Decrease of ARC protein expression in the striatum after tone fear conditioning

Introduction. The involvement of the striatal system in S-R learning is usually based on neural plasticity related to immediate early-genes (IEGs). Previous studies also have shown that the dorsal striatum plays a role in tone fear conditioning (TFC). Objectives. Given that IEg expression in dorsal striatum supports S-R learning we analyzed early molecular consolidation events in the striatum by measuring the protein levels of the EGR1, C-Fos, and Arc in the striatum 30 and 90 minutes after the TFC training. Additionally, to minimize a dorsal hippocampal possible interference, glutamatergic transmission was disrupted during fear conditioning training using the NMDA receptor antagonist AP5 injection into hippocampus.

Method. Wistar rats received AP5 or saline injection in the hippocampus five minutes before undergoing tone fear conditioning (tone and foot-shock pairings) or tone only. Results. Animals that received tone and footshock pairings presented a decrease in ARC protein 30 minutes after training when compared to the tone groups. AP5 treated group exposed to tone only condition presented a decrease in EGR protein 90 minutes after training when compared to the saline and tone. No differences were observed in FOS protein levels. Conclusions. Our results suggest that it is possible that some interaction between striatum and hippocampus in processing tone experience and that reduced levels of ARC could be related to the associative features of this pavlovian task. Keywords. Memory; tone fear conditioning; striatum; immediate-early genes

INTRODUCTION
There is extensive evidence of the involvement of the striatal system in learning and memory processes, mainly in the acquisition of habits [1][2][3][4] . These studies showed that response and cue learning are usually associated with the striatum. Damaging this structure in rats impairs learning in mazes that require a response solution [1][2][3][4] . It also impairs learning in the cue version of the water maze task, but not in the hidden platform spatial version 3 . These data suggest that the hippocampus and the dorsal striatum are parts of systems that differ in the type of memory they mediate.
These striatal dependent tasks involve stimulus-response (S-R) learning.
The involvement of the striatal system in S-R learning is usually based on neural plasticity related to immediate early-genes (IEGs), including IEG mRNA or protein products as C-Fos, EGR1 (Early growth response protein 1, also known as ZIF268, zinc finger protein 268) and ARC (Activityregulated cytoskeletal associated protein) [5][6][7][8] . During memory acquisition and consolidation, intracellular molecular downstream events result in IEG induction, leading to the gene expression and protein synthesis required for memory formation. Several stimuli can induce the IEGs, such as neurotransmitters, peptides, and growth factors; their expression indicates a role in memory processing and behavior 9 .
The idea that the striatal system is also important in the fear conditioning paradigm, a form of CS-US associative learning, in addition to its already well-established role in S-R learning, has become stronger 2,10-12 . This learning involves pairing an aversive stimulus (such as an electric shock -US) with a neutral stimulus (such as a tone -CS). After a stimuli association, the tone acquires the capacity to elicit behavioral and physiological fear responses 10,11 . However, the conditioned responses involve autonomic (e.g., heart rate and respiratory frequency) and somatomotor responses such as freezing, defined as complete absence of movement, except for that required for respiration 13 . Freezing time is used as an index of learning in classical conditioning paradigm, and previous studies in our laboratory showed that bilateral lesion of the dorsal striatum (DS) decreases the freezing response to a conditioned tone without interfering with the context fear conditioning (CFC) 10 and that the indirect amygdala striatum pathway is mediating this learning 11 . Moreover, medial striatum receives projections from auditory inputs. Later, other studies have corroborated the involvement of the striatum in tone fear conditioning 12 .
Given that IEG expression in dorsal striatum supports S-R learning we analyzed early molecular consolidation events in the striatum by measuring the protein levels of the EGR1, Fos, and Arc in the striatum after the TFC training.
In humans, a negative reaction to a sound, in which individuals experience intense emotional experiences and autonomic arousal when they are confronted with specific sounds can be consider misophonia, a condition that has been associated some psychological disorders, including anxiety and obsessive-compulsive disorders 14 . Some studies suggest that misophonia is associated with altered brain activity in the auditory cortex and corticostriatal areas 15,16 .
Although this condition has not yet been observed directly in animal models, the study of tone fear conditioning could contribute to clarify our understanding of this pathology and can provide directions for future research into the mechanisms of misophonia.

Subjects
Three-month-old male Wistar rats, weighing 300-400g, were bred and raised in animal facility of CEDEME (Centro de Committee (#0926/08), following international rules for animal use and care.

Apparatus
The conditioning chamber consists of a black acrylic box, measuring 22x24x35cm, with a transparent acrylic lid.
The floor is made of parallel metallic rods, each 0.4cm in diameter, 1.2cm apart, and connected to a scrambler electric shock generator (AVS Projetos). For the tone conditioning test, a white cylindrical chamber (diameter 35cm, height 60cm) covered with a transparent acrylic lid was used. The two apparatus were cleaned with different substances and kept in different rooms. A tone (60dB) was produced using a buzzer, placed outside the two apparatuses. Tone alone generates no alterations on rat behavior, just a subtle startle response first time it is presented.

Procedures
Conditioning Training. Training consists of place the animals individually in the conditioning chamber, where they remained for two min before any tone or tone-shock pairing was delivered. After this period, five tone or tone-footshock pairings were given. The tone (CS) sounded for 5 seconds (60dB) and during the last second a foot shock (US; 0.6mA/1 s) was delivered ending together with the tone. The interval between pairings was 30 seconds. These procedures were performed just as described previously in Ferreira et al studies 10,11 .
After the pairings, the rats remained in the conditioning chamber for an additional minute and were then returned to their home cages. The control group animals, the SOUND group, were put in the conditioning chamber and exposed only to buzzer tone (no footshock was delivered, 60dB sound

Experiment 1B. IEGs expression in the striatum after 90 min of Fear Conditioning Training
In Experiment 1B, the rats underwent the same experimental design described above, except that the structures were collected 90 minutes after TFC training.
In the analysis performed in the striatum by western blot, a reduced expression of EGR1 was observed ( Figure   1D). TREATMENT x GROUP interaction (F (1,16) =8.35; p=0.010) was significant. Post-hoc analysis showed that the SOUND-AP5 group presented a reduced expression of EGR1 protein compared to the SOUND-SAL group (p=0.001).
Interestingly, no reduced expression was observed in the PAIRED-AP5 group when compared to the PAIRED-SAL.
There were no differences for TREATMENT (F (1,16) 1E). Also no differences in protein expression of FOS were observed in the striatum (Fig. 1F)  There is a decreased expression of ARC on PAIRED group when compared do SOUND group. In panel D. We also observed a decreased expression of EGR1 protein level in the SOUND-AP5 group No other differences were detected. n in parenthesis represents the total number of pool replications. All graphs show mean + SE. (* = p<0.05).

Experiment 2. Effects of intrahippocampal AP5 administration on CFC and TFC
The guide cannulae placement in the dorsal hippocampus is represented in Figure 2.  Figure 3A). TFC were conducted 24hs after the CFC. After three minutes, the same training tone was emitted. As Figure   3B shows, AP5 administration did not alter freezing time in

DISCUSSION
The main results of the present study are: 1) animals that received tone and footshock pairings presented a decrease in ARC protein 30 minutes after training when compared to the tone groups. 2) AP5 treated group exposed to tone only condition presented a decrease in EGR protein 90 minutes after training when compared to the respective control group, but spared TFC conditioned freezing response.
The present study showed that 30min after the training of TFC there was a reduction on the Arc protein expression in the striatum. The effect direction was the opposite of expected by the hypothesis generated in this study. If the involvement of the striatum in TFC was like the one observed in S-R tasks 5-8 we expected an increased expression of IEGs.
Although the result was unexpected, the two PAIRED groups exhibited this similar profile, reinforcing the idea that some aspect of the TFC training led to a decrease of ARC protein expression when compared to animals that were exposed to the training environment and the tones.  19,20 . Because the decrease of ARC occurred in PAIRED groups regardless of the administration of AP5 or saline administration, the observed effect in both groups seems to be a consequence of the training procedures and not the hippocampus manipulation.
Given that it has been shown that the striatum is involved in TFC 10-12 , the decreased expression of ARC after a CS-US associative learning shows that tone fear conditioning induces a different profile of genes expression in the striatum from the ones observed in the literature, where increased expression of IEGs has been observed, most of them using appetitive tasks 6,7 . To our knowledge, this is the first study to have observed decreased IEG expression in the striatum after TFC. Thus, is possible that Pavlovian and instrumental conditioning differ in the way the striatum mobilizes IEG expression. In an S-R learning task in a plus maze, an increase in c-fos in the dorsal striatum was observed.
However, this same study did not find differences in Zif-268 (i.e. EGR1) expression 7 . Other work observed an increased ARC protein levels in the striatum after a discrimination task in a plus maze 6 . The subregions of dorsal striatum also take part in drug-seeking neural plasticity of behavior. In cocaine self-administration protocols, ARC expression is decreased in dorsal medial striatum (DMS) but not in dorsal lateral striatum (DLS) in addiction-vulnerable animals 21,22 . In these studies, gene and protein expressions were evaluated after an extensive behavioral protocol (self-administration training, progressive ratio, extinction, and finally, cue induced-reinstatement of lever pressing). Despite the typical instrumental learning used for an extended period of this task acquisition, these animals were exposed to a CS-US classical conditioning protocol, in which a sound signalized the cocaine availability. After that, long extinction training was conducted, and IEGs were analyzed 24 hours after the test of cue induced-reinstatement of lever press in a protocol very similar to test a CS-US association (typical pavlovian conditioning Besides the ARC protein differential expression, the striatum is critical for S-R learning 1-4 , but it does not seem to be critical for TFC for the acquisition of the association between CS and US, once lesions in the striatum after training did not alter heart rate pattern 25 . In the same way, lesions in the striatum impaired the acquisition of conditioned eyelid response in classical conditioning, but no effect was observed in the heart rate conditioned response 26 .
However, the striatum has a critical role when the conditioning is evaluated by the conditioned somatomotor response 10,11 . Thus, we speculate that the organization of the somatic-response aspect at the conditioned response can be related to a decreased expression of IEGs in the striatum.
In experiment 1B we observed no EGR1 alteration when comparing GROUPS, but we show a decreased EGR levels 90min after the animals received TONE-AP5. Taken together, these results suggest that hippocampal NMDA activity, which is important for contextual learning 19,20 , can still have some influence on the acquisition/consolidation of fear conditioning and even affect striatal function. In addition, it has been proposed that hippocampus and striatum may potentially interact and that, depending on the learning situation, this interaction may be cooperative or competitive with each other during learning 27,28 .

CONCLUSION
Overall, we observed that after short periods of TFC training, there is a reduction in IEGs. Our results also suggest that it is possible that some interaction between striatum and hippocampus in processing tone experience and that reduced levels of ARC could be related to the associative features of this pavlovian task.

ACKNOWLEDGMENT
The authors would like to thank José Bernardo Costa for technical assistance during the behavior experiments.
This study was financially supported by grant