Limits...
Auditory synapses to song premotor neurons are gated off during vocalization in zebra finches.

Hamaguchi K, Tschida KA, Yoon I, Donald BR, Mooney R - Elife (2014)

Bottom Line: A potential site for this interaction is the song premotor nucleus HVC, which receives auditory input and contains neurons (HVCX cells) that innervate an anterior forebrain pathway (AFP) important to feedback-dependent vocal plasticity.Using intracellular recordings in singing zebra finches, we found that DAF failed to perturb singing-related synaptic activity of HVCX cells, although many of these cells responded to auditory stimuli in non-singing states.These findings support a model in which the AFP accesses feedback independent of HVC.

View Article: PubMed Central - PubMed

Affiliation: Department of Neurobiology, Duke University Medical Center, Durham, United States.

ABSTRACT
Songbirds use auditory feedback to learn and maintain their songs, but how feedback interacts with vocal motor circuitry remains unclear. A potential site for this interaction is the song premotor nucleus HVC, which receives auditory input and contains neurons (HVCX cells) that innervate an anterior forebrain pathway (AFP) important to feedback-dependent vocal plasticity. Although the singing-related output of HVCX cells is unaltered by distorted auditory feedback (DAF), deafening gradually weakens synapses on HVCX cells, raising the possibility that they integrate feedback only at subthreshold levels during singing. Using intracellular recordings in singing zebra finches, we found that DAF failed to perturb singing-related synaptic activity of HVCX cells, although many of these cells responded to auditory stimuli in non-singing states. Moreover, in vivo multiphoton imaging revealed that deafening-induced changes to HVCX synapses require intact AFP output. These findings support a model in which the AFP accesses feedback independent of HVC. DOI: http://dx.doi.org/10.7554/eLife.01833.001.

Show MeSH

Related in: MedlinePlus

Synaptic inputs onto HVCX cells encode future syllable onsets.(A) An example of a stereotyped syllable sequence, with syllables underlined by orange bars. (B) Pseudo-color representation of the membrane potential patterns relative to baseline while the bird sang many repetitions of the syllable sequence shown in (A). Black dots; syllable onsets. Data are aligned to a specific syllable and sorted from long to short motifs, which reflect natural variations in zebra finch song tempo. (C) Same data shown in (B), but dPSP onsets are shown by black dots, and the timing of individual syllables is represented by the orange regions. (D) Syllable onset- and offset-triggered average of dPSP onset rate (n = 53 HVCX cells). (E) Example spectrograms from catch and ‘hit-and-truncated’ songs. (F) Trial-averaged subthreshold activity during hit-and-truncated songs showed a clear deviation from activity during catch trials. (G) Cumulative histogram of syllable onset timings relative to DAF onset in hit trials including delayed syllables (red dashed line), hit trials without delayed syllables (red solid line), and catch trials (black line). (H) Examples of trial-averaged subthreshold activity reveal that delayed motor output drives changes in the subthreshold activity (compare black and red dashed lines). After removing hit trials with delayed syllables, the subthreshold activity in hit and catch conditions is nearly identical (black vs red solid lines).DOI:http://dx.doi.org/10.7554/eLife.01833.013
© Copyright Policy
Related In: Results  -  Collection

License
getmorefigures.php?uid=PMC3927426&req=5

fig6: Synaptic inputs onto HVCX cells encode future syllable onsets.(A) An example of a stereotyped syllable sequence, with syllables underlined by orange bars. (B) Pseudo-color representation of the membrane potential patterns relative to baseline while the bird sang many repetitions of the syllable sequence shown in (A). Black dots; syllable onsets. Data are aligned to a specific syllable and sorted from long to short motifs, which reflect natural variations in zebra finch song tempo. (C) Same data shown in (B), but dPSP onsets are shown by black dots, and the timing of individual syllables is represented by the orange regions. (D) Syllable onset- and offset-triggered average of dPSP onset rate (n = 53 HVCX cells). (E) Example spectrograms from catch and ‘hit-and-truncated’ songs. (F) Trial-averaged subthreshold activity during hit-and-truncated songs showed a clear deviation from activity during catch trials. (G) Cumulative histogram of syllable onset timings relative to DAF onset in hit trials including delayed syllables (red dashed line), hit trials without delayed syllables (red solid line), and catch trials (black line). (H) Examples of trial-averaged subthreshold activity reveal that delayed motor output drives changes in the subthreshold activity (compare black and red dashed lines). After removing hit trials with delayed syllables, the subthreshold activity in hit and catch conditions is nearly identical (black vs red solid lines).DOI:http://dx.doi.org/10.7554/eLife.01833.013

Mentions: Our findings indicate that HVCX cells are insensitive to auditory feedback perturbation during singing, raising the possibility that they primarily encode motor-related information. One expectation of a motor-related signal is that it should precede sound onset during singing. To test this idea, we examined the relationship between the onsets of dPSPs and syllables during bouts of singing (Figure 6A–C). Population-averaged cross-correlation analysis revealed that dPSP onsets in HVCX cells occur on average ∼25 ms before syllable onsets (n = 53 cells, Figure 6D), similar to the findings from prior extracellular studies (Kozhevnikov and Fee, 2007; Fujimoto et al., 2011). In contrast, dPSP onsets were not correlated with syllable offsets (Figure 6D), indicating that singing-related synaptic inputs to HVCX cells are correlated to future, rather than ongoing or past, vocal output. Another expectation of a motor-related signal is that, should change in parallel to changes in vocal output. As previously mentioned, DAF can trigger acute effects, such as truncation or slowing of song tempo (Figure 6E,G) (Sakata and Brainard, 2008), which are distinct from the slower changes in song that are associated with feedback-dependent error correction mechanisms. In fact, we observed significant changes in membrane potential fluctuations of HVCX cells when DAF-triggered song truncation or delays in syllable onsets (Figure 6F,H). Therefore, the intracellular method we used is capable of detecting DAF-related changes in the subthreshold activity of HVCX cells, but only when DAF triggers acute vocal motor effects. Taken together, these findings suggest that synaptic inputs to HVCX cells do not convey auditory feedback signals during singing and instead encode information about future vocal output, consistent with a motor-related signal.10.7554/eLife.01833.013Figure 6.Synaptic inputs onto HVCX cells encode future syllable onsets.


Auditory synapses to song premotor neurons are gated off during vocalization in zebra finches.

Hamaguchi K, Tschida KA, Yoon I, Donald BR, Mooney R - Elife (2014)

Synaptic inputs onto HVCX cells encode future syllable onsets.(A) An example of a stereotyped syllable sequence, with syllables underlined by orange bars. (B) Pseudo-color representation of the membrane potential patterns relative to baseline while the bird sang many repetitions of the syllable sequence shown in (A). Black dots; syllable onsets. Data are aligned to a specific syllable and sorted from long to short motifs, which reflect natural variations in zebra finch song tempo. (C) Same data shown in (B), but dPSP onsets are shown by black dots, and the timing of individual syllables is represented by the orange regions. (D) Syllable onset- and offset-triggered average of dPSP onset rate (n = 53 HVCX cells). (E) Example spectrograms from catch and ‘hit-and-truncated’ songs. (F) Trial-averaged subthreshold activity during hit-and-truncated songs showed a clear deviation from activity during catch trials. (G) Cumulative histogram of syllable onset timings relative to DAF onset in hit trials including delayed syllables (red dashed line), hit trials without delayed syllables (red solid line), and catch trials (black line). (H) Examples of trial-averaged subthreshold activity reveal that delayed motor output drives changes in the subthreshold activity (compare black and red dashed lines). After removing hit trials with delayed syllables, the subthreshold activity in hit and catch conditions is nearly identical (black vs red solid lines).DOI:http://dx.doi.org/10.7554/eLife.01833.013
© Copyright Policy
Related In: Results  -  Collection

License
Show All Figures
getmorefigures.php?uid=PMC3927426&req=5

fig6: Synaptic inputs onto HVCX cells encode future syllable onsets.(A) An example of a stereotyped syllable sequence, with syllables underlined by orange bars. (B) Pseudo-color representation of the membrane potential patterns relative to baseline while the bird sang many repetitions of the syllable sequence shown in (A). Black dots; syllable onsets. Data are aligned to a specific syllable and sorted from long to short motifs, which reflect natural variations in zebra finch song tempo. (C) Same data shown in (B), but dPSP onsets are shown by black dots, and the timing of individual syllables is represented by the orange regions. (D) Syllable onset- and offset-triggered average of dPSP onset rate (n = 53 HVCX cells). (E) Example spectrograms from catch and ‘hit-and-truncated’ songs. (F) Trial-averaged subthreshold activity during hit-and-truncated songs showed a clear deviation from activity during catch trials. (G) Cumulative histogram of syllable onset timings relative to DAF onset in hit trials including delayed syllables (red dashed line), hit trials without delayed syllables (red solid line), and catch trials (black line). (H) Examples of trial-averaged subthreshold activity reveal that delayed motor output drives changes in the subthreshold activity (compare black and red dashed lines). After removing hit trials with delayed syllables, the subthreshold activity in hit and catch conditions is nearly identical (black vs red solid lines).DOI:http://dx.doi.org/10.7554/eLife.01833.013
Mentions: Our findings indicate that HVCX cells are insensitive to auditory feedback perturbation during singing, raising the possibility that they primarily encode motor-related information. One expectation of a motor-related signal is that it should precede sound onset during singing. To test this idea, we examined the relationship between the onsets of dPSPs and syllables during bouts of singing (Figure 6A–C). Population-averaged cross-correlation analysis revealed that dPSP onsets in HVCX cells occur on average ∼25 ms before syllable onsets (n = 53 cells, Figure 6D), similar to the findings from prior extracellular studies (Kozhevnikov and Fee, 2007; Fujimoto et al., 2011). In contrast, dPSP onsets were not correlated with syllable offsets (Figure 6D), indicating that singing-related synaptic inputs to HVCX cells are correlated to future, rather than ongoing or past, vocal output. Another expectation of a motor-related signal is that, should change in parallel to changes in vocal output. As previously mentioned, DAF can trigger acute effects, such as truncation or slowing of song tempo (Figure 6E,G) (Sakata and Brainard, 2008), which are distinct from the slower changes in song that are associated with feedback-dependent error correction mechanisms. In fact, we observed significant changes in membrane potential fluctuations of HVCX cells when DAF-triggered song truncation or delays in syllable onsets (Figure 6F,H). Therefore, the intracellular method we used is capable of detecting DAF-related changes in the subthreshold activity of HVCX cells, but only when DAF triggers acute vocal motor effects. Taken together, these findings suggest that synaptic inputs to HVCX cells do not convey auditory feedback signals during singing and instead encode information about future vocal output, consistent with a motor-related signal.10.7554/eLife.01833.013Figure 6.Synaptic inputs onto HVCX cells encode future syllable onsets.

Bottom Line: A potential site for this interaction is the song premotor nucleus HVC, which receives auditory input and contains neurons (HVCX cells) that innervate an anterior forebrain pathway (AFP) important to feedback-dependent vocal plasticity.Using intracellular recordings in singing zebra finches, we found that DAF failed to perturb singing-related synaptic activity of HVCX cells, although many of these cells responded to auditory stimuli in non-singing states.These findings support a model in which the AFP accesses feedback independent of HVC.

View Article: PubMed Central - PubMed

Affiliation: Department of Neurobiology, Duke University Medical Center, Durham, United States.

ABSTRACT
Songbirds use auditory feedback to learn and maintain their songs, but how feedback interacts with vocal motor circuitry remains unclear. A potential site for this interaction is the song premotor nucleus HVC, which receives auditory input and contains neurons (HVCX cells) that innervate an anterior forebrain pathway (AFP) important to feedback-dependent vocal plasticity. Although the singing-related output of HVCX cells is unaltered by distorted auditory feedback (DAF), deafening gradually weakens synapses on HVCX cells, raising the possibility that they integrate feedback only at subthreshold levels during singing. Using intracellular recordings in singing zebra finches, we found that DAF failed to perturb singing-related synaptic activity of HVCX cells, although many of these cells responded to auditory stimuli in non-singing states. Moreover, in vivo multiphoton imaging revealed that deafening-induced changes to HVCX synapses require intact AFP output. These findings support a model in which the AFP accesses feedback independent of HVC. DOI: http://dx.doi.org/10.7554/eLife.01833.001.

Show MeSH
Related in: MedlinePlus