Skip to main page content
Access keys NCBI Homepage MyNCBI Homepage Main Content Main Navigation
. 2011;6(12):e29132.
doi: 10.1371/journal.pone.0029132. Epub 2011 Dec 28.

Habituation of the C-start Response in Larval Zebrafish Exhibits Several Distinct Phases and Sensitivity to NMDA Receptor Blockade

Affiliations
Free PMC article

Habituation of the C-start Response in Larval Zebrafish Exhibits Several Distinct Phases and Sensitivity to NMDA Receptor Blockade

Adam C Roberts et al. PLoS One. .
Free PMC article

Abstract

The zebrafish larva has been a valuable model system for genetic and molecular studies of development. More recently, biologists have begun to exploit the surprisingly rich behavioral repertoire of zebrafish larvae to investigate behavior. One prominent behavior exhibited by zebrafish early in development is a rapid escape reflex (the C-start). This reflex is mediated by a relatively simple neural circuit, and is therefore an attractive model behavior for neurobiological investigations of simple forms of learning and memory. Here, we describe two forms of short-lived habituation of the C-start in response to brief pulses of auditory stimuli. A rapid form, persisting for ≥1 min but <15 min, was induced by 120 pulses delivered at 0.5-2.0 Hz. A more extended form (termed "short-term habituation" here), which persisted for ≥25 min but <1 h, was induced by spaced training. The spaced training consisted of 10 blocks of auditory pulses delivered at 1 Hz (5 min interblock interval, 900 pulses per block). We found that these two temporally distinguishable forms of habituation are mediated by different cellular mechanisms. The short-term form depends on activation of N-methyl-d-aspartate receptors (NMDARs), whereas the rapid form does not.

Conflict of interest statement

Competing Interests: The authors have declared that no competing interests exist.

Figures

Figure 1
Figure 1. The C-start reflex in larval zebrafish and the experimental apparatus used to study habituation of the reflex.
(A) Representative C-start reflex in a zebrafish larva. Frames commence at the presentation of an auditory/vibrational (AV) pulse; the white dot indicates the initiation of the C-start reflex. Images were recorded at 1000 frames/s. For illustration purposes, only every other frame is shown. (B) Experimental apparatus used to elicit the C-start. The plastic wells containing the zebrafish were placed on top of a light box, which was used to achieve adequate visual contrast during video recording. Each well was filled with ∼3 ml of E3 and contained one zebrafish. We were able to record the responses of as many as 24 zebrafish in a single session.
Figure 2
Figure 2. Zebrafish do not exhibit a C-start response to auditory stimuli before 4 dpf, and subsequently are most responsive to a 200-Hz auditory stimulus.
(A) Mean response rate of larval zebrafish to AV stimulation during development. No escape responses were elicited 3 dpf (n = 8), whereas the response rate at 4 dpf and 5–6 dpf was 0.92 and 1.0, respectively. (B) Mean volume (dB) of auditory pulses that elicit a C-start response in larvae 4 dpf (n = 12) and 5–6 dpf (n = 8). It required significantly stronger auditory stimulation to evoke a C-start in the younger fish than in the older fish (t [18] = 2.28, p<0.05). One 4 dpf zebrafish failed to respond to auditory pulses, and therefore were assigned a threshold of 135 dB, the highest volume tested. (Data in Figures 2A and 2B are from the same experiments.) The asterisk indicates a significant difference between the groups. Error bars in this and subsequent graphs represent SEM. (C) Mean volume (dB) of sound pulses of different auditory frequencies used to elicit a C-start reflex: 50 Hz (n = 8), 200 Hz (n = 8), 500 Hz (n = 8), and 1000 Hz (n = 8). A one-way ANOVA performed on the group data was significant (F [3,28] = 4.58, p<0.01). The mean threshold to elicit escape responses was 104.13±3.38 dB in the 50 Hz group, 92.5±1.32 dB in the 200 Hz group, 104.5±2.93 dB in the 500 Hz group, and 96.5±2.93 dB in the 1000 Hz group. SNK post hoc tests showed that a 200 Hz stimulus elicited escape responses at a significantly lower threshold than did either the 50 Hz stimulus or the 500 Hz stimulus (p<0.05), as indicated by the asterisk.
Figure 3
Figure 3. Low-frequency stimulation is more effective at eliciting rapid habituation than high-frequency stimulation.
(A) Mean response rates of larval zebrafish (6–7 dpf) after 120 pulses of AV stimulation at different frequencies, tested 10 s after the last training pulse. The response rates were: 0.0167 Hz, 0.35±0.10 (n = 23); 0.1 Hz, 0.29±0.10 (n = 24); 0.5 Hz, 0.04±0.04 (n = 24); 1 Hz, 0.04±0.04 (n = 24); 2 Hz, 0.04±0.04 (n = 24); 10 Hz, 0.35±0.10 (n = 23); and 60 Hz, 0.33±0.10, (n = 24). (B) Mean rate of C-start responses 10 s after habituation training (1 Hz, 120 pulses). The Trained group (n = 24) exhibited a significantly lower response rate than did the Test alone (n = 24; t [46] = 6.44, p<0.0001), as indicated by the asterisk. The data for the Trained group came from the experiments shown in (A). (C) Mean response rates of larval zebrafish to pulses of different auditory frequencies, tested 1 min after training. The data are from the same experiments presented in (A). The response rates were: 0.0167 Hz, 0.26±0.09 (n = 23); 0.1 Hz, 0.29±0.10 (n = 24); 0.5 Hz, 0.25±0.13 (n = 24); 1 Hz, 0.38±0.10 (n = 24); 2 Hz, 0.29±0.10 (n = 24); 10 Hz, 0.70±0.10 (n = 23); and 60 Hz, 0.58±0.10 (n = 24). (D) Mean response rate 1 min after habituation training (1 Hz, 120 pulses). The Trained group (n = 24) exhibited significantly greater habituation than did the Test alone group (n = 24; t [46] = 2.07, p<0.05), as indicated by the asterisk. The data for the Trained group were from the experiments in (C). (E) Persistence of rapid habituation following training at 1 Hz (120 pulses). The pretest response rate was based on the response to the first stimulus of the habituation training, and posttest rates were based on the responses of the same animals (n = 12) at 10 s, 1 min, and 15 min after training. To determine when the response rate returned to baseline levels, we performed a repeated measures ANOVA (F [3,33] = 12.72, p<0.0001). SNK post hoc tests used to probe for significant differences among the responses to the various tests indicated that habituation was present at 10 s and 1 min after training compared to the pretest and 15 min posttest (p<0.05 for each comparison), whereas no habituation was present at the 15 min posttest compared to the pretest response rate (p>0.05). The asterisk indicates a significant difference between the 10 s test and the pretest and 15 min test values, whereas a pound sign indicates a significant difference between the 1 min test and the pretest and 15 min test values. (F) Number of pulses required to elicit rapid habituation (1-Hz stimulation). To determine the most effective number of pulses to elicit habituation (1-min posttest), we first performed a one-way ANOVA, which showed that the groups differences were significant (F [3, 92] = 9.89, p<0.0001). SNK post hoc tests indicated that the group that received 120 AV pulses had a significantly lower response rate than the groups that received only 20 or 60 AV pulses (p<0.05 for each comparison), as indicated by an asterisk. The group that received 900 pulses was significantly less responsive than the group that received 20 pulses (p<0.05), as indicated by a pound sign.
Figure 4
Figure 4. Pharmacological antagonism of NMDA receptors does not disrupt rapid habituation.
(A) Response rates during habituation training (1 Hz, 120 pulses) in APV Trained (n = 7) and E3 Trained (n = 9) larvae. Blockade of NMDARs did not alter habituation to the first 10 AV stimuli. The response rate of the APV Trained group was 0.21±0.04, whereas that of the E3 Trained group was 0.26±0.03 (t [14] = 0.79, p>0.4). (B) In addition, there was no difference between the APV Trained group (n = 19) and the E3 Trained group (n = 21) on the 1-min posttest (t [38] = 0.62, p>0.5). (Note that the data in this graph include fish from [A], in which the responses during training were recorded and other fish in which only the posttest response was recorded.)
Figure 5
Figure 5. Spaced training elicited short-term habituation that depends on NMDAR activation.
(A) Protocol for spaced training. (B) Habituation Indices (HI) for larvae after spaced training at different stimulus frequencies. The HI was determined by measuring the response rate of the zebrafish prior to training (3 tests, 5 min ISI) and after training (15–25 min after training; 3 tests, 5 min ISI). The posttest response rates were then subtracted from the pretest response rate to calculate the HI. The HIs for the 15–25 min posttests were: 0.5 Hz, 0.06±0.10 (n = 6); 1 Hz, −0.54±0.13 (n = 8); 2 Hz, −0.34±0.11 (n = 8); 10 Hz, 0.08±0.08 (n = 8); and 50 Hz, 0.05±.09 (n = 7). (C) Mean HI after spaced training (1 Hz) during the 15–25-min and 60–70-min posttests. There were three groups: a group that was tested 15–25 min after training (15–25 min Trained, n = 8), a group (Test alone, n = 6) that did not receive habituation training, and a group (60–70 min Trained, n = 24) that received habituation training and was tested 60–70 min after training. A one-way ANOVA indicated that there were significant differences among the three groups (F [2,35] = 6.76, p<0.01). SNK post hoc tests indicated that the 15–25 min Trained group showed greater habituation than did the Test alone and the 60–70 min Trained groups (p<0.05 for each comparison), as indicated by the asterisk. The data used for the 15–25 min Trained group was from the experiments presented in (A). (D) Blockade of NMDARs disrupted the habituation induced by spaced training. The HI of the APV Trained group (n = 15) was −0.04±0.10, whereas it was −0.33±0.09 for the E3 Trained group (n = 15) (t [28] = 2.16; p<0.05). The asterisk indicates a statistically significant difference. (E) APV treatment did not disrupt the responsiveness of untrained fish. There were no differences between the HIs of the untrained (Test alone) group treated with APV (n = 14) and the Test alone group treated with E3 (n = 14). The mean HI for the untrained APV-treated group was −0.07±0.06), and −0.07±.07 for the untrained E3-treated group (t [26] = 0.008, p>0.9).
Figure 6
Figure 6. Model of the M-cell circuitry showing the potential sites of synaptic change that underlie habituation in the zebrafish.
Asterisks indicate potential sites of synaptic plasticity at excitatory chemical/electrical synapses (open symbols) and inhibitory synapses (filled symbols). Nonsynaptic changes, including ephaptic inhibition of the M-cell and decreased excitability of the sensory afferents or M-cell, may also contribute to behavioral habituation. AC = axon cap.

Similar articles

See all similar articles

Cited by 20 articles

See all "Cited by" articles

References

    1. Bourne JN, Harris KM. Balancing structure and function at hippocampal dendritic spines. Annual Review of Neuroscience. 2008;31:47–67. - PMC - PubMed
    1. Newpher TM, Ehlers MD. Spine microdomains for postsynaptic signaling and plasticity. Trends in Cell Biology. 2009;19:218–227. - PubMed
    1. Kessels HW, Malinow R. Synaptic AMPA receptor plasticity and behavior. Neuron. 2009;61:340–350. - PMC - PubMed
    1. Mozzachiodi R, Byrne JH. More than synaptic plasticity: role of nonsynaptic plasticity in learning and memory. Trends in Neurosciences. 2010;33:17–26. - PMC - PubMed
    1. Silva AJ, Zhou Y, Rogerson T, Shobe J, Balaji J. Molecular and cellular approaches to memory allocation in neural circuits. Science. 2009;326:391–395. - PMC - PubMed

Publication types

MeSH terms

Substances

Feedback