Skip to main content
NCBI home page
NIHPA Author Manuscripts logoLink to NIHPA Author Manuscripts
. Author manuscript; available in PMC: 2020 Sep 1.
Published in final edited form as: Genes Brain Behav. 2018 Nov 28;18(7):e12531. doi: 10.1111/gbb.12531

A novel role for acid-sensing ion channels (ASIC1A) in Pavlovian reward conditioning

Ali Ghobbeh 1,2, Rebecca J Taugher 1,2, Syed M Alam 1,2, Rong Fan 1,2, Ryan T LaLumiere 7, John A Wemmie 1,2,3,4,5,6
PMCID: PMC6818262  NIHMSID: NIHMS1055939  PMID: 30375184

Abstract

Pavlovian fear conditioning has been shown to depend on ASIC1A, however it is unknown whether conditioning to rewarding stimuli also depends on ASIC1A. Here we tested the hypothesis that ASIC1A contributes to Pavlovian conditioning to a non-drug reward. We found effects of ASIC1A disruption depended on the relationship between the conditional stimulus (CS) and the unconditional stimulus (US), which was varied between five experiments. In experiment 1, when the CS preceded the US signaling an upcoming reward, Asic1a−/− mice exhibited a deficit in conditioning compared to Asic1a+/+ mice. Alternatively, in experiment 2, when the CS co-initiated with the US and signaled immediate reward availability, the Asic1a−/− mice exhibited an increase in conditioned responses compared to Asic1a+/+ mice, which contrasted with the deficits in the first experiment. Furthermore, in experiments 3 and 4, when the CS partially overlapped in time with the US, or the CS was shortened and co-initiated with the US, the Asic1a−/− mice did not differ from control mice. The contrasting outcomes were likely due to differences in conditioning because in experiment 5 neither the Asic1a−/− nor Asic1a+/+ mice acquired conditioned responses when the CS and US were explicitly unpaired. Taken together, these results suggest that the effects of ASIC1A disruption on reward conditioning depend on the temporal relationship between the CS and US. Furthermore, these results suggest that ASIC1A plays a critical, yet nuanced role in Pavlovian conditioning. More research will be needed to deconstruct the roles of ASIC1A in these fundamental forms of learning and memory.

Keywords: ASIC1A, Learning and Memory, Pavlovian, Reward

Graphical Abstract

graphic file with name nihms-1055939-f0008.jpg

Introduction

Classical Pavlovian conditioning is a fundamental form of learning and memory, during which an association is made between at least two previously unrelated stimuli, the conditional stimulus (CS) and unconditional stimulus (US) (Pavlov, 1927). The CS is neutral and does not evoke a response on its own. The US in contrast, evokes a response, the unconditional response (UR), independent of previous learning and can be either inherently aversive (e.g., foot shock) or rewarding (e.g., drug or food). With repeated CS-US pairings, a Pavlovian association is learned such that the CS acquires the ability to produce a response on its own, the conditional response (CR), that may resemble the UR (Cardinal et al., 2002).

Accumulating data suggest that Pavlovian conditioning depends on acid-sensing ion channel-1A (ASIC1A), a member of the degenerin-epithelial Na+ channel (DEG-ENaC) family (Waldmann et al., 1997). ASIC1A is intrinsically pH-sensitive, permeable to cations including Na+ and Ca2+, and required for currents evoked by extracellular acidosis in central nervous system neurons (Wemmie et al., 2003, Wemmie et al., 2002). ASIC1A is expressed throughout the brain, including areas implicated in Pavlovian learning and memory such as the amygdala (Du et al., 2014, Coryell et al., 2008, Wemmie et al., 2003, Ziemann et al., 2009), bed nucleus of the stria terminalis (BNST) (Taugher et al., 2014, Taugher et al., 2015), hippocampus (Wemmie et al., 2002, Liu et al., 2016), insula (Li et al., 2016), and nucleus accumbens (NAc) (Kreple et al., 2014). In addition, ASIC1A has been localized to dendritic spines, and implicated in synaptic plasticity and neurotransmission mediated by protons released from neurotransmitter containing vesicles (Du et al., 2017, Kreple et al., 2014, Zha et al., 2006, Wemmie et al., 2002, Du et al., 2014). Effects of ASIC1A disruption on dendritic spine density and glutamate receptor composition and function have been reported (Kreple et al., 2014, Du et al., 2014, Zha et al., 2006, Gonzalez-Inchauspe et al., 2017), as well as effects on long-term potentiation (Chiang et al., 2015, Wemmie et al., 2002, Du et al., 2014) and long-term depression (Li et al., 2016). Thus, through its location and function, ASIC1A may be well positioned to influence Pavlovian conditioning.

Consistent with the above observations, effects of ASIC1A on Pavlovian conditioning to aversive stimuli have been described. Disrupting ASIC1A in mice reduced cued and contextual fear conditioning (Wemmie et al., 2004, Wemmie et al., 2003, Taugher et al., 2017, Price et al., 2014, Coryell et al., 2008, Chiang et al., 2015). These deficits were reproduced with amygdala-specific ASIC1A disruption (Chiang et al., 2015) and a deficit in contextual fear memory was rescued by expressing ASIC1A in the amygdala (Coryell et al., 2009). In addition, ASIC1A disruption impaired eyeblink conditioning (Wemmie et al., 2002), and deficits in conditioned place avoidance (CPA) to carbon dioxide (CO2) (Taugher et al., 2014), a paradigm in which mice learn to avoid a context paired with CO2, an aversive US. In contrast, Asic1a−/− mice acquired conditioned taste aversion normally, although extinction was impaired (Li et al., 2016). Together, these findings suggest an important role for ASIC1A in Pavlovian conditioning to aversive stimuli.

Considering these findings, it is conceivable that ASIC1A may also play an important role in Pavlovian conditioning to rewarding stimuli. Supporting this possibility, ASIC1A is abundant in reward-related brain areas such as the NAc, and recent work indicates differential locomotor sensitivity to the effects of cocaine (Jiang et al., 2013) and enhanced conditioned place preference (CPP) to cocaine and morphine in mice lacking ASIC1A (Kreple et al., 2014). Similarly, selective ASIC1A deletion in the NAc increases cocaine CPP, and restoring ASIC1A to the NAc of Asic1a−/− mice returns cocaine CPP to wild-type levels (Kreple et al., 2014). These results suggest that ASIC1A reduces conditioning to drug rewards. Importantly, these effects contrast sharply with the apparent ability of ASIC1A to promote aversive conditioning (Wemmie et al., 2004, Wemmie et al., 2003, Taugher et al., 2017, Price et al., 2014, Coryell et al., 2008, Chiang et al., 2015). Thus, ASIC1A may have differential effects depending on whether the US is rewarding or aversive, although more study is needed to test this possibility.

Here we used a non-drug reward to test the role of ASIC1A in Pavlovian conditioning. Given the effects of ASIC1A on conditioning to drugs of abuse, we hypothesized that loss of ASIC1A would enhance conditioning to a non-drug reward. However, our results suggest that this is not necessarily the case, as loss of ASIC1A either impaired or enhanced conditioning depending on the timing of the relationship between the CS and US.

Materials and Methods:

Mice.

Asic1a−/− mice were originally generated as described (Wemmie et al., 2002). Both the Asic1a−/− and Asic1a+/+ mice were generated from homozygous parents which are maintained on a congenic C57BL/6J background (Taugher et al., 2017). The lines have been backcrossed to standard C57BL/6J mice from JAX greater than 10 times, and are routinely backcrossed to this strain to maintain congenicity. Mice were housed in groups of 2–5 mice and kept on a 12-hour light-dark cycle, with all experiments occurring during the light cycle. Mice were fed standard chow and water ad libitum. All experimental groups were age (12–16 weeks) and sex-matched. Sample sizes are presented in figure legends and ranged from 20 to 26 mice per group. Animal care met the standards set by the National Institutes of Health, and all experiments were approved by the University of Iowa Animal Care and Use Committee.

Pavlovian conditioning apparatus.

Pavlovian conditioning experiments were conducted in Med-Associates operant chambers (21.6×17.8× 12.7 cm; Med Associates, St. Albans, VT) contained within sound attenuating cubicles (Med Associates, St. Albans, VT). Operant chambers were equipped with rod floors, two plexi-glass side walls, front and back metal walls, and a liquid dipper arm that delivered approximately 10 μl of vanilla-flavored Ensure (diluted in water to a 1:1 ratio) from a reservoir. On the back wall, there was a speaker (65db, 2900 Hz). A house light was placed outside of the left Plexiglas side wall, providing light to the chamber. In the middle of the front wall was a food hole flanked on either side by lever slits, above which cue lights were mounted. For all experiments described here, the CS was a tone and cue light. The CS was always 5 seconds in duration except for experiment 4, when it was 0.5 seconds. The US was the Ensure solution presented for 15 seconds. The food hole was fitted with infrared photobeam sensors that counted the number and time of head entries into the food hole. 24 hours before experiments began, mice were given the Ensure solution for reward habituation and to prevent neophobia during the experiment.

Chamber acclimatization.

On the first day of each experimental paradigm, mice were acclimated to the operant chambers for 10 minutes, with the house light off, before beginning the session. At the 10-minute mark, the Med-Associates program was initiated with the illumination of the house light signaling the beginning of the experimental session. The 10-minute acclimation period was excluded during subsequent sessions.

Pavlovian conditioning.

A separate cohort of mice was used for each experiment. Asic1a+/+ and Asic1a−/− mice underwent a 45-minute training session per day consisting of 20 CS-US pairings over the course of eight experimental sessions. The temporal relationship between the CS and US were varied between experiments. The inter-trial interval (ITI) between CS-US pairings varied, with an average ITI of 2 minutes. 24 hours after the final training session mice underwent an extinction probe test during which the experimental parameters were the same as training except that the CS was presented in the absence of the US.

Behavioral measures.

Head entries were recorded by MED-PC software (Med Associates, St. Albans, VT). The number of CS trials in which at least one head entry was recorded were quantified for each session of each experiment. Conditioned response latencies were assessed as the time to the first head entry after a CS presentation. Latencies were averaged across all 20 trials in each session. CS trials in which no head entry occurred were given a default latency of 5 seconds. During training, conditioned response (CR) scores were calculated by subtracting the head entry rate during the ITI from the head entry rate during the CS, adapted from (Darvas et al., 2014). For the extinction test, CR scores were calculated by subtracting the head entry rate during 15 seconds of ITI immediately preceding the CS from the head entry rate during the 15 seconds following CS onset (the time period during training when the US would have been present).

Statistical analysis.

All values are plotted as mean ± SEM. For all tests, p < 0.05 was considered significant. One-way analysis of variance (ANOVA) was used to assess differences in head entry responses over time for a given genotype. Two-way ANOVA was used to test for interactions between genotype and time, or genotype and sex. When significant effects were observed in the context of the full ANOVA, post hoc Fisher’s LSD was used to identify differences between genotypes during individual sessions. For the extinction experiment, unpaired t-test was used to compare average CR scores between genotypes and Wilcoxon signed rank test was used to determine if group medians differed significantly from 0. All analyses were performed using GraphPad Prism (version 7) software.

Results

ASIC1A disruption impairs conditioning to non-drug reward

To test the hypothesis that ASIC1A plays an important role in Pavlovian reward conditioning, we started with a task similar to that described previously by others (Darvas et al., 2014, Holland, 1977, Holland, 1980) in which the conditional stimulus (CS) preceded the unconditional stimulus (US), a highly palatable liquid food reward (vanilla-flavored Ensure) (Fig. 1A). We compared Asic1a+/+ and Asic1a−/− mice across sessions consisting of 20 CS-US pairings per session for eight experimental sessions. We assessed head entries during the CS and found a statistically significant genotype by time interaction (Fig. 1B) (F(7,287) = 2.324, p = 0.0255). Furthermore, this effect was driven by the increased head entries in the Asic1a+/+ group (F(7, 140) = 3.042, p = 0.0052) but not the Asic1a−/− group, which failed to increase head entries over the eight training sessions (F(7, 147) = 1.168, p = 0.3324). To rule out whether the difference between genotypes was due to multiple head entries per trial, we also quantified the number of trials with a head entry (Fig. 1C), and continued to observe a statistically significant genotype by time interaction (F(7, 287) = 3.274, p = 0.0023). We also quantified conditioned response latency, defined as time from CS onset to the first subsequent head entry, and found conditioned response latencies mirrored the head entry results, such that the Asic1a+/+ mice developed significantly shorter latencies than Asic1a−/− mice across training sessions (Fig. 1D) (F(7,287) = 3.644, p = 0.0009).

Figure 1:

Figure 1:

Open in a new tab

Experiment 1. (A) Sequential pairing of CS and US. (B) Comparison of head entries during the CS over 8 experimental sessions in Asic1a+/+ and Asic1a−/− mice. Asic1a−/− mice exhibit significantly less conditioned head entries than Asic1a+/+ mice. (C) Comparison of CS trials with a head entry. Asic1a−/− mice exhibit significantly less CS trials with a head entry than Asic1a+/+ mice. (D) Comparison of conditioned response latency. Asic1a−/− mice exhibit significantly higher conditioned response latency than Asic1a+/+ mice. (E) Comparison of head entries during the US. Asic1a+/+ and Asic1a−/− mice exhibit a similar increase in head entries. (F) Head entry distributions during the CS and US in sessions 1 and 8. (Fishers LSD, *p < 0.05, **p < 0.001; n = 21 Asic1a+/+, 22 Asic1a−/−).

We next examined whether sex affected the outcome, and found no effect. For example, during the eighth experimental session there was an effect of genotype (F(1,39) = 4.882, p = 0.0331) but no effect of sex (F(1,39) = 0.1622, p = 0.6893), and no interaction F(1,39) = 0.665, p = 0.4198). Similar analyses of sex were conducted for all subsequent experiments, and no effect was observed (data not shown).

Because the CS signaled upcoming reward availability, we further assessed whether the differences in head entries during the CS translated into differences in head entries when the reward was actually available (Fig. 1E). There was an effect of genotype (F(1,41) = 4.829, p = 0.0337) and an effect of time (F(7,287) = 5.918, p < 0.0001), but no interaction (F(7,287) = 0.8317, p = 0.5617). Despite the lower number of head entries exhibited by the Asic1a−/− mice during the US, the increase in head entries over time suggests that some learning occurred.

To more closely examine the behavioral changes that developed, we assessed the distribution of head entries across one-second intervals spanning the CS and US, averaged for all 20 trials per session, for sessions 1 and 8 (Fig. 1F). Both genotypes increased head entries during the CS-US presentations in session 8 compared to session 1. In addition, in session 8 there was a noticeable deficit in head entries in the Asic1a−/− mice during the CS, and a delay in the peak head entry response. Taken together, these data suggest that disrupting ASIC1A altered the acquisition of CS-evoked responses.

Asic1a−/− mice exhibit greater responses when CS and US co-initiate and overlap

To test whether the behavioral differences in the Asic1a−/− mice might be sensitive to the temporal relationship between the CS and US, we performed a second experiment with a new cohort of mice, in which we altered the timing of the CS and US presentations. In this paradigm, the CS onset signaled that the reward was immediately available (Fig. 2A), and hence minimized the delay between the start of the CS and the ability to sample the US. Otherwise, this second experiment was identical to the first. Over 8 training sessions, we found that both genotypes increased head entries during the CS (Fig. 2B) (Asic1a+/+, F(7,182) = 15.2, p < 0.0001; Asic1a−/−, F(7,168) = 30.77, p < 0.0001). Moreover, the Asic1a−/− mice exhibited more head entries than the Asic1a+/+ mice (genotype by time interaction, F(7,350) = 2.83, p = 0.0070). Similar results were observed when we quantified CS trials with a head entry (Fig. 2C) (F(7, 350) = 4.737, p < 0.0001) and CS latency (Fig. 2D) (F(7,350) = 3.191, p = 0.0027). We also examined head entries after the CS, but while the US was still available (10 seconds) and found that head entries did not differ between genotypes (Fig. 2E). There was an effect of time (F(7,350) = 7.068, p < 0.0001) but no effect of genotype (F(1,50) = 1.331, p = 0.2541) and no interaction (F(7,350) = 0.8918, p = 0.5130). The observations above were supported by head entry distributions during CS-US presentations (Fig. 2F). Therefore, in contrast to the first experiment, disrupting ASIC1A improved CS-evoked responses in experiment 2. These findings suggest that ASIC1A disruption can have differing effects that depend on the temporal relationship between the CS and US.

Figure 2:

Figure 2:

Open in a new tab

Experiment 2. (A) Co-initiating and overlapping CS and US. (B) Comparison of head entries during the CS over 8 experimental sessions in Asic1a+/+ and Asic1a−/− mice. Asic1a−/− mice exhibit significantly more conditioned head entries than Asic1a+/+ mice. (C) Comparison of CS trials with a head entry. Asic1a−/− mice exhibit significantly more CS trials with a head entry than Asic1a+/+ mice. (D) Comparison of conditioned response latency. Asic1a−/− mice exhibit a significant decrease in conditioned response latency than Asic1a+/+ mice. (E) Comparison of head entries during the last 10 seconds of the US. Asic1a+/+ and Asic1a−/− mice exhibit a similar increase in head entries. (F) Head entry distributions during the CS and US in sessions 1 and 8. (Fishers LSD, *p < 0.05, **p < 0.001; n = 26 Asic1a+/+, 26 Asic1a−/−).

Sensitivity to CS-US relationship

The opposing effects of ASIC1A disruption between the first and second experiments led us to ask what aspects of the CS-US relationship might be responsible. Two features of the second experiment differed from the first. One difference was that in the first experiment there was no overlap between the CS and US, whereas in the second experiment the CS and US overlapped in time. Another difference was that in the second experiment the CS and US were co-initiated. Therefore, to test whether the overlap or co-initiation of the CS and US mattered, we performed two additional experiments, referred to as experiments 3 and 4. In experiment 3, the CS preceded the US by 2.5 seconds and overlapped with the US by 2.5 seconds (Fig. 3A). In experiment 4, the CS was shortened to 0.5 secs and was co-initiated with the US (Fig. 4A). Other aspects of experiments 3 and 4 were identical to experiments 1 and 2, including the duration of the US. We hypothesized that either the overlap or co-initiation would produce an effect similar to experiment 2 and lead to more head entries in the Asic1a−/− mice.

Figure 3:

Figure 3:

Open in a new tab

Experiment 3. (A) Overlapping CS and US (B) Comparison of head entries during the CS over 8 experimental sessions in Asic1a+/+ and Asic1a−/− mice. Asic1a+/+ and Asic1a−/− mice exhibit a similar increase in head entries. (C) Comparison of CS trials with a head entry. Asic1a+/+ and Asic1a−/− mice exhibit a similar increase in CS trials with a head entry. (D) Comparison of conditioned response latency. Asic1a+/+ and Asic1a−/− mice exhibit a similar decrease in conditioned response latency. (E) Comparison of head entries during the last 12.5 seconds of the US. Asic1a+/+ and Asic1a−/− mice exhibit a similar increase in head entries. (F) Head entry distributions during the CS and US in sessions 1 and 8. (n = 27 Asic1a+/+, 25 Asic1a−/−).

Figure 4:

Figure 4:

Open in a new tab

Experiment 4. (A) Short CS co-initiating with US. (B) Comparison of head entries during the CS and the next 4.5 seconds across experimental sessions in Asic1a+/+ and Asic1a−/− mice. Asic1a+/+ and Asic1a−/− mice exhibit a similar increase in head entries. (C) Comparison of CS trials with a head entry. Asic1a+/+ and Asic1a−/− mice exhibit an increase in CS trials with a head entry. (D) Comparison of conditioned response latency. Asic1a+/+ and Asic1a−/− mice exhibit a similar decrease in conditioned response latency. (E) Comparison of head entries during the last 10 seconds of the US. Both Asic1a+/+ and Asic1a−/− mice exhibit an increase in head entries across training days. (F) Head entry distributions during the CS and US in sessions 1 and 8 (Fishers LSD, **p < 0.001; n = 23 Asic1a+/+, 20 Asic1a−/−).

In experiment 3, we found evidence of conditioning in both Asic1a+/+ and Asic1a−/− mice, suggested by the increase in head entries during the CS with training (Fig. 3B) (Asic1a+/+ mice, F(7,154) = 8.573, p < 0.0001; Asic1a−/− mice, F(7,133) = 6.321, p < 0.0001). In addition, while there was an effect of time (F(1,287) = 13.21, p < 0.0001), there was no effect of genotype (F(1,41) = 2.526, p = 0.1196) and no interaction (F(1,287) = 1.583, p = 0.1400). Similar outcomes were observed upon quantification of CS trials with a head entry (Fig. 3C) (time F(7, 287) = 25.6, p < 0.0001; genotype F(1,41) = 0.1108, p = 0.7409; interaction F(7, 287) = 0.8823, p = 0.5207) and CS latency (Fig. 3D) (time F(7, 287) = 19.38, p < 0.0001; genotype F(1,41) = 1.542, p = 0.2214; interaction F(7,287) = 1.24, p = 0.2809). As for head entries during the US (Fig. 3E), there was an effect of time (F(7,287) = 14.32, p < 0.0001) but no effect of genotype (F(1,41) = 0.2271, p = 0.6362) and no interaction (F(1,287) = 1.249, p = 0.2758). Head entry distribution across CS-US presentations (Fig. 3F) paralleled the results in Figures 3B, C, and E and suggested similar learning in both genotypes. Together, these data suggest that overlapping the CS and US resulted in at least some conditioning in the Asic1a−/− mice. However, overlapping the CS and US was not sufficient to reproduce the effects in experiment 2, in which the Asic1a−/− mice exhibited significantly more head entries than wild-types during the CS.

In experiment 4, in which a short CS was co-initiated with the US, we assessed head entries during the 0.5 sec CS plus the following 4.5 sec, and also found evidence of conditioning to the CS in both groups (Fig. 4B) (Asic1a+/+ mice, F(7,119) = 4.737, p < 0.0001; Asic1a−/− mice, F(7,133) = 4.712, p < 0.0001). However, while there was an effect of time (F(7,252) = 8.819, p < 0.0001), there was no effect of genotype (F(1,36) = 0.2541, p = 0.6173) and no interaction (F(7,252) = 0.5431, p = 0.8013). Similar outcomes were observed upon quantification of CS trials with a head entry (Fig. 4C) (time F(7,252) = 12.16, p < 0.0001; genotype F(1,36) = 0.2039, p = 0.6543; interaction F(7,252) = 0.3724, p = 0.9179) and CS latency (Fig. 4D) (time F(7,252) = 10.45, p < 0.0001; genotype F(1,36) = 0.3303, p = 0.5691; interaction F(7,252) = 0.358, p = 0.9257). When we assessed head entries during the US, we found a significant genotype by time interaction (F(1,252) = 2.985, p = 0.0050) (Fig. 4E). The timing of head entry distribution during CS-US presentations for sessions 1 and 8 is also shown (Fig. 4F). Together, the results of experiment 4 suggest that co-initiating the CS and US by itself was also not sufficient to reproduce the elevated CS responses in the Asic1a−/− mice observed in experiment 2. Thus, in experiment 2 both co-initiation and overlap of the CS and US may have combined to produce the exaggerated CS-associated head entries in the Asic1a−/− mice.

Dependence on CS-US pairing

The results thus far led us to ask to what degree the previous outcomes depended on learning the association between the CS and US. Therefore, to further assess the role of ASIC1A in Pavlovian reward conditioning we performed a final experiment in which the CS and US were explicitly unpaired (Fig. 5A). Here, the CS never immediately preceded or overlapped with reward delivery, and thus did not predict that a reward was available. We hypothesized that if the results of the previous experiments reflected conditioning, then responses to the CS should not increase over time when the CS and US are explicitly unpaired. Over eight experimental sessions, we assessed head entries during the CS (Fig. 5B) and found that neither genotype increased head entry responses over time (Asic1a+/+ mice, F(7,147) = 1.69, p = 0.1156; Asic1a−/− mice, F(7,168) = 0.5737, p = 0.7766). Similarly, no improvement with training was observed in the Asic1a+/+ or Asic1a−/− mice in CS trials with a head entry (Fig. 5C) (F(7,315) = 1.542, p = 0.1524) or CS latency (Fig. 5D) (F(7,315) = 1.646, p = 0.1220). Head entry distribution during the CS presentations also suggested no conditioning to the unreinforced CS (Fig. 5E). Together these observations suggest that in the previous experiments the behavioral responses during the CS did indeed reflect Pavlovian conditioning.

Figure 5:

Figure 5:

Open in a new tab

Experiment 5. (A) Explicitly unpaired paradigm. (B) Comparison of head entries during the CS across experimental sessions in Asic1a+/+ and Asic1a−/−. Neither Asic1a+/+ nor Asic1a−/− mice exhibit an increase in conditioned head entries. (C) Comparison of CS trials with a head entry. Neither Asic1a+/+ nor Asic1a−/− mice exhibit an increase in CS trials with a head entry. (D) Comparison of conditioned response latency. Neither Asic1a+/+ nor Asic1a−/− mice exhibit a decrease in conditioned response latency. (E) Head entry distributions during the CS for sessions 1and 8. (F) Comparison of head entries during the first 5 seconds of the US. Asic1a+/+ and Asic1a−/− mice exhibit an increase in head entries. (G) Comparison of head entries during the entire 15 seconds of the US. Asic1a+/+ and Asic1a−/− mice exhibit an increase in head entries. (H) Head entry distributions during the US for sessions 1 and 8 (Fishers LSD, *p < 0.05; n = 22 Asic1a+/+, 25 Asic1a−/−).

Interestingly when we analyzed head entries during the US (Fig. 5 F, G, H), there was an effect of time in both genotypes (Fig. F: Asic1a+/+ mice F(7,147) = 13.31, p < 0.0001; Asic1a−/− mice F(7,168) = 13.79, p < 0.0001) (Fig G: Asic1a+/+ mice F(7,147) = 7.186, p < 0.0001; Asic1a−/− mice F(7,168) = 11.31, p < 0.0001). Thus, despite the absence of conditioning to the CS, both groups of mice still acquired an improved ability to locate the reward across training sessions, suggesting the animals adopted an alternative learning strategy, independent of the explicit cues. Of note, there was a genotype by time interaction when we quantified the responses during the first 5 seconds of the US (Fig. 5F) (F(7,315) = 2.526, p = 0.0153) and when we quantified responses during the entire 15 seconds of the US (Fig. 5G) (F(7,315) = 2.505, p = 0.0161), which was largely due to increased head entries in the Asic1a−/− mice during sessions 4 through 6. This observation might suggest US-associated learning also depended on ASIC1A, although by the end of training (session 8) the two genotypes did not differ.

Conditioned response scores

To account for non-specific head entries during the ITI, for each experiment we also assessed conditioned response (CR) score, adapted from (Darvas et al., 2014). Using this score (Figs. 6A, B), the results closely paralleled CS-associated head-entries (Figs 1–5). Figure 6 further illustrates the uniqueness of experiment 2, in which the CS and US were co-initiated and overlapped for 5 seconds; this paradigm produced the greatest increase in CR score in both genotypes, as observed by comparing CR scores for experiment 2 to the experiment with next highest CR score (Fig. 6A, B) (Asic1a+/+ mice, experiment 2 vs experiment 1, F(1,46) = 12.1, p = 0.0011; Asic1a−/− mice, experiment 2 vs experiment 3, F(1,43) = 61.67, p < 0.0001). Moreover, these data further illustrate the difference in experiment 2 between the Asic1a−/− mice and their Asic1a+/+ counterparts (F(7,350) = 3.245, p = 0.0024).

Figure 6:

Figure 6:

Open in a new tab

Analysis of CR scores. (A) CR scores of Asic1a+/+ mice. CR scores were greatest in experiment 2. (B) CR scores of Asic1a−/− mice. CR scores were greatest in experiment 2. Experiments 1 through 5 indicated by symbols.

Conditioned responses in the absence of reinforcement

To further assess conditioning, we quantified CR scores when CS presentations were not reinforced with a US during a single extinction session 24 hours after the final training session (Fig. 7). In the absence of reinforcement, the head entries evoked by the CS provided a measure of conditioning, although concurrent extinction also likely reduced conditioned responses. Evidence of Pavlovian conditioning was regarded as a mean CR score significantly greater than zero. In all five experiments, only 3 groups met this criterion for conditioning (Fig. 7A); these consisted of the Asic1a+/+ mice in experiment 1 (Wilcoxon signed rank test, W = 128, p = 0.0241), the Asic1a+/+ mice in experiment 2 (W = 297, p < 0.0001), and the Asic1a−/− mice in experiment 2 (W = 331, p < 0.0001). As with the other measures of conditioning presented in the preceding figures, the greatest CR scores were exhibited by the Asic1a−/− mice in experiment 2, which were significantly greater than their Asic1a+/+counterparts (t(51) = 2.235, p = 0.0229). For a more granular perspective, we quantified average distribution of head entries for all 20 unreinforced CS trials for experiments 1, 2, and 5 (Fig. 7B) and CS-evoked responses versus trial number (Fig. 7C). In both of these figures, experiment 2 stood out for producing the most striking CS-evoked responses and for elevated CS-evoked head entries in the Asic1a−/− mice. Extinction is harder to discern because of variability in head entry responses across trials (Fig. 7C).

Figure 7:

Figure 7:

Open in a new tab

CR scores and head entries with the absence of reinforcement during a single extinction session. (A) Average CR scores for each genotype are presented for each experiment, indicated along X-axis. CR scores statistically different from zero were indicated by #. Asic1a−/− mice in experiment 2 exhibited a significantly greater average CR score than the Asic1a+/+ mice in experiment 2 (indicated by *). (B) Head entry distributions averaged for all 20 unreinforced CS trials for experiments 1, 2, and 5. (C) CS-evoked responses versus trial number for experiments 1, 2, and 5.

Discussion

To our knowledge these experiments are the first to assess the role of ASIC1A in Pavlovian reward conditioning to a non-drug reward. The results are consistent with the hypothesis that ASIC1A plays a role in reward conditioning. However, the effects of ASIC1A disruption were more nuanced than initially anticipated. Consistent with previous work in which Asic1a−/− mice exhibited increases in conditioning in some paradigms (cocaine CPP), yet decreases in conditioning in others (fear conditioning), here we found evidence of both impairment and enhancement. Interestingly, in these experiments the role played by ASIC1A was sensitive to relatively subtle variations in the temporal relationship between the CS and US. In experiment 1, when the CS preceded the US, Asic1a−/− mice showed reduced head entry responses compared to Asic1a+/+ mice, suggesting that the Asic1a−/− mice did not learn the CS-US relationship normally. In contrast, in experiment 2, when the CS co-initiated with the US and signaled that the reward was immediately available, the Asic1a−/− mice had more head entries, suggesting that their conditioning was enhanced. To test whether the timing of the CS and US mattered, we further manipulated the temporal relationship between the CS and US in experiments 3 and 4, and in these paradigms, we observed no apparent differences between the Asic1a+/+ and Asic1a−/− mice. Taken together, these experiments suggest that the loss of ASIC1A may provide advantages in certain reward learning situations but may also confer disadvantages in other situations, such as when there is no overlap between the CS and US.

To confirm that the responses observed during the CS reflected conditioning, experiment 5 tested explicit unpairing of the CS and US, and thus provided an important control for experiments 1 through 4. As expected, in experiment 5 head entries did not increase over time in response to the unpaired CS. However head entries during the US did increase over time, suggesting that in the absence of explicit cues the mice adopted an alternative learning strategy. Unfortunately, our experimental design did not allow us to identify how the animals acquired the ability to find the reward. We suspect the mice may have used implicit cues, such as the odor of the Ensure reward, which we could not eliminate. A recent paper suggests ASIC1A disruption may impair olfaction (Vann and Xiong, 2018), however we have previously found no differences of Asic1a−/− mice compared to matched Asic1a+/+ mice in the ability to locate buried food by its odor, to detect scent from mice of the opposite sex, and to avoid the aversive odor of butyric acid (Coryell et al., 2007). Regardless of these considerations of olfaction, the learning observed in the unpaired paradigm peaked earlier in the Asic1a−/− mice, which by itself is an interesting observation as it suggests that this alternative learning might also be sensitive to ASIC1A-disruption. Importantly, the behavioral responses to the unpaired US in experiment 5 did not reach the high levels observed in experiment 2 when the US overlapped and co-initiated with the CS, suggesting that the results depended on the CS-US pairing.

One advantage of our approach was that the vanilla-flavored Ensure produced sufficient motivation to support Pavlovian conditioning without food or water restriction, which would have altered appetitive drive. Because the CS-associated responses in the Asic1a−/− mice were reduced, increased, or normal depending on the paradigm, it is unlikely that the differences seen here were due to abnormalities in motivation. Supporting comparable appetitive drive for food and water, the average weight of Asic1a−/− mice does not differ from Asic1a+/+ mice (Kreple et al., 2014). In addition, because the CS and US remained the same, and the experiments differed primarily in the timing of the CS and US presentation, it is unlikely that observed behavioral differences were due to sensory dysfunction, inability to detect the CS, or an inability to express head entries. Furthermore, we have previously found hearing to be grossly normal in the Asic1a−/− mice (Coryell et al., 2007), whereas others have not observed vision-based abnormalities that would disrupt detection of visual cues in our experiments (Render et al., 2010).

It is interesting to consider how the present results might relate to our previous findings with Asic1a−/− mice in Pavlovian fear conditioning and conditioned place preference. The results here suggest that altering the timing of the CS-US relationship might influence the outcomes in those other paradigms. However in those conditioning paradigms, the nature of the CS and US was very different than those studied here. For example, in fear conditioning, the CS was 20 seconds (auditory tone) or continuous (context), and the footshock (US) was only 1 second (Wemmie et al., 2004, Wemmie et al., 2003, Taugher et al., 2017). Also, with conditioned place preference the CS (context) was continuous and overlapped with the US (drug reward), although the temporal relationship between the CS and US was not as precise as the relationship here (Kreple et al., 2014). Nevertheless the results in this manuscript suggest it may be informative to vary the CS-US relationships in those other aversive and appetitive conditioning paradigms.

Although we cannot presently rule out altered brain development as the cause of the behavioral differences observed here, we have previously found that brain morphology and structure are grossly normal in Asic1a−/− mice (Wemmie et al., 2003, Wemmie et al., 2002). Moreover, restoring ASIC1A expression to specific brain sites in adult Asic1a−/− mice normalizes several behaviors and physiological measures, including cocaine CPP and AMPA-to-NMDA receptor ratio (Kreple et al., 2014). Similarly, disrupting or inhibiting ASIC1A in adult mice produces effects resembling those observed in the Asic1a−/− mice in multiple behaviors (Kreple et al., 2014) (Taugher et al., 2014, Taugher et al., 2015, Dwyer et al., 2009). Thus, ASIC1A has important effects on behavior and brain function that are independent of brain development.

It is not yet clear where in the brain ASIC1A acts to affect the behaviors studied here. Candidate sites include the amygdala and NAc, which are implicated in various features of Pavlovian learning (Wassum and Izquierdo, 2015, Day and Carelli, 2007, Parkinson et al., 1999) and where ASIC1A plays a role in Pavlovian conditioning (Taugher et al., 2017, Wemmie et al., 2004, Wemmie et al., 2003, Wemmie et al., 2002). Additional candidate sites include the lateral hypothalamus and insula, as these sites express ASIC1A (Song et al., 2012, Li et al., 2016) and are implicated in the acquisition and expression of conditioned head entry behaviors (Sharpe et al., 2017, Kusumoto-Yoshida et al., 2015). Additional work will be needed to determine precise sites of ASIC1A action in Pavlovian reward conditioning.

As for subcellular sites of action, we suspect that the synapse is the best candidate in the behaviors studied here. Recently, a portion of the excitatory post-synaptic current (EPSC) was found to depend on ASIC1A. This cation current is sensitive to ASIC antagonists, pH buffering capacity, and carbonic anhydrase inhibitors, and has been reported in the NAc (Kreple et al., 2014), lateral amygdala (Du et al., 2014), and Calyx of Held (Gonzalez-Inchauspe et al., 2017). Protons released from synaptic vesicles during neurotransmission (Du et al., 2014, Sinning and Hubner, 2013) provide the most-likely mechanism for activating ASIC1A during the EPSC. Perhaps related to the ASIC1A-dependent synaptic current, we found a number of synaptic abnormalities in NAc medium spiny neurons following ASIC1A disruption including increases in: AMPA to NMDA receptor ratio, AMPA receptor rectification index, miniature EPSC frequency, and dendritic spine density (Kreple et al., 2014). Also consistent with a synaptic role for ASIC1A, loss or inhibition of ASIC1A reduces long-term potentiation in the amygdala (Du et al., 2014, Chiang et al., 2015) and hippocampus (Wemmie et al., 2002, Liu et al., 2016), (but see (Wu et al., 2013)). Loss of ASIC1A is also implicated in long-term depression in the insular cortex (Li et al., 2016). Thus, ASIC1A is implicated in both synaptic potentiation and depression in multiple brain areas, which might help explain the diverse effects of ASIC1A disruption observed in these studies.

In conclusion, more needs to be learned about how ASIC1A contributes to brain function and behavior. It will be valuable to discern how disrupting ASIC1A can reduce some types of learning and memory, yet enhance others. Because the structure and function of these channels are well conserved between species, including humans (Li et al., 2010), we expect that ASIC1A plays an important role in the human brain. Thus far, the human Asic1a gene has been implicated in panic disorder in studies with limited sample sizes (Smoller et al., 2014, Leibold et al., 2017) but the importance of Pavlovian conditioning in human disorders (Romaniuk et al., 2010, Rabinak et al., 2017, Meyer et al., 2016, Lynch et al., 1973) and the effects of ASIC1A on Pavlovian conditioning in mice suggests that ASIC1A will ultimately be shown to play roles in a variety of psychiatric illnesses.

Acknowledgements:

R.J.T. and S.M.A. were supported by (NIMH 5T32MH019113). R.T.L. was supported by NIDA 5R01DA037216. J.A.W. was supported by the Department of Veterans Affairs (Merit Award), the NIMH (5R01MH085724), NHLBI (R01HL113863), (NIDA 5R01DA037216) and a NARSAD Independent Investigator Award. We thank John Freeman, Edward Wasserman, Margaret Fuller, Subhash Gupta, Gail Harmata, and Anne J. Ledtje for their constructive feedback, and Leah VanDenBosch for her assistance.

Footnotes

Conflict of Interest: The authors declare no competing financial interests.

References

  1. CARDINAL RN, PARKINSON JA, HALL J & EVERITT BJ 2002. Emotion and motivation: the role of the amygdala, ventral striatum, and prefrontal cortex. Neurosci Biobehav Rev, 26, 321–52. [DOI] [PubMed] [Google Scholar]
  2. CHIANG PH, CHIEN TC, CHEN CC, YANAGAWA Y & LIEN CC 2015. ASIC-dependent LTP at multiple glutamatergic synapses in amygdala network is required for fear memory. Scientific Reports, 5. [DOI] [PMC free article] [PubMed] [Google Scholar]
  3. CORYELL MW, WUNSCH AM, HAENFLER JM, ALLEN JE, MCBRIDE JL, DAVIDSON BL & WEMMIE JA 2008. Restoring Acid-sensing ion channel-1a in the amygdala of knock-out mice rescues fear memory but not unconditioned fear responses. J Neurosci, 28, 13738–41. [DOI] [PMC free article] [PubMed] [Google Scholar]
  4. CORYELL MW, WUNSCH AM, HAENFLER JM, ALLEN JE, SCHNIZLER M, ZIEMANN AE, COOK MN, DUNNING JP, PRICE MP, RAINIER JD, LIU Z, LIGHT AR, LANGBEHN DR & WEMMIE JA 2009. Acid-sensing ion channel-1a in the amygdala, a novel therapeutic target in depression-related behavior. J Neurosci, 29, 5381–8. [DOI] [PMC free article] [PubMed] [Google Scholar]
  5. CORYELL MW, ZIEMANN AE, WESTMORELAND PJ, HAENFLER JM, KURJAKOVIC Z, ZHA XM, PRICE M, SCHNIZLER MK & WEMMIE JA 2007. Targeting ASIC1a reduces innate fear and alters neuronal activity in the fear circuit. Biol Psychiatry, 62, 1140–8. [DOI] [PubMed] [Google Scholar]
  6. DARVAS M, WUNSCH AM, GIBBS JT & PALMITER RD 2014. Dopamine dependency for acquisition and performance of Pavlovian conditioned response. Proc Natl Acad Sci U S A, 111, 2764–9. [DOI] [PMC free article] [PubMed] [Google Scholar]
  7. DAY JJ & CARELLI RM 2007. The nucleus accumbens and Pavlovian reward learning. Neuroscientist, 13, 148–59. [DOI] [PMC free article] [PubMed] [Google Scholar]
  8. DU J, PRICE MP, TAUGHER RJ, GRIGSBY D, ASH JJ, STARK AC, HOSSAIN SAAD MZ, SINGH K, MANDAL J, WEMMIE JA & WELSH MJ 2017. Transient acidosis while retrieving a fear-related memory enhances its lability. Elife, 6. [DOI] [PMC free article] [PubMed] [Google Scholar]
  9. DU JY, REZNIKOV LR, PRICE MP, ZHA XM, LU Y, MONINGER TO, WEMMIE JA & WELSH MJ 2014. Protons are a neurotransmitter that regulates synaptic plasticity in the lateral amygdala. Proceedings of the National Academy of Sciences of the United States of America, 111, 8961–8966. [DOI] [PMC free article] [PubMed] [Google Scholar]
  10. DWYER JM, RIZZO SJ, NEAL SJ, LIN Q, JOW F, ARIAS RL, ROSENZWEIG-LIPSON S, DUNLOP J & BEYER CE 2009. Acid sensing ion channel (ASIC) inhibitors exhibit anxiolytic-like activity in preclinical pharmacological models. Psychopharmacology (Berl), 203, 41–52. [DOI] [PubMed] [Google Scholar]
  11. GONZALEZ-INCHAUSPE C, URBANO FJ, DI GUILMI MN & UCHITEL OD 2017. Acid-Sensing Ion Channels Activated by Evoked Released Protons Modulate Synaptic Transmission at the Mouse Calyx of Held Synapse. Journal of Neuroscience, 37, 2589–2599. [DOI] [PMC free article] [PubMed] [Google Scholar]
  12. HOLLAND PC 1977. Conditioned stimulus as a determinant of the form of the Pavlovian conditioned response. Journal of Experimental Psychology: Animal Behavior Processes, 3, 77–104. [DOI] [PubMed] [Google Scholar]
  13. HOLLAND PC 1980. Cs-Us Interval as a Determinant of the Form of Pavlovian Appetitive Conditioned-Responses. Journal of Experimental Psychology-Animal Behavior Processes, 6, 155–174. [PubMed] [Google Scholar]
  14. JIANG Q, WANG CM, FIBUCH EE, WANG JQ & CHU XP 2013. Differential Regulation of Locomotor Activity to Acute and Chronic Cocaine Administration by Acid-Sensing Ion Channel 1a and 2 in Adult Mice. Neuroscience, 246, 170–178. [DOI] [PMC free article] [PubMed] [Google Scholar]
  15. KREPLE CJ, LU Y, TAUGHER RJ, SCHWAGER-GUTMAN AL, DU J, STUMP M, WANG Y, GHOBBEH A, FAN R, COSME CV, SOWERS LP, WELSH MJ, RADLEY JJ, LALUMIERE RT & WEMMIE JA 2014. Acid-sensing ion channels contribute to synaptic transmission and inhibit cocaine-evoked plasticity. Nat Neurosci, 17, 1083–91. [DOI] [PMC free article] [PubMed] [Google Scholar]
  16. KUSUMOTO-YOSHIDA I, LIU H, CHEN BT, FONTANINI A & BONCI A 2015. Central role for the insular cortex in mediating conditioned responses to anticipatory cues. Proc Natl Acad Sci U S A, 112, 1190–5. [DOI] [PMC free article] [PubMed] [Google Scholar]
  17. LEIBOLD NK, VAN DEN HOVE D, VIECHTBAUER W, KENIS G, GOOSSENS L, LANGE I, KNUTS I, SMEETS HJ, MYIN-GERMEYS I, STEINBUSCH HW & SCHRUERS KR 2017. Amiloride-sensitive cation channel 2 genotype affects the response to a carbon dioxide panic challenge. J Psychopharmacol, 31, 1294–1301. [DOI] [PubMed] [Google Scholar]
  18. LI M, INOUE K, BRANIGAN D, KRATZER E, HANSEN JC, CHEN JW, SIMON RP & XIONG ZG 2010. Acid-sensing ion channels in acidosis-induced injury of human brain neurons. J Cereb Blood Flow Metab, 30, 1247–60. [DOI] [PMC free article] [PubMed] [Google Scholar]
  19. LI WG, LIU MG, DENG SN, LIU YM, SHANG L, DING J, HSU TT, JIANG Q, LI Y, LI F, ZHU MX & XU TL 2016. ASIC1a regulates insular long-term depression and is required for the extinction of conditioned taste aversion. Nature Communications, 7. [DOI] [PMC free article] [PubMed] [Google Scholar]
  20. LIU MG, LI HS, LI WG, WU YJ, DENG SN, HUANG C, MAXIMYUK O, SUKACH V, KRISHTAL O, ZHU MX & XU TL 2016. Acid-sensing ion channel 1a contributes to hippocampal LTP inducibility through multiple mechanisms. Scientific Reports, 6. [DOI] [PMC free article] [PubMed] [Google Scholar]
  21. LYNCH JJ, FERTZIGER AP, TEITELBAUM HA, CULLEN JW & GANTT WH 1973. Pavlovian conditioning of drug reactions: some implications for problems of drug addiction. Cond Reflex, 8, 211–23. [DOI] [PubMed] [Google Scholar]
  22. MEYER PJ, KING CP & FERRARIO CR 2016. Motivational Processes Underlying Substance Abuse Disorder. Curr Top Behav Neurosci, 27, 473–506. [DOI] [PMC free article] [PubMed] [Google Scholar]
  23. PARKINSON JA, OLMSTEAD MC, BURNS LH, ROBBINS TW & EVERITT BJ 1999. Dissociation in Effects of Lesions of the Nucleus Accumbens Core and Shell on Appetitive Pavlovian Approach Behavior and the Potentiation of Conditioned Reinforcement and Locomotor Activity byd-Amphetamine. The Journal of Neuroscience, 19, 2401–2411. [DOI] [PMC free article] [PubMed] [Google Scholar]
  24. PAVLOV I 1927. Conditioned reflexes: an investigation of the physiological activity of the cerebral cortex, Oxford, England: Oxford Univ. Press. [DOI] [PMC free article] [PubMed] [Google Scholar]
  25. PRICE MP, GONG H, PARSONS MG, KUNDERT JR, REZNIKOV LR, BERNARDINELLI L, CHALONER K, BUCHANAN GF, WEMMIE JA, RICHERSON GB, CASSELL MD & WELSH MJ 2014. Localization and behaviors in null mice suggest that ASIC1 and ASIC2 modulate responses to aversive stimuli. Genes Brain and Behavior, 13, 179–194. [DOI] [PMC free article] [PubMed] [Google Scholar]
  26. RABINAK CA, MORI S, LYONS M, MILAD MR & PHAN KL 2017. Acquisition of CS-US contingencies during Pavlovian fear conditioning and extinction in social anxiety disorder and posttraumatic stress disorder. J Affect Disord, 207, 76–85. [DOI] [PMC free article] [PubMed] [Google Scholar]
  27. RENDER JA, HOWE KR, WUNSCH AM, GUIONAUD S, COX PJ & WEMMIE JA 2010. Histologic examination of the eye of acid-sensing ion channel 1a knockout mice. Int J Physiol Pathophysiol Pharmacol, 2, 69–72. [PMC free article] [PubMed] [Google Scholar]
  28. ROMANIUK L, HONEY GD, KING JR, WHALLEY HC, MCINTOSH AM, LEVITA L, HUGHES M, JOHNSTONE EC, DAY M, LAWRIE SM & HALL J 2010. Midbrain activation during Pavlovian conditioning and delusional symptoms in schizophrenia. Arch Gen Psychiatry, 67, 1246–54. [DOI] [PubMed] [Google Scholar]
  29. SHARPE MJ, MARCHANT NJ, WHITAKER LR, RICHIE CT, ZHANG YJ, CAMPBELL EJ, KOIVULA PP, NECARSULMER JC, MEJIAS-APONTE C, MORALES M, PICKEL J, SMITH JC, NIV Y, SHAHAM Y, HARVEY BK & SCHOENBAUM G 2017. Lateral Hypothalamic GABAergic Neurons Encode Reward Predictions that Are Relayed to the Ventral Tegmental Area to Regulate Learning. Current Biology, 27. [DOI] [PMC free article] [PubMed] [Google Scholar]
  30. SINNING A & HUBNER CA 2013. Minireview: pH and synaptic transmission. FEBS Lett, 587, 1923–8. [DOI] [PubMed] [Google Scholar]
  31. SMOLLER JW, GALLAGHER PJ, DUNCAN LE, MCGRATH LM, HADDAD SA, HOLMES AJ, WOLF AB, HILKER S, BLOCK SR, WEILL S, YOUNG S, CHOI EY, ROSENBAUM JF, BIEDERMAN J, FARAONE SV, ROFFMAN JL, MANFRO GG, BLAYA C, HIRSHFELD-BECKER DR, STEIN MB, VAN AMERINGEN M, TOLIN DF, OTTO MW, POLLACK MH, SIMON NM, BUCKNER RL, ONGUR D & COHEN BM 2014. The Human Ortholog of Acid-Sensing Ion Channel Gene ASIC1a Is Associated With Panic Disorder and Amygdala Structure and Function. Biological Psychiatry, 76, 902–910. [DOI] [PMC free article] [PubMed] [Google Scholar]
  32. SONG N, ZHANG G, GENG W, LIU Z, JIN W, LI L, CAO Y, ZHU D, YU J & SHEN L 2012. Acid sensing ion channel 1 in lateral hypothalamus contributes to breathing control. PLoS One, 7, e39982. [DOI] [PMC free article] [PubMed] [Google Scholar]
  33. TAUGHER RJ, GHOBBEH A, SOWERS LP, FAN R & WEMMIE JA 2015. ASIC1A in the bed nucleus of the stria terminalis mediates TMT-evoked freezing. Front Neurosci, 9, 239. [DOI] [PMC free article] [PubMed] [Google Scholar]
  34. TAUGHER RJ, LU Y, FAN R, GHOBBEH A, KREPLE CJ, FARACI FM & WEMMIE JA 2017. ASIC1A in neurons is critical for fear-related behaviors. Genes Brain Behav, 16, 745–755. [DOI] [PMC free article] [PubMed] [Google Scholar]
  35. TAUGHER RJ, LU Y, WANG YM, KREPLE CJ, GHOBBEH A, FAN R, SOWERS LP & WEMMIE JA 2014. The Bed Nucleus of the Stria Terminalis Is Critical for Anxiety-Related Behavior Evoked by CO2 and Acidosis. Journal of Neuroscience, 34, 10247–10255. [DOI] [PMC free article] [PubMed] [Google Scholar]
  36. VANN KT & XIONG ZG 2018. Acid-sensing ion channel 1 contributes to normal olfactory function. Behav Brain Res, 337, 246–251. [DOI] [PMC free article] [PubMed] [Google Scholar]
  37. WALDMANN R, CHAMPIGNY G, BASSILANA F, HEURTEAUX C & LAZDUNSKI M 1997. A proton-gated cation channel involved in acid-sensing. Nature, 386, 173–7. [DOI] [PubMed] [Google Scholar]
  38. WASSUM KM & IZQUIERDO A 2015. The basolateral amygdala in reward learning and addiction. Neuroscience and Biobehavioral Reviews, 57, 271–283. [DOI] [PMC free article] [PubMed] [Google Scholar]
  39. WEMMIE JA, ASKWITH CC, LAMANI E, CASSELL MD, FREEMAN JH & WELSH MJ 2003. Acid-sensing ion channel 1 is localized in brain regions with high synaptic density and contributes to fear conditioning. Journal of Neuroscience, 23, 5496–5502. [DOI] [PMC free article] [PubMed] [Google Scholar]
  40. WEMMIE JA, CHEN J, ASKWITH CC, HRUSKA-HAGEMAN AM, PRICE MP, NOLAN BC, YODER PG, LAMANI E, HOSHI T, FREEMAN JH JR. & WELSH MJ 2002. The acid-activated ion channel ASIC contributes to synaptic plasticity, learning, and memory. Neuron, 34, 463–77. [DOI] [PubMed] [Google Scholar]
  41. WEMMIE JA, CORYELL MW, ASKWITH CC, LAMANI E, LEONARD AS, SIGMUND CD & WELSH MJ 2004. Overexpression of acid-sensing ion channel 1a in transgenic mice increases acquired fear-related behavior. Proc Natl Acad Sci U S A, 101, 3621–6. [DOI] [PMC free article] [PubMed] [Google Scholar]
  42. WU PY, HUANG YY, CHEN CC, HSU TT, LIN YC, WENG JY, CHIEN TC, CHENG IH & LIEN CC 2013. Acid-sensing ion channel-1a is not required for normal hippocampal LTP and spatial memory. J Neurosci, 33, 1828–32. [DOI] [PMC free article] [PubMed] [Google Scholar]
  43. ZHA XM, WEMMIE JA, GREEN SH & WELSH MJ 2006. Acid-sensing ion channel 1a is a postsynaptic proton receptor that affects the density of dendritic spines. Proc Natl Acad Sci U S A, 103, 16556–61. [DOI] [PMC free article] [PubMed] [Google Scholar]
  44. ZIEMANN AE, ALLEN JE, DAHDALEH NS, DREBOT II, CORYELL MW, WUNSCH AM, LYNCH CM, FARACI FM, HOWARD MA, WELSH MJ & WEMMIE JA 2009. The Amygdala Is a Chemosensor that Detects Carbon Dioxide and Acidosis to Elicit Fear Behavior. Cell, 139, 1012–1021. [DOI] [PMC free article] [PubMed] [Google Scholar]

RESOURCES