| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Department of Biology and Neuroscience Graduate Group, University of Pennsylvania, Philadelphia, Pennsylvania 19104
Submitted 22 September 2003; accepted in final form 8 January 2004
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONMETHODSRESULTSDISCUSSIONACKNOWLEDGMENTSREFERENCES |
|---|
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONACKNOWLEDGMENTSREFERENCES |
|---|
; Coenen and Vendrik 1972; Livingstone and Hubel 1981; Morrow and Casey 1992; Poggio and Mountcastle 1963; Steriade et al. 1969; Swadlow and Weyand 1987). Sensory processing in the avian song system is also profoundly affected by changes in behavioral state (Cardin and Schmidt 2003b; Dave et al. 1998; Nick and Konishi 2001; Rauske et al. 2003; Schmidt and Konishi 1998). Behavioral state dynamically modulates both the strength and tuning of auditory responses in nucleus HVC, a high-order sensorimotor integration area of the song system (Cardin and Schmidt 2003b). HVC population auditory responses are variable and unselective during wakefulness but stable and highly selective for the bird's own song (BOS) over all other auditory stimuli during sedation. In addition, HVC auditory responsiveness is completely suppressed by arousal.
HVC is part of both the motor and anterior forebrain pathways of the song system and receives projections from the ascending auditory pathway (see Fig. 1A). Field L, the primary avian forebrain auditory area, is directly innervated by the auditory thalamic nucleus Ovoidalis (Ov) and is tonotopically organized (Zaretsky and Konishi 1976). Field L generally responds to many different complex stimuli (Grace et al. 2003; Janata and Margoliash 1999; Lewicki and Arthur 1996; Lim and Kim 1997) and is reciprocally connected with the caudal lateral mesopallium (CLM), which sends projections to the song system nucleus interfacialis (NIf) (Vates et al. 1996). NIf, which projects directly to HVC, has been proposed as a major source of auditory input to HVC (Janata and Margoliash 1999).
|
; Schmidt and Konishi 1998). This suggests that the observed changes in HVC activity may result from direct modulation within HVC, modulation of auditory activity in an afferent structure such as NIf, or simultaneous modulation in HVC and NIf or other afferent structures. Previous studies have suggested that HVC may be the initial site of modulation in this pathway (Dave et al. 1998; Schmidt and Konishi 1998). However, immunohistochemical evidence suggests that NIf is a candidate site of modulatory effects. Neuromodulatory inputs to NIf from cholinergic (Ryan and Arnold 1981) and catecholaminergic (Harding et al. 1998; Mello et al. 1998; Soha et al. 1996) systems could play a role in regulating auditory input to HVC.
In this study, we investigated the relationship between NIf and HVC by recording simultaneously in both structures during wakefulness and sedation. We present evidence that NIf auditory responses are co-modulated with HVC responses by changes in behavioral state. In addition, pharmacological manipulations in NIf lead to modulation of HVC auditory responsiveness. Some of these data have been previously presented in abstract form (Cardin and Schmidt 2002).
|
METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONACKNOWLEDGMENTSREFERENCES |
|---|
Adult male zebra finches (Taeniopygia guttata) ranging from 120 to 500 days of age were obtained from our breeding colony and from a local supplier. Birds were housed under constant 12:12 light:dark conditions and given food and water ad libitum. All procedures described here were approved by an Institutional Animal Care and Use Committee at the University of Pennsylvania.
Auditory stimuli
The song of each bird was recorded in a sound attenuation chamber and digitized at 40 kHz with Goldwave (Goldwave, St. Johns, Canada). For BOS stimuli, two song motifs were presented in normal orientation so that the bird heard the motifs as he would when singing. For reversed-BOS stimuli (REV), a two-motif segment was played backward. Auditory stimuli were presented at 70-dB sound pressure level (SPL) peak intensity. SPL measurements were made with A weighting.
Chronic recordings
Electrodes and equipment were as described in detail by Cardin and Schmidt (2003b) and Schmidt and Konishi (1998). Briefly, the bird was anesthetized with 0.07 ml ketamine/xylazine (40 mg/kg ketamine and 8 mg/kg xylazine, Phoenix Pharmaceuticals, St. Joseph, MO), and nichrome wire electrodes (0.2–0.7 M
) were implanted in HVC and NIf. These low-impedance electrodes record multiunit activity from a population of neurons. HVC and NIf were located in all experiments by stereotaxic coordinates and by their distinctive, coordinated activity patterns (see Fig. 1B). Electrodes were secured in place with dental cement and attached to a connector (Omnetics Connector, Minneapolis, MN) on the skull. A silver ground wire was inserted under the skull and cemented in place.
Birds were allowed to recover for several days following surgery before being placed in a recording chamber and connected to a lightweight operational amplifier (Texas Instruments, Dallas, TX) and a mercury commutator. Large multiunit neural signals were band-passed between 500 Hz and 10 kHz and digitized at 20 kHz. The birds remained connected for 3–5 days at a time and were kept on a consistent light cycle throughout the experiments. They were handled and disturbed as little as possible to observe normal variations in behavior and arousal state.
During awake auditory trials, the birds were continuously observed on a video monitor to ensure that their eyes were open and that they were moving in the chamber. All awake auditory trials were conducted with the lights on. Any trials on which a bird did any of the following were discarded: vocalized, closed its eyes for 
2 consecutive s, tucked its head under a wing, or crouched on the floor of the cage without moving. Auditory experiments during wakefulness consisted of 30–60 identical auditory stimulus trials. Each auditory stimulus trial comprised 2 s of baseline recording followed by 
2 s of auditory stimulus and 2 additional s of baseline recording. The time interval between the end of one trial and the beginning of the next was varied during each experiment and ranged from 5 s to 3 min. An auditory response was defined as the summed neural response to a set of auditory trials.
For arousal experiments, the implanted birds were given an intramuscular injection of diazepam (Abbott Laboratories, North Chicago, IL) and allowed to rest for 15–20 min. Under diazepam sedation, HVC and NIf demonstrate robust, song-selective auditory responses. As described in previous work (Cardin and Schmidt 2003b), this dose of diazepam resulted in a state of mild sedation from which birds were easily aroused. Arousal experiments consisted of a block of 30–60 identical BOS stimuli. The time interval between trials ranged from 5 s to 3 min. At the beginning of randomly selected trials, the bird was aroused by a light touch to the chest feathers with a padded tool.
Acute recordings
Surgery procedures were as described in detail by Cardin and Schmidt (2003b). A head post was cemented to the skull of each bird during a short initial surgery under ketamine-xylazine anesthesia. On the day of an auditory experiment, the bird was given an intramuscular injection of 7.5 mg/kg diazepam and 0.1 ml 5% dextrose (Abbott Laboratories). The bird was secured in a stereotaxic apparatus by the head post and received a small initial injection of 2% lidocaine hydrochloride (Phoenix Pharmaceuticals) subcutaneously to the scalp. To ensure normal audition and decrease discomfort, neither ear bars nor a beak bar were used.
Windows were opened in the skull over HVC and NIf and small incisions made in the dura in each location, through which glass (5–20 M) electrodes were lowered. Sharp glass (World Precision Instruments, New Haven, CT) electrodes were pulled on a Micropipette Puller P-97 (Sutter Instrument, Novato, CA). Glass electrodes were filled with isotonic saline. Once the electrodes were in place, a series of auditory trials using the BOS and REV stimuli was run to confirm song-selective HVC auditory responses.
Prior to the beginning of each acute recording experiment, we assessed the effectiveness of an air puff to the chest in arousing the bird as described in Cardin and Schmidt (2003b). A bird was considered aroused by the air puff if we observed behavioral indicators such as the eyes opening for several seconds, feathers ruffling, and tail movements. Behavioral indicators of arousal lasted 3–10 s. When the air puff was confirmed to consistently arouse the bird, the stereotaxic apparatus was moved into the sound-attenuating chamber, and the door was closed.
Arousal experiments were composed of 30–60 identical auditory stimulus trials. All auditory stimuli were examples of the bird's song recorded in a sound attenuation chamber and digitized at 40 kHz. The stimuli were played at 70 dB (SPL) through an HLS410 speaker (JBL, Northridge, CA) in the sound chamber via a Crown D-45 Amplifier (Crown International, Elkhart, IN). The time interval between trials was varied during each experiment and ranged from 5 s to 3 min. A set of 30 auditory trials usually spanned 10–18 min. On randomly interleaved trials, the bird was aroused by a puff of air to the chest 1 s before the onset of the auditory stimulus. Neural data from the acute experiments was amplified and bandpassed between 500 Hz and 10 kHz and digitized at 20 kHz. Acquisition software was written in Labview (National Instruments) by A. Leonardo.
Pharmacology
For pharmacology experiments, birds were lightly sedated with diazepam and secured in the stereotaxic apparatus with a head post as described above. Electrodes were placed in HVC and NIf, and a series of auditory stimulus trials was run to confirm song-selective auditory responses in both areas. The NIf electrode was replaced by a glass pipette connected to a Hamilton syringe containing the drug solution. A small amount of drug (0.2 µl) was injected into NIf over the course of 2 min. We used either muscimol (8 mM in 0.9% saline, Sigma-Aldrich, St. Louis, MO) or bicuculline methiodide (10 mM in 0.9% saline, Sigma-Aldrich). After waiting for an additional 2 min, the pipette was slowly removed. Several additional sets of auditory trials were run to assess the HVC auditory response.
During all pharmacology experiments, the drug solution contained 10% biotinylated dextran amine (BDA; Molecular Probes, Eugene, OR). The bird was killed immediately following the experiment, and the tissue was fixed and cryoprotected. Staining for BDA allowed confirmation of the location and approximate spread of the microinjection (see Fig. 12A). Data from experiments in which the staining spread beyond the borders of NIf were discarded.
|
After chronically implanted birds had been recorded for 2–10 days, they were deeply anesthetized with 0.1 ml 50 mg/ml Nembutal (Abbott Laboratories) and transcardially perfused with 0.9% saline and 4% paraformaldehyde. Brains were cryoprotected in 30% sucrose and sectioned at 50 µm on a freezing microtome. Electrode placement was confirmed using cresyl violet staining. For identification of microinjection sites, sectioned tissue was processed with an avidin-biotin-horseradish peroxidase complex kit followed by a reaction with a peroxidase substrate kit (Vector Laboratories, Burlingame, CA).
Data analysis
Data were analyzed as described in detail by Cardin and Schmidt (2003b) using Matlab (The Mathworks, Natick, MA) routines written by J. A. Cardin and M. F. Schmidt. Spike events in the multiunit data from acute and chronic experiments were measured by using a peak-detection algorithm. For each data set, the threshold was visually positioned at a point clearly above background noise but low enough to detect all observed spike events. Peristimulus time histograms (PSTHs) were calculated by binning spike events (bin size = 10 ms) during each trial and summing the resulting raster plots over 15–30 auditory trials. Bin size did not affect the results of the data analysis described below.
Auditory responses in awake birds were frequently observed to involve changes in mean firing rate, changes in the temporal distribution of spike events, or both (Cardin and Schmidt 2003b). We therefore evaluated auditory responses based on both response strength and variance measurements taken from PSTHs. The number of spikes in each PSTH bin was defined as the bin total. Thus we could compare the set of bin totals from the stimulus period to the set of bin totals from the baseline period. For each PSTH, we performed two calculations: 1) an unpaired t-test comparing the set of bin totals representing the BOS stimulus period to the set of bin totals representing a baseline period of equivalent duration and 2) an F test to compare the variance of the sets of PSTH bin totals representing the summed neural activity during the same stimulus and baseline periods. The recorded response was defined as a significant auditory response if P < 0.01 for either of the above tests.
Index values
To quantitatively represent changes in both firing rate and temporal distribution of spike events during auditory responses in awake and sedated birds, we used a combination of response strength and variance measurements (see Cardin and Schmidt 2003b). Variance (V) was measured as the variance of the set of PSTH bin totals resulting from a set of auditory trials. The response strength index (RSINDEX) for each auditory trial was calculated by dividing the difference between mean BOS and baseline firing rates by their sum
 |
is the mean firing rate during BOS stimulus presentation, and 
is the mean firing rate during the preceding baseline period. RSINDEX for each set of auditory trials is shown as mean ± SE. While the RSINDEX measure does not include an assessment of variance, conclusions reached using a z-score method did not differ from those presented here. The variance index (VINDEX) for each set of auditory trials was calculated as follows
 |
Index values range from –1 to 1. An RSINDEX or VINDEX value of 0 indicates no difference between the BOS response and baseline activity. A positive RSINDEX value indicates a greater response during the BOS presentations than during the baseline periods, while a negative RSINDEX value indicates a smaller response during the BOS stimulus presentations than during the baseline periods. Similarly, a positive VINDEX value indicates a greater variance of bin totals during the stimulus period than during the baseline period, while a negative VINDEX value indicates lower variance of bin totals during the stimulus period than during the baseline period.
Response pattern
To compare HVC and NIf auditory responses to successive sets of BOS stimuli at different times, the patterns of successive auditory responses at a given recording site were compared in two ways. Each auditory response was the summed neural activity during a set of 30–60 auditory trials. Each PSTH resulting from a set of auditory trials was normalized by dividing all bin totals by the mean of the baseline bin totals. To investigate the prevalence of variability of awake auditory responses, we used the set of normalized bin totals corresponding to the BOS stimulus period of the PSTH. The sets of PSTH bin totals were compared with the corresponding bin totals from the next response by a paired t-test.
To evaluate the similarity of successive auditory response patterns, we performed linear correlation analyses of pairs of PSTHs normalized as described above. Several pairs of responses were analyzed for many birds. To avoid biasing the analysis toward birds whose auditory responses were sampled most often, we computed mean r values for each recording site in each bird before comparing data sets from different behavioral states. All linear correlations were Pearson correlations.
We evaluated the relationship between NIf and HVC auditory responsiveness on two time scales. For the longer time scale of variation from set to set of auditory trials over a day, we used index values to represent the HVC and NIf auditory responses to each set of auditory trials. Linear correlations were performed on the paired HVC and NIf RSINDEX and VINDEX measurements taken over the course of the day for each bird.
For the short time scale of stimulus-evoked changes in HVC and NIf firing rates during individual trials within a set, we assessed RS at each recording site for all the auditory trials in a set. RS is a simple measure of the change in mean firing rate during the auditory stimulus on each trial
 |
The relationship between HVC and NIf RS measurements was tested by linear correlation. All linear correlations were Pearson correlations.
Spontaneous activity
To assess spontaneous firing rates in NIf and HVC, we used samples of neural activity recorded in the absence of auditory stimuli. Thirty 2-s samples of spontaneous multiunit population activity were taken from recordings of NIf and HVC activity from each bird during each behavioral state. A mean firing rate was calculated for HVC and NIf for each bird in units of spike counts/s. NIf and HVC firing rates were compared across the population of birds by paired t-test. Mean firing rates for the population of birds were calculated by finding the mean firing rate for all NIf and all HVC samples.
Selectivity
The selectivity of awake auditory responses was measured only if the auditory response to the BOS stimulus was significantly different from baseline activity. We did not observe any instances of a significant response to the REV stimulus in the absence of a response to the BOS stimulus. The selectivity of a recording site for BOS versus REV stimuli was measured using d' values (Green and Swets 1966; Janata and Margoliash 1999; Mooney 2000; Solis and Doupe 1997; Theunissen and Doupe 1998). The d' value comparing the response to BOS stimuli relative to REV stimuli was calculated as follows
 |
is the mean response strength to a set of BOS auditory stimuli and was calculated as described above. 
is the mean response strength to the corresponding set of REV stimuli. 
and 
represent the variance of and , respectively. A d' value of 0.5 or greater was used as the criterion for a response selective for BOS over REV stimuli (Solis and Doupe 1997). Population d' values are shown as mean ± SE.
|
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONACKNOWLEDGMENTSREFERENCES |
|---|
Characterization of NIf auditory responses in awake birds
To investigate the relationship between NIf and HVC during wakefulness, we recorded simultaneous multiunit activity from ipsilateral NIf-HVC pairs in 23 chronically implanted birds. Electrode locations were verified in three ways, as shown in the examples from one chronically implanted bird in Fig. 1, B–D. First, NIf and HVC were identified by their coordinated bursts of activity during sedation, as observed by Janata and Margoliash (1999). Figure 1B shows an example of spontaneous multiunit activity recorded simultaneously from chronically implanted electrodes in an ipsilateral NIf-HVC pair during light sedation. Second, as observed previously (McCasland 1987), both NIf and HVC demonstrated robust premotor activity during song. Figure 1C shows premotor activity in NIf and HVC during vocal production of several introductory notes and four complete song motifs. Finally, electrode location was confirmed by cresyl violet histology. An example of a chronic electrode track to NIf is shown in Fig. 1D. This study only includes data from recording sites confirmed by all three criteria.
We observed that neurons in both NIf and HVC can demonstrate auditory responses to the BOS stimulus during wakefulness. Each bird was presented with 1–12 sets of BOS stimuli over the course of 1–4 days. In total, 85 sets of auditory stimuli were presented to 23 chronically implanted birds during wakefulness. We measured multiunit neural responses to each set of BOS stimulus trials to assess whether they were significantly different from baseline activity on the basis of response strength, variance, or both (see METHODS). Using this method, 96.5% (82/85) of the sets of auditory trials generated significant auditory responses in HVC, and 98.8% (84/85) of the sets generated significant auditory responses in the ipsilateral NIf. All birds (23/23) demonstrated at least one significant awake auditory response in both HVC and NIf.
While the prevalence of awake auditory responses in HVC and NIf was similar, the distribution of response characteristics was quite different. In both our previous (Cardin and Schmidt 2003b) and current experiments, many HVC auditory responses involved not an increase in mean firing rate but a change in the temporal distribution of spike events. However, auditory responses recorded simultaneously in NIf predominantly involved a change in mean firing rate (Fig. 2). To quantify the distribution of these response characteristics, we used only the auditory responses previously determined to be significantly different from baseline activity, as described above. We found that 17.1% (14/82) of the HVC auditory responses were significant for response strength alone (RS only), 57.3% (47/82) were significant for response strength and variance (RS and V), and 25.6% (21/82) were significant for variance alone (V only; Fig. 2, left). In contrast, most NIf auditory responses fell into the RS and V category: 90.6% (76/84) were significant for both response strength and variance, while 7.1% (6/84) were significant for response strength alone, and only 2.3% (2/84) were significant for variance alone (Fig. 2, right). Thus NIf auditory responses almost always involved a change in both mean firing rate and temporal distribution of spike events, while one-fourth of the HVC responses involved only a change in the temporal distribution of spike events.
|
Our previous work has indicated that HVC auditory responses vary during wakefulness. Because HVC receives a strong input from NIf, this modulation of HVC activity may reflect varying auditory responsiveness in NIf. To compare the stability of NIf and HVC auditory responses in awake birds, we presented 20 chronically implanted birds with 2–12 sets of 30 identical BOS stimuli at different times on a single day. The mean interval between the end of one set of auditory stimuli and the beginning of the next set was 73.6 ± 31.2 min. While we observed auditory responses in both HVC and NIf in awake birds, auditory responses in both nuclei varied greatly over time. Figure 3A shows an example of simultaneous recordings in an ipsilateral HVC and NIf pair during two awake auditory trial sets separated by 10 min. At the first time point (left panels), both NIf (top) and HVC (bottom) demonstrated robust, if different, auditory responses to 30 presentations of the BOS stimulus. However, at the second time point (right panels), the auditory responses were greatly diminished in both nuclei.
|
Next, we used linear correlation analyses to test the hypothesis that the observed modulation in auditory responsiveness is specific to wakefulness. Figure 3B shows mean correlation coefficients for pairs of NIf and HVC auditory responses during wakefulness and sedation. Only immediately successive pairs of auditory responses recorded within a short interval (5–180 min) were compared. In both NIf and HVC, pairs of auditory responses were much less correlated in awake than in sedated birds. The mean r value for pairs of NIf auditory responses was 0.39 ± 0.03 during wakefulness (n = 49 comparisons in 20 birds) and 0.71 ± 0.04 during sedation (n = 7 comparisons in 6 birds; P < 0.01; Mann-Whitney U test; Fig. 3B, left). Similarly, the mean r value for pairs of successive awake HVC auditory responses (0.33 ± 0.03; n = 49 comparisons in 20 birds) was significantly smaller than that for sedated HVC responses (0.71 ± 0.05; n = 7 comparisons in 6 birds; P < 0.01; Mann-Whitney U test; Fig. 3B,right). The mean interval between awake responses was 48.2 ± 16.3 min, while the mean interval between sedated responses was 41.6 ± 4.9 min. These results indicate that the observed modulation of auditory responses in both NIf and HVC is predominantly restricted to the state of wakefulness.
Co-variation of auditory responses in NIf and HVC during wakefulness
The above data indicate that both NIf and HVC responses vary in awake birds. To evaluate the relationship between awake auditory responses in NIf and HVC, we compared NIf and HVC auditory responses from individual birds over two ranges of time intervals: 1) long intervals, from set to set of auditory trials during the day, and 2) short intervals, over the course of the 30 auditory trials within a set. We calculated index values to quantitatively represent the response strength (RSINDEX) and variance (VINDEX) properties of neural responses to each set of auditory trials. Index values represent all possible auditory responsiveness measurements within a range of –1 to 1 (see METHODS). Only data from birds with six or more awake auditory trial sets on the same day (n = 4 birds) were used. The mean interval between sets of awake auditory trials was 31.5 ± 10.4 min.
The example in Fig. 4 illustrates the observed co-variation of NIf and HVC auditory responsiveness over 1 day. Figure 4 shows NIf and HVC RSINDEX and VINDEX data from one bird during 11 sets of awake auditory trials in 1 day. Auditory responsiveness values in NIf and HVC were highly correlated (P < 0.01) over the day. The correlation coefficients for NIf and HVC RSINDEX (Fig. 4A) and VINDEX (Fig. 4B) values were 0.84 and 0.90, respectively (Pearson linear correlation). Analysis of auditory responses from all four birds showed that, in each bird, both NIf and HVC RSINDEX values (r = 0.74 ± 0.07) and VINDEX values (r = 0.78 ± 0.07) were significantly correlated over the course of 1 day (P < 0.01). Taken together, these results indicate that awake auditory responsiveness in NIf and HVC co-varies over the relatively long time scale of 1 day.
|
Behavioral state alters the relationship between neural activity in NIf and HVC
HVC auditory responses are modulated by changes in behavioral state (Cardin and Schmidt 2003b; Nick and Konishi 2001; Rauske et al. 2003; Schmidt and Konishi 1998). We therefore explored the impact of behavioral state on the relationship between spontaneous and evoked neural activity in NIf and HVC. To quantify spontaneous firing rates in NIf and HVC during wakefulness, we measured mean multiunit firing rate in the absence of auditory stimuli in 23 awake chronically implanted birds (see METHODS). A subset of 14 birds was then sedated, and multiunit spontaneous activity was again measured. Figure 5A shows examples of spontaneous firing rates in NIf and HVC during wakefulness (top) and light sedation in one bird (bottom). During wakefulness, NIf demonstrated a higher level of population activity than HVC. During sedation in the same bird, spontaneous activity in NIf and HVC exhibited coordinated bursting activity, as shown previously by Janata and Margoliash (1999). The graphs to the right show mean multiunit NIf and HVC activity levels for the entire group of birds during wakefulness and sedation. Mean spontaneous multiunit NIf firing rates in awake birds (142.5 ± 26.8 spike events/s) were significantly higher than during sedation (27.5 ± 6.8 spike events/s; P < 0.01; n = 14 birds; paired t-test). In contrast, there was no significant difference between mean multiunit HVC firing rates in awake (18.2 ± 4.0 spike events/s) and sedated (16.9 ± 4.2 spike events/s) birds, despite the change from bursting to tonic activity (NS; n = 14 birds; paired t-test). Thus while the overall level of activity in HVC did not change with behavioral state, multiunit firing rates in NIf were significantly greater during wakefulness than during sedation.
|
A subset (n = 14) of the chronically implanted birds was lightly sedated after completion of the awake auditory trials. During sedation, the relationship between NIf and HVC responses was altered. Figure 5C shows mean NIf and HVC RSINDEX and VINDEX values for awake and sedated auditory responses. NIf RSINDEX and VINDEX values were significantly greater than HVC values during wakefulness (P < 0.01; paired t-test). However, during sedation, this relationship was reversed, and HVC RSINDEX and VINDEX values were greater than NIf values (P < 0.01; paired t-test). These data suggest that the relationship between auditory responsiveness in NIf and HVC is greatly altered by changes in behavioral state.
Auditory response patterns in NIf and HVC are more similar during sedation than during wakefulness
While the degree of auditory responsiveness in NIf and HVC co-varied during wakefulness, the specific patterns of simultaneously recorded NIf and HVC auditory responses could still be quite different, as shown by the example in Fig. 3A (left). To investigate the effect of behavioral state on the relationship between NIf and HVC response patterns, we compared simultaneously recorded auditory response PSTHs from NIf and HVC by linear correlation (see METHODS). Figure 6A shows examples of simultaneous NIf and HVC auditory response PSTHs in one bird during wakefulness and sedation. While the awake response patterns are quite different (r = 0.21), with peaks at different time points, the sedated response patterns are more similar (r = 0.63; Pearson linear correlation). Analysis of data from all birds demonstrated that the mean correlation coefficient for pairs of simultaneous NIf and HVC responses was significantly smaller during wakefulness (r = 0.31 ± 0.03; n = 85 comparisons in 23 birds) than during sedation (r = 0.62 ± 0.05; n = 22 comparisons in 14 birds; P < 0.01; Mann-Whitney U test; Fig. 6B). These results suggest that during sedation, both auditory responsiveness, as measured by index values, and response patterns are similar in NIf and HVC. In contrast, while overall NIf and HVC auditory responsiveness values co-vary during wakefulness, the patterns of simultaneous awake auditory responses in NIf and HVC differ significantly.
|
We have observed that multiunit activity in HVC is highly selective for the BOS stimulus during sedation but unselective during wakefulness (Cardin and Schmidt 2003b). Because NIf has been proposed as a main auditory input to HVC, we predicted that NIf would not demonstrate song selectivity during wakefulness. To test this hypothesis, we presented 51 sets of randomly intermixed BOS and REV stimuli to 23 chronically implanted awake birds. Song selectivity was assessed by calculating d' values; a d' of >0.5 was considered selective for BOS over REV, while a value of less than –0.5 was considered selective for REV over BOS (see METHODS). Figure 7A shows an example of awake auditory responses in NIf to interleaved presentations of BOS and REV stimuli. In this bird, the NIf responses to the BOS (left) and REV (right) stimuli were quite similar (d' = 0.09). Figure 7B (left) shows NIf d' values plotted against HVC d' values for each of the 51 BOS-REV comparisons. While we observed some (n = 14/51) auditory responses with NIf or HVC d' >0.5 in awake birds, mean d' measurements for NIf (0.11 ± 0.06) and HVC (0.22 ± 0.05) in awake birds revealed that both structures were predominantly unselective. NIf d' values during wakefulness were significantly smaller than those from HVC (P < 0.01; paired t-test). During sedation, both NIf and HVC demonstrated significant song selectivity for the BOS stimulus. Mean sedated NIf d' was 0.87 ± 0.16, while mean sedated HVC d' was 2.89 ± 0.45 (n = 12 comparisons in 12 birds; P < 0.01; paired t-test; Fig. 7B, right). Thus both NIf and HVC demonstrate song selectivity during sedation but not during wakefulness. These results indicate that behavioral state profoundly affects the fundamental characteristics of auditory processing throughout this ascending auditory pathway.
|
Our observations of co-variation of NIf and HVC auditory activity in awake birds suggest that auditory activity in NIf, like that in HVC, may be modulated by behavioral state. To explore this hypothesis, we used an acute recording paradigm in which we recorded simultaneously from ipsilateral NIf and HVC pairs while repeatedly changing the bird's behavioral state. All recording sites in both NIf and HVC demonstrated a song-selective response to the BOS stimulus prior to the beginning of the arousal experiments (data not shown). In these experiments, lightly sedated birds were intermittently aroused by an air puff on randomly selected auditory trials (see METHODS). Using this arousal paradigm, we recorded from 13 pairs of NIf and HVC multiunit recording sites in nine birds. All of the NIf and HVC recording sites (13/13) showed a significant suppression of auditory responsiveness following arousal. Figure 8 shows data from a typical NIf-HVC arousal experiment. During resting trials (left panels), both NIf (top) and HVC (bottom) demonstrated robust responses to the BOS stimulus. In contrast, auditory responses in both NIf and HVC were completely suppressed by arousal on randomly intermixed trials (right panels).
|
|
|
|
Neural activity in NIf affects HVC auditory responsiveness
Taken together, the preceding results from both awake and sedated experiments indicate that the observed modulation of HVC auditory responses is correlated with changes in NIf auditory responsiveness. However, little is known about the relationship between neural activity in NIf and HVC. We recorded HVC auditory responses in lightly sedated birds while pharmacologically manipulating neural activity in the ipsilateral NIf. In all pharmacology experiments, the injected solution included BDA. Experiments in which BDA staining extended beyond NIf were excluded. An example of a BDA-labeled injection site is depicted in Fig. 12A, left. As shown by comparison to the cresyl violet staining of the same section of tissue (right), the injection site was restricted within the borders of NIf.
We first assessed the effects of suppressing activity in NIf on auditory responsiveness in HVC. For these experiments, we recorded multiunit activity in HVC while injecting muscimol, a GABAA receptor agonist, to the ipsilateral NIf (n = 5 NIf-HVC pairs in 4 birds). Figure 12B shows data from a typical experiment. During baseline recording, there was a robust HVC auditory response to the BOS stimulus (Fig. 12B, left). Immediately following a small injection (0.2 µl) of muscimol (8 mM in 0.9% saline) to the ipsilateral NIf, the entire HVC auditory response, and indeed most spontaneous HVC activity, was eliminated (Fig. 12B, right). In all experiments (5/5), muscimol injection to NIf completely eliminated all ipsilateral HVC auditory responses (Fig. 12C). Mean HVC RSINDEX was 0.38 ± 0.04 before muscimol and 0.05 ± 0.03 after muscimol injection to NIf (P < 0.01; paired t-test; Fig. 12C, left). Mean HVC VINDEX before muscimol was 0.64 ± 0.1 and after muscimol to NIf was –0.09 ± 0.16 (P < 0.01; paired t-test; Fig. 12C, right). After 15–40 min of recovery, both mean HVC RSINDEX (0.34 ± 0.03) and mean HVC VINDEX (0.44 ± 0.09) returned to preinjection levels. Injection of muscimol to NIf also had a suppressive effect on spontaneous activity in HVC. While spontaneous HVC activity was robust prior to drug injection (33.7 ± 9.9 spike events/s), it was significantly lower following muscimol treatment (5.12 ± 3.9 spike events/s; P < 0.01; paired t-test).
The above results suggest that HVC auditory responsiveness depends on neural activity in NIf. To further test this hypothesis, we injected bicuculline, a GABAA receptor antagonist, into NIf while recording from the ipsilateral HVC (n = 4 NIf-HVC pairs in 4 birds). Injection of a small (0.2 µl) amount of bicuculline (10 mM in 0.9% saline) to NIf caused an increased auditory response to the BOS stimulus in the ipsilateral HVC. Figure 13A shows data from a typical experiment. During baseline recording, an HVC auditory response to the BOS stimulus was recorded (Fig. 13A, left). After a low dose of bicuculline was injected to the ipsilateral NIf, the HVC auditory response was increased (Fig. 13A, right), and an additional PSTH peak appeared in response to the initial portion of the auditory stimulus (arrow). In all experiments (4/4), bicuculline injection to NIf increased the ipsilateral HVc auditory response (Fig. 13B). Mean HVC RSINDEX before bicuculline was 0.27 ± 0.03 and after bicuculline injection to NIf was 0.41 ± 0.01 (P < 0.01; paired t-test; Fig. 13B, left). Mean HVC VINDEX before bicuculline was 0.50 ± 0.06 and after bicuculline to NIf was 0.75 ± 0.04 (P < 0.01; paired t-test; Fig. 13B, right). After 30–50 min of recovery, both mean HVC RSINDEX (0.33 ± 0.02) and mean HVC VINDEX (0.61 ± 0.05) returned to preinjection levels.
|
As a control, we injected saline alone into NIf and recorded from the ipsilateral HVC. Injections of small amounts of 0.9% saline (0.2 µl) had no effect on HVC auditory responsiveness (n = 4 NIf-HVC pairs in 4 birds). Mean HVC RSINDEX during the baseline period was 0.56 ± 0.04 and after saline was 0.53 ± 0.06 (NS; paired t-test). After 15–30 min of recovery, mean HVC RSINDEX was 0.51 ± 0.05. Similarly, mean HVC VINDEX during the baseline period was 0.82 ± 0.03 and after saline was 0.76 ± 0.06 (NS; paired t-test). After recovery, mean HVC VINDEX was 0.77 ± 0.06. Thus the injection of vehicle alone to NIf does not affect auditory responses in the ipsilateral HVC.
An increase in GABAergic activation in NIf thus leads to a profound suppression of auditory responsiveness in HVC, suggesting that NIf inputs are necessary for the expression of HVC auditory responses. The reduction of spontaneous activity in HVC after muscimol to NIf indicates that input from NIf plays an important role in determining spontaneous HVC activity under sedation. In contrast, decreasing GABAergic activity in NIf leads to an increase in HVC auditory responsiveness in the absence of any change in spontaneous neural activity. These data support a role for NIf in providing auditory input to HVC and suggest that modulation of NIf activity is one possible mechanism by which HVC auditory responsiveness may be regulated.
|
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONACKNOWLEDGMENTSREFERENCES |
|---|
NIf input is necessary for spontaneous and auditory activity in HVC
A previous study by Janata and Margoliash (1999) characterized the properties of NIf neurons in urethane-anesthetized birds. They found that NIf responded preferentially to the BOS stimulus but was not as song selective as HVC in the same birds. They also observed correlated firing patterns in NIf and HVC. In our experiments, we observed similar characteristics in sedated birds. Coincident bursting in NIf and HVC might be caused by NIf driving HVC activity. Alternatively, both structures might be driven by a common input such as Uva. In the latter case, NIf might provide a depolarizing offset to HVC neurons, allowing input from another source to drive HVC bursting. The almost complete loss of spontaneous HVC activity following injection of muscimol to NIf suggests that NIf inputs are necessary for spontaneous HVC activity under sedation. In contrast, levels of spontaneous activity in NIf during wakefulness are greatly increased, and NIf and HVC activity does not display any obvious degree of coordination.
Because NIf provides the strongest known anatomical input to HVC from any auditory structure (Fortune and Margoliash 1995; Nottebohm et al. 1982), it has been proposed as a major source of auditory input to HVC (Janata and Margoliash 1999). However, Field L projects to the HVC shelf region into which some HVC dendrites extend, suggesting a second possible route for auditory input to reach HVC (Fortune and Margoliash 1995; Kelley and Nottebohm 1979). Preliminary data have suggested that injections of GABA, lidocaine, or muscimol into NIf, but not Field L, affect the spontaneous and evoked activity of the ipsilateral HVC (Boco and Margoliash 2001; Fee et al. 2002). Likewise, we found in this study that injecting small doses of the GABAA receptor agonist muscimol into NIf eliminated all auditory HVC activity. In addition, small injections of the GABAA receptor antagonist bicuculline into NIf led to an increase in HVC auditory responsiveness. Together, these results suggest that NIf is a major source of auditory input to HVC. While it is possible that NIf inputs simply provide a depolarizing offset that allows another auditory input to drive HVC, it seems unlikely that the main auditory input to HVC comes from another source. Uva, the thalamic structure that projects to both NIf and HVC, does not demonstrate auditory responses (Wild 1994). MMAN, another forebrain structure that projects to HVC, demonstrates longer auditory response latencies than those observed in HVC (Vates et al. 1997). Finally, inputs to HVC from Fie
, NIf auditory response measurements; 
, HVC auditory response measurements. A: NIf and HVC RSINDEX values co-varied over the day (r = 0.84; P < 0.01; linear correlation). B: similarly, NIf and HVC VINDEX values co-varied over the day (r = 0.90, P < 0.01; linear correlation). Dotted lines, baseline activity levels.
