrificed 60-min following the test, along with a control group of rats who arrived at the same time but never received any behavioral training or testing. Animals were deeply anesthetized with isoflurane and brains were removed and immediately flash frozen. Tissue was later sliced and dissected for immunofluorescence and western blotting.
Immunofluorescence
Immunofluorescence proceeded similarly to previous work (
The anterior retrosplenial cortex encodes event-related information and the posterior retrosplenial cortex encodes context-related information during memory formation.
). Tissue was sliced in 40-micron sections and mounted onto charged slides. Slides were fixed in a 10% buffered formalin before being rehydrated in wash buffer (PBS+0.05% Tween-20) and permeabilized (PBS+0.3% Triton X) for 15-min and incubated in blocking solution (PBS+0.7% NGS). Slides were incubated in zif268/EGR-1 antibody (Cell Signaling, 1:400) solution (PBS+0.3% Triton X+5% NGS) overnight at 4ºC. The next day, tubes were given 2h at room temperature before incubation in secondary antibody solution for 2h. Slides were rinsed with wash buffer, a DAPI counterstain was applied, and coverslipped. Images were captured from the CA1 region of the DH and the BLA on the Leica THUNDER imager system using a 20x objective using LAS-X (Leica). Images were exported as 12-bit TIFF files and converted to a binary image via Gaussian filtering (sigmas: 6, 3) then quantified using the “Analyze Particles” plugin in ImageJ. zif268 activity was normalized as a proportion of DAPI present in the same section. Groups showed no difference in DAPI levels in the DH, F=1.61, p=0.188, or BLA, F=1.11, p=0.347.
Synaptosomal preparation
K48 polyubiquitin tagging within the synaptic compartment is associated with memory reconsolidation-like processes (
Distinct subcellular changes in proteasome activity and linkage-specific protein polyubiquitination in the amygdala during the consolidation and reconsolidation of a fear memory.
Neurobiology of Learning and Memory.2019; 157: 1-11
). Amygdalae were homogenized in TEVP buffer with 320mM sucrose and centrifuged at 1000x g for 10minutes. The supernatant was removed and centrifuged at 10,000x g for 10minutes, and the remaining pellet was denatured in lysis buffer (all in 100ml DDH20; 0.605g Tris-HCl, 0.25g sodium deoxycholate, 0.876g NaCl, 1μg/ml PMSF, 1μg/ml leupeptin, 1μg/ml aprotinin, 10ml 10% SDS). Protein levels were measured with a protein assay kit (Bio-Rad).
Western blotting
Following synaptosomal preparation, proteins were loaded onto a 7.5% SDS/PAGE gel and then to a membrane using a transfer apparatus (Bio-Rad). Membranes were incubated in blocking buffer (50% LI-COR TBS blocking buffer, 50% TBS+0.1% Tween-20) for 1h then incubated in K48 (1:500, Cell Signaling) or βactin (1:1000, Cell Signaling) primary solutions overnight at 4°C. Membranes were rinsed and washed three times for 5min in wash buffer (TBS+0.1% Tween-20) then incubated in appropriate secondary (1:15,000, IRDye 800CW goat anti-rabbit, IRDye 680RD goat anti-mouse, LI-COR) antibody for 1h at room temperature. Images were captured using the Odyssey Fc near-infrared system (LI-COR). Densitometry was performed using Image Studio. K48 was first normalized to actin (in which no group differences were observed, F=1.14, p=0.279).
All immunofluorescent and western blot data were expressed as a percentage relative to naïve animals who arrived in the colony at the same time, were housed and maintained the same, and were sacrificed at the same time as experimental rats (N=9; M=5, F=4). These animals were never given any behavioral experience or removed from the colony prior to tissue collection.
Data analysis
All data were analyzed with ANOVAs or t-tests (alpha=0.05) using SPSS. Planned comparisons assessed between- and within-group differences following main effects or interactions. Data are presented as group means and stratified by sex in Supplementary Figures 1-5.
Results
Weak shock exposure and extinction reduce context fear responding.
We first compared the weak-shock procedure to extinction following contextual fear conditioning (design in Figure 1A). Groups were compared to a No Exposure control that received training and testing, but remained in their homecage during Day 2.
Training
All animals similarly increased their freezing during training (Figure 1B). This was confirmed by a 3(Group) x 3(Time Period) ANOVA which found a main effect of time period, F(
)=261.27, p<0.001, but no main effect of group nor an interaction, Fs<1.
Day 2 (Weak Shock or Extinction). To assess responding throughout this session (Figure 1C), a 2(Group) x 11(Minute) ANOVA was conducted. This found main effects of group, F(
)=5.05, p=0.038, and minute, F(10,170)=9.25, p<0.001, but no interaction between the two, F<1. The extinction group froze less than the weak shock group, but both groups decreased freezing throughout the session.
Test. A one-way ANOVA compared freezing during the test session (Figure 1D) and found an interaction, F(
)=6.27, p=0.006. While the no exposure group froze more than the weak shock, p=0.007, and extinction, p=0.004, groups, they did not differ from each other, p=0.720. To examine how responding changed within-group between Phase 2 and the test, a 2(Group) x 2(Session) ANOVA was conducted (Figure 1E). This found a main effect of session, F(
)=2.35, p=0.14, demonstrating that while both groups decreased responding from Day 2 to testing, this decrease was larger following weak shock. Planned comparisons demonstrated that while the weak shock group showed a significant decrease during this time (p<0.001) this was only a trend in the extinction group (p=0.064).
Rats were then designated as either having decreased or increased responding between these days. If animals did not change responding by more than 5% in either direction, they were designated as no change. 90% of the weak shock condition reduced their freezing between-session and 56% of the extinction group did (Figure 3A).
Figure 3Within-subject changes between day 2 and testing demonstrate unique behavioral outcomes of extinction and weak shock. A) 90% of animals in the weak shock group and 56% of animals in the extinction group show a decrease in responding between day 2 and testing. B) 100% of animals in the weak shock group and 83% of animals in the extinction group show a decrease in responding between day 2 and testing. C) 88% of animals that received weak shock in the same context and 62.5% of animals that received weak shock in an alternate context show a decrease in responding between day 2 and testing. D) 75% of animals that received extinction in the same context show a decrease in responding between day 2 and testing. For the first and only time in this set of experiments, a majority of animals (57%) in the alternate group showed an increase in responding between day 2 and testing.
Exposure to the weak shock results in unconditional freezing to the context without creating a persistent fear memory.
We next aimed to test if the weak shock itself was enough to support a fear memory. All animals were first conditioned with the weak (0.3mA) shock (design in Figure 1F). As before, both the weak shock and extinction groups were compared to a group that did not receive behavioral manipulations between training and testing.
Training
Fear conditioning with the 0.3mA shock increased freezing throughout the training session (Figure 1G), confirmed by a 3(Group) x 3(Time Period) ANOVA which found a main effect of time period, F(
)=109.09, p<0.001, but no effect of group or interaction, Fs<1. This suggests that weak shock can produce unconditional freezing.
Day 2 (Weak Shock or Extinction). A 2(Group) x 11(Minute) ANOVA assessed responding throughout this session (Figure 1H). This found a main effect of minute, F(10,100)=5.30, p<0.001, and an interaction, F(10,100)=2.10, p=0.031, but no effect of group, F<1, suggesting the weak shock sustained more consistent levels of freezing during this session.
Test. A one-way ANOVA found no differences between the groups during testing (Figure 1I). All three groups showed low freezing to the context, suggesting that conditioning with the 0.3mA shock failed to create a persistent fear memory. A 2(Group) x 2(Session) ANOVA was conducted to assess behavioral decreases between day 2 and testing (Figure 1J). This found a main effect of session, F(
)=17.30, p=0.002, but no effect of group or interaction, Fs<1. Both groups weak shock (p=0.018) and extinction (p=0.012) decreased freezing between day 2 and testing. 100% of the weak shock group and 83% of the extinction group decreased their freezing between Day 2 and testing (Figure 3B).
Effects of weak shock exposure transfer across contexts.
Extinction learning is characterized by its context dependency. However, contextual novelty has been shown to facilitate reconsolidation-like processes (
). We therefore compared weak shock in either the same or an alternate context (design depicted in Figure 2A).
Figure 2Weak shock exposure conducted outside of the training context can reduce behavior in the training context. A) Behavioral design. All animals received 5 context-shock pairings during training. The next day animals received 10 weak shocks in either the same context as training (Same) or a novel context (Alternate). Group sizes were as follows: Same, N = 8 (M = 4, F = 4); Alternate, N = 8 (M = 4, F = 4). B) Both groups increased their freezing throughout the training session. C) Groups did not differ during the weak shock phase. D) Groups did not differ during the testing phase. E) Both groups decreased their freezing between day 2 and testing. F) Behavioral design. All animals received 5 context-shock pairings during training. The next day animals received exposure to either the same context as training (Same) or a novel context (Alternate). Group sizes were as follows: Same, N = 8 (M = 4, F = 4); Alternate, N = 7 (M = 4, F = 3). G) Both groups increased their freezing throughout the training session. H) While both groups gradually decreased their freezing throughout the second day, overall animals in the same context froze more than animals in the alternate context. I) Groups did not differ during the testing phase. J) While animals that received extinction in the same context decreased their responding between day 2 and the test, the opposite pattern was observed in animals that received exposure to the alternate context. *p < 0.05, **p < 0.01, ***p < 0.001.
Animals increased their freezing throughout training (Figure 2B). This was confirmed by a 2(Group) x 3(Time Period) ANOVA that found a main effect of time period F(
)=60.73, p<0.001, but no main effect of group nor an interaction, largest F=1.42, p=0.259.
Weak shock
A 2(Group) x 11(Minute) ANOVA assessed responding throughout the session (Figure 2C). Both groups decreased their responding throughout the session, indicated by a main effect of minute, F(10,140)=6.36, p<0.001. There was no effect of group or interaction, Fs < 1. During the first minute of the session before any shocks were introduced, animals in the alternate context froze less than animals in the acquisition context (p=0.055) suggesting that animals could discern the contexts prior to the first shock.
Test. An independent-samples t-test found no difference between the groups during testing, t(
)=0.261, p=0.798 (Figure 2D). A 2(Group) x 2(Session) ANOVA conducted to assess within-group differences between weak shock and testing (Figure 2E) found a main effect of session F(
)=11.64, p=0.004, but no effect of group nor interaction, Fs<1. Planned comparisons showed both same (p=0.021) and alternate (p=0.043) groups decreased freezing over this time. 88% of animals in the same context reduced their freezing between sessions and 62.5% of animals that received weak shock in the alternate context did (Figure 3C).
Exposure to a second context in the absence of the weak shock increases between-session freezing.
We next wanted to confirm that exposure to the alternate context in the absence of the UCS did not have the same effect as weak shock exposure in the alternate context (design depicted in Figure 2F), akin to an ABA renewal-like design. However, given that the alternate context should not elicit a fear response, exposure to the second context does not necessarily constitute extinction.
Training
Animals increased their freezing throughout training (Figure 2G). A 2(Group) x 3(Time Period) ANOVA that found a main effect of time period, F(
)=86.31, p<0.001, but no effect of group or interaction, Fs<1.
Extinction
A 2(Group) x 11(Minute) ANOVA assessed responding throughout the session (Figure 2H) and found main effects of both time, F(10,130)=4.91, p<0.001, and group, F(
)=20.50, p<0.001, but no interaction, F<1. Overall responding was higher in the same context than the alternate context and both groups decreased freezing over time.
Test
While groups showed no differences in overall freezing during the test session (Figure 2I), t(
)=0.739, p=0.473, a 2(Group) x 2(Session) ANOVA conducted to examine within-group differences between the extinction and test session (Figure 2J) found an interaction, F(
)=5.71, p=0.033, but no effect of session, F<1. Planned comparisons demonstrated that the same group decreased freezing throughout this time (p=0.006) and that a nonstatistical trend was observed in the opposite direction in the alternate group (p=0.091). 75% of animals that received exposure to the same context showed reduced freezing between session but only 29% of the animals that had exposure to an alternate context did. This was the only group run within this series of experiments in which a majority (57%) of the animals increased their responding between sessions (Figure 3D).
Amygdala molecular profiles differ following behavioral reduction achieved through either weak shock or extinction.
Animals were sacrificed 60 minutes following the test in Figure 1D and tissue was collected for immunofluorescence and western blotting. Activity in the DH has been associated with context learning, such that placement in a context alone is enough to drive zif268 expression in the CA1 region (
Impairment of the context preexposure facilitation effect in juvenile rats by neonatal alcohol exposure is associated with decreased Egr-1 mRNA expression in the prefrontal cortex.
). Further, immediate early gene zif268 is elevated in the BLA following fear conditioning, and increased fear is associated with higher BLA zif268 expression (
Cellular imaging of zif268 expression in the hippocampus and amygdala during contextual and cued fear memory retrieval: Selective activation of hippocampal CA1 neurons during the recall of contextual memories.
). It was previously demonstrated that K48, a polyubiquitin chain involved in proteasomal degradation, tagging was upregulated in the synapse following memory retrieval, but not formation (
Distinct subcellular changes in proteasome activity and linkage-specific protein polyubiquitination in the amygdala during the consolidation and reconsolidation of a fear memory.
Neurobiology of Learning and Memory.2019; 157: 1-11
). This polyubiquitin chain is an essential component for the synaptic changes underlying persistent fear memory modifications. Further, the presentation of a few lower-intensity shocks following context fear learning that results in elevated fear increases zif268-K48 co-expression (
). Collectively, this suggests that BLA activity and K48 tagging might be associated with interference during retrieval when reconsolidation-like, but not consolidation-like, effects occur.
zif268 activity was generally elevated in the DH in groups that received testing. zif268 was quantified as a proportion of DAPI in that same section and then expressed as a percentage of the control group, who received no behavioral training or testing (Figure 4A). A one-way ANOVA was significant, F(3,285)=4.15, p=0.007. No exposure (p<0.001) and extinction (p=0.008) groups showed increased zif268, the weak shock group only showed a trend (p=0.074). This result might be attributable to the context-independency of the behavioral reduction produced by the weak shock.
Figure 4A) zif268 in the DH expressed as a proportion of total DAPI relative to the naïve control group 60 minutes following testing. Elevated zif268 activity in the DH was observed in the No Exposure and Extinction groups relative to a naïve control group, but not in the weak shock group. B) Representative images from the DH in each group, with DAPI staining in blue and zif268 staining in green. C) zif268 in the BLA expressed as a proportion of total DAPI relative to the naïve control group 60 minutes following testing. While all three groups showed elevated zif268 activity in the BLA relative to a naïve control group, both weak shock and extinction reduced zif268. D) Representative images from the BLA in each group, with DAPI staining in blue and zif268 staining in green. *p < 0.05, **p < 0.01, ****p < 0.0001. Scale bar = 100uM.
Both extinction and weak shock reduced zif268 expression in the BLA. Total zif268 present in each section was quantified as described above (Figure 4C; representative images in Figure 4D). A one-way ANOVA was significant, F(3,296)=7.91, p<0.001. Every group showed increased zif268 activity relative to the control group (No Exposure: p<0.001; Weak Shock: p=0.003; Extinction: p=0.005). While animals in the No Exposure group had more zif268 expression than the animals in the weak shock group, p=0.031, this was only a trend relative to extinction, p=0.059, suggesting both extinction and weak shock exposure can reduce BLA activity in addition to decreasing freezing. In both groups Extinction (r=0.713, p=0.031) and No Exposure (r=0.666, p=0.035) DH activity was correlated with BLA activity. This was not the case in either the weak shock (r=0.429, p=0.215) or control (r=0.192, p=0.619) groups. This suggests that coordinated activity between the DH and BLA that occurs in context fear expression both before and after extinction is disrupted following the weak shock procedure, likely due to the context-independency of this effect.
Exposure to the weak shock produced a long-lasting upregulation in BLA K48 polyubiquitin tagging. We next examined K48 within the synaptic compartment of the BLA. All experimental groups were again compared to a naïve control group that received no behavioral training or testing. A one-way ANOVA found group differences, F(
)=2.82, p=0.057 (Figure 5B; representative lanes in Figure 5C). The weak shock group showed elevated K48 polyubiquitination relative to both no exposure (p=0.017) and extinction (p=0.020) groups, and slightly elevated relative to the control group (p=0.051). Memory retrieval impacted synaptic function differently in this group relative to both the no exposure group and the extinction group, despite showing behavioral responding similar to the extinction group (Figure 1D).
Figure 5A) Schematic of the BLA region where tissue was collected when animals were sacrificed 60 minutes following testing. B) Mean levels of synaptic K48 expression normalized to actin relative to a naïve control group. K48 polyubiquitin tagging was upregulated only in the weak shock group following the test. C) Representative blots from each group. *p < 0.05.
We found that both weak shock and extinction reduced freezing relative to a group that received fear conditioning with no behavioral manipulation between training and testing. We then found that while a weak shock resulted in a small increase in contextual freezing the day following training, this memory was not persistent as evidenced by low levels of freezing in the no exposure group during testing. Finally, we found that weak shock exposure in a novel context reduced responding to the training context. While zif268 was generally elevated in all experimental groups relative to naïve controls in both the DH and BLA, behavior reduced through either extinction or weak shock exposure corresponded with reduced zif268 expression in the BLA, suggesting the amygdala was the key site of plasticity related to behavioral change. We found increased synaptic K48 polyubiquitin tagging only in the weak shock group in the BLA, demonstrating this procedure produced long-term changes within the synapse of the BLA to promote behavior change.
When rats received the weak shock exposure in the same or alternate context as training, we found that while animals initially discriminated between the two contexts, presentation of the weak shock led to freezing similar to the group who received the weak shock in the original context. This is particularly interesting because extinction, by contrast, is constrained to the extinction environment. This context-dependency of extinction learning suggests that extinction results in the formation of a new inhibitory memory. This might be partially explained by the pattern of results examining cellular activity in the DH. zif268 expression was elevated in all but the weak shock group. Together, these results suggest that the weak shock effect does not rely entirely on new inhibitory memory like that created in extinction and instead acts on the original representation of the UCS to reduce fear responding. However, we should note that the procedure here did not directly compare UCS deflation in an alternate context to simple exposure to that context. Future work directly comparing UCS deflation to extinction in an alternate context will be especially fruitful in understanding how each procedure affects behavioral relapse.
These effects share similarities with previous work (
) in which a light predicted a loud noise, resulting in fear to the light. Following this conditioning, animals received presentations of the loud noise alone. Following this UCS habituation, animals responded less to the light relative to animals that received no habituation. However, that experiment used the same stimulus in both training and in the habituation phase. It is unlikely that using the same intensity of UCS on Day 2 here would reduce freezing behavior and instead it would likely result in enhanced fear conditioning, as we observed strong contextual fear conditioning with just five context-strong shock pairings. Instead, the theoretical account that can explain the present findings most completely is a UCS deflation account that suggests rats are modulating their responding to the context based on the updated value of the UCS (
). Under this account, the rats are flexibly gating their responding to the context to represent the less aversive 0.3mA shock that they experienced following fear conditioning with the strong shock.https://psycnet.apa.org/record/1973-10772-001
This work is the first to directly test how reducing behaviors through presentations of a less intense UCS directly compares to behavioral reduction achieved through extinction. Future work should investigate how weak shock exposure affects relapse effects, including spontaneous recovery and reinstatement. While the present experiments show context fear conditioning can be reduced with weak shocks in a novel context, it has yet to be determined if this same process will work with cued fear conditioning. While a similar procedure was previously employed (
). Systematic work directly comparing the weak shock UCS deflation to extinction as it pertains to relapse effects will therefore need to be conducted. The present results and others (
Cellular imaging of zif268 expression in the hippocampus and amygdala during contextual and cued fear memory retrieval: Selective activation of hippocampal CA1 neurons during the recall of contextual memories.
The anterior retrosplenial cortex encodes event-related information and the posterior retrosplenial cortex encodes context-related information during memory formation.
Distinct subcellular changes in proteasome activity and linkage-specific protein polyubiquitination in the amygdala during the consolidation and reconsolidation of a fear memory.
Neurobiology of Learning and Memory.2019; 157: 1-11
Impairment of the context preexposure facilitation effect in juvenile rats by neonatal alcohol exposure is associated with decreased Egr-1 mRNA expression in the prefrontal cortex.
G.R.B. and S.T. designed the experiments. G.R.B., E.M., P.K.R., N.C.F., and S.T. collected the data. S.T. analyzed the data. G.R.B and S.T. wrote the manuscript with input from N.C.F.
We would like to thank the Purdue University Department of Psychological Sciences animal care staff for excellent animal care and husbandry.
Financial Disclosures
This work was supported by the Purdue Department of Psychological Sciences and the Purdue Institute for Integrative Neuroscience, as well as the Rosalind Franklin University of Medicine and Science Discipline of Physiology and Biophysics. The authors report no biomedical financial interests or potential conflicts of interest.
00001-0&id=gr1.jpg){kind=link}
00001-0&id=gr3.jpg){kind=link}
00001-0&id=gr2.jpg){kind=link}
00001-0&id=gr4.jpg){kind=link}
00001-0&id=gr5.jpg){kind=link}