| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
 |
ABSTRACT |
|---|
 TopAbstract IntroductionMethodsResultsDiscussionReferences |
|---|
|
INTRODUCTION |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionReferences |
|---|
Repeated stimulation of carotid chemoreceptors, or electrical stimulation of the carotid sinus nerve or brainstem mid-line, produces a long-term facilitation of phrenic motoneurone activity in animals vagotomized, injected with a neuromuscular blocking agent and artificially ventilated (Millhorn et al. 1980; Hayashi et al. 1993; Bach & Mitchell, 1996; Morris et al. 1996a). This respiratory memory persists for more than an hour. Induction of long-term facilitation is blocked or attenuated by serotonin antagonists in cats (Millhorn et al. 1980). Serotonin is known to have long-lasting excitatory effects on various respiratory motoneurones (Holtman et al. 1986; Monteau et al. 1990; Berger et al. 1992; Lindsay & Feldman, 1993); it increases the response of phrenic motoneurones, which innervate the diaphragm, to non-serotonergic excitation (Mitchell et al. 1992).
Recent research suggests that much of the expression of long-term facilitation results from the effects of serotonin on motoneurones (Fuller et al. 2000a). Ketanserin, a serotonin competitive inhibitor, administered before, but not after, episodic hypoxia blocks the motor expression of long-term facilitation (Fuller et al. 2000b). It was suggested that brainstem respiratory neural networks do not participate in long-term facilitation beyond its induction. This study addressed the alternative hypothesis that changes in medullary respiratory neural networks that persist beyond stimulation of carotid chemoreceptors contribute to long-term facilitation.
Neurones in the ventral respiratory group (VRG) of the medulla, including Bötzinger (Böt) and pre-Bötzinger regions, have been implicated in the process of respiratory rhythm generation (Bianchi et al. 1995; Rekling & Feldman, 1998). Bulbospinal neurones in both the VRG and the dorsal respiratory group (DRG) of the nucleus tractus solitarii (NTS) provide drive to phrenic inspiratory motoneurones (Cohen et al. 1974; Davies et al. 1985). Serotonergic neurones are found in the raphe nuclei (Palkovits et al. 1974; Jacobs & Azmitia, 1992) and project to the DRG, Böt-VRG (Connelly et al. 1989; Smith et al. 1989) and spinal cord (Morrison & Gebber, 1984; Holtman et al. 1986). Stimulation of carotid chemoreceptors produces transient firing rate changes in raphe neurones that show evidence of functional connectivity to Böt-VRG and NTS neurones as well as phrenic motoneurones (Morris et al. 1996a).
We measured firing rates of Böt-VRG, DRG and raphe neurones before and after induction of long-term facilitation to test for maintained rate changes appropriate for roles in the expression of this respiratory memory. A second aim was to screen correlated spike train data for non-random timing relationships that could be interpreted as changes in the effective connectivity of the respiratory network associated with long-term facilitation. Preliminary accounts of the results have been published (Morris et al. 1993, 1996c).
|
METHODS |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionReferences |
|---|
All experiments were performed under protocols approved by the University of South Florida's Animal Care and Use Committee. Other results on responses to carotid chemoreceptor stimulation from these experiments, some of which were done in collaboration with A. Arata, have been reported previously (Morris et al. 1996a,b); most of the methods were described in detail therein. The data analysed here were obtained from nine adult cats initially anaesthetized with sodium thiopental (22.0 mg kg-1, I.V.). Anaesthesia was maintained with Dial-urethane (allobarbital-CIBA, 60.0 mg kg-1; urethane, 240 mg kg-1). Cats were vagotomized, artificially ventilated and injected with a bolus of the neuromuscular blocking agent gallamine triethiodide (2.2 mg kg-1; I.V.) followed by constant infusion (0.4 mg kg-1 h-1). Animals received dexamethasone (2.0 mg kg-1) and atropine (0.5 mg kg-1). Arterial blood pressure, PO2, PCO2, pH, [HCO3-] and end-tidal CO2 were monitored. Cats were respirated with 100 % O2 to counteract the hypoxia due to ventilation-perfusion mismatching common with positive pressure ventilation in an open chest preparation. End-tidal CO2 was continuously monitored and maintained at the lowest value that would maintain a cycling phrenic nerve discharge; this value was between 28 and 35 mmHg. When necessary, sodium bicarbonate solution (8.4 %) was infused to correct metabolic acidosis, and low molecular weight dextran solution was infused to maintain a mean systemic blood pressure of at least 100 mmHg. Core body temperature was maintained at 38.0 ± 0.5 °C. The animals were killed by an overdose of sodium pentobarbitone.
Neuronal impulses were monitored extracellularly with planar arrays of six to eight individual tungsten microelectrodes (3-5 M
) in n. raphe obscurus, rostral VRG (the Bötzinger and pre-Bötzinger regions), caudal VRG (region of nucleus ambiguus and retroambigualis), and the NTS. Signals were amplified, filtered (100-5000 Hz band pass) and recorded on FM instrumentation recorders together with efferent phrenic nerve activity, a stimulus marker, systemic blood pressure and, in some experiments, carotid blood pressure. Common synchronization pulses (5 Hz) were recorded on each tape.
Multiunit efferent phrenic nerve activity was amplified and fed into a resistor-capacitor 'leaky' integrator with a time constant of 200 ms. A moving, smoothed first derivative was used to detect the start and end of inspiration (Korten & Haddad, 1989) from the digitized, integrated phrenic nerve signal. In some experiments, whole nerve phrenic multiunit activity and the synchronization pulse channel were also digitized with 16 bit precision at 5 kHz for subsequent spike-triggered averaging. Spike trains were subjected to two statistical evaluations of respiratory modulation and a measure of respiratory modulation, 
2, was calculated (Orem & Dick, 1983; Morris et al. 1996b). Single neurones with no statistically significant respiratory modulation of their firing rates were classified as not respiratory modulated (NRM). Cycle-triggered histograms were used to classify cells with a statistically significant respiratory modulation according to the phase (inspiration, I, or expiration, E) during which they were more active. Neurones with peak firing rates in the first half of the phase and a longer period of decrementing than of augmenting activity were classified as decrementing (Decr) cells. Cells with peak firing rates in the second half of the phase were denoted as augmenting (Aug) neurones. Respiratory cells that discharged in patterns not easily categorized were denoted as Others (Segers et al. 1987).
Carotid chemoreceptors were stimulated in series of three to six stimuli at 3-5 min intervals with 200 µl of CO2-saturated 0.9 % saline solution, injected via the external carotid artery over a period of 30 s. The criterion for induction of long-term facilitation was a persistent increase in the amplitudes of peak integrated phrenic activity at least 2 S.D. above mean prestimulus control values with the same end-tidal CO2.
The firing rate of each neurone during expression of long-term facilitation was compared with control values. All respiratory cycles were measured during the control period prior to the first stimulation (first sample, duration 10 min) and following the afterdischarge of the last chemoreceptor stimulus period (second sample, at least 5 min post-stimulus, duration 10 min). The samples consisted of the average firing rates of each spike train during each respiratory cycle. Student's t test was then used to determine if there were sufficient differences between the two samples to reject the null hypothesis that there was no change in firing rate.
Measures of short time scale correlation
Three sets of cross-correlograms were calculated for each neurone pair (Perkel et al. 1967). Spike train data samples were drawn from equal time segments (300 s) at the beginning and end of a control period and after induction of long-term facilitation. The post-long-term facilitation sample was at least 5 min after and within 15 min of the final chemoreceptor stimulation. The detectability index (DI) of each significant primary correlogram feature was calculated (Aertsen & Gerstein, 1985). The percentage difference between DI values from the beginning and the end of the control interval and the difference between the end of control and the sample recorded after induction of long-term facilitation were calculated. A sign test (P < 0.05) was applied to test whether the change in DIs was greater after long-term facilitation more often than chance. Autocorrelograms were calculated for each neurone and used as an aid in the interpretation of all cross-correlation histograms. Unrectified and full wave rectified spike-triggered averages of digitized multiunit phrenic activity were calculated and visually inspected for significant features with latencies up to 1 s (Cohen et al. 1974).
A related method was used to test individual pairs of neurones for changes in short time scale correlations following induction of long-term facilitation. Spike trains composed of interleaved successive samples drawn from before and after induction of long-term facilitation of phrenic nerve activity were constructed (Fig. 1). The mid-point of each such interval became a 'virtual' stimulus followed by spike train data recorded during expression of long-term facilitation. Timing codes were added to mark the beginning of each reconstructed interval and served as 'triggers' for calculation of the joint peristimulus time histogram (JPSTH) for each correlated pair of spike trains (Aertsen et al. 1989). Each JPSTH was normalized by dividing the difference between the JPSTH and the PSTH product by the standard deviation of the PSTH product. This procedure is intended to compensate for changes in correlation expected solely as a consequence of changes in the firing rates of the recorded neurones. Normalized JPSTHs were evaluated for significance (Aertsen et al. 1989).
 |
View larger version [in this window] [in a new window] |
|
|
A, integrated efferent phrenic nerve activity and firing rate histograms from 5 neurones recorded simultaneously in n. raphe obscurus during induction of long-term facilitation. Rectangles mark samples of control (left) and post-induction (right) data selected for further analysis. B, each sample was divided into successive equal duration segments; specific values of segment durations are indicated in Results. C, successive data segments from the two samples were interleaved and virtual 'trigger' and 'stimulus' events were added at the start and middle of each pair of interleaved segments, respectively. This data structure allowed use of the joint peristimulus time histogram (JPSTH) with the surprise significance test (see Methods) to evaluate pairs of spike trains for changes in non-random timing relationships after memory induction.
| ||
Correlations among some neurones were evaluated further with the gravity method for the analysis and representation of groups of simultaneously monitored neurones and their dynamic associations (Gerstein & Aertsen, 1985; Gerstein et al. 1985). Each of N neurones is represented as a particle in N space. Each particle is initially equidistant from all other particles and has a time-varying charge that is a filtered version of its spike activity. The charged particles exert forces on each other and move as though in a viscous fluid. Non-random timing relationships in the spike trains were indicated by significant particle condensation (Lindsey et al. 1992).
Next, neurones represented by particle pairs that aggregated significantly (P < 0.01) were screened for evidence of recurring transient assembly configurations (Lindsey et al. 1997). Successive elements in each row of a two-dimensional array contained the aggregation velocity of each pair of particles during an interval represented by the column. The velocities in each column served, in turn, as a template that was compared to all other columns for a match. Velocity values were nulled if the particle pair never exhibited significant aggregation, or if they were less than a nulling threshold of 3.25 % of the maximum velocity detected in the data set in order to filter out arbitrarily small fluctuations in aggregation velocity. In this work, a column was judged a match and tested for significance if more than 90 % of the non-nulled velocities in the template column matched in the target column and more than 75 % of the matched pairs had velocities within a range of ±25 % of the corresponding template velocities. A column had to have at least two velocities to be used as a template. The null hypothesis was that the number of repeating patterns found when a particular template column was compared to the other columns was not greater than expected by chance. The null hypothesis was rejected if the maximum number of matches for a particular column was greater than the number found when the same column was instead compared to the columns in all 100 pair-wise shuffled condensation profile sets.
|
RESULTS |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionReferences |
|---|
The firing rates of 87 of 105 neurones (83 %), measured in 10 recordings from nine animals, changed following induction of long-term facilitation (Table 1). Up to 30 neurones were recorded simultaneously. Firing rate histograms of 24 neurones recorded simultaneously in four brainstem sites together with integrated phrenic nerve activity are shown in Fig. 2B. The first segment includes control data and firing rates during the first interval of peripheral chemoreceptor stimulation. The second segment includes activity just prior to, during and immediately following the fifth and final period of stimulation. The third segment shows firing rate histograms during expression of long-term facilitation in efferent phrenic activity; this record began 6 min after the final stimulus.
 |
View larger version [in this window] [in a new window] |
|
|
A, schematic dorsal view of the cat brainstem and electrode arrays in nucleus raphe obscurus, rostral and caudal VRG, and the DRG of the NTS. B, data segments show firing rate histograms of 24 neurones recorded simultaneously at the indicated sites during the first (I) and fifth (II) period of carotid chemoreceptor stimulation and 6 min following the fifth and final stimulus and induction of long-term facilitation (III). C, offset peak in average of phrenic nerve activity triggered by 57 004 spikes in the neurone designated DRG 3 in panel B. D, central trough in average of phrenic nerve activity triggered by 3739 spikes in a rostral VRG I-Decr neurone. The trigger cell had a lower firing rate following induction of long-term facilitation relative to the preceding control period.
| ||
Averages of phrenic multiunit activity triggered by the spike trains of 11 caudal VRG and DRG inspiratory neurones had peaks with positive time lags. Nine of these inspiratory trigger neurones had increased firing rates after induction of long-term facilitation. An example of an offset peak in an average triggered by a DRG neurone is shown in Fig. 2C. The activity of DRG neurone 3 increased with long-term facilitation. Because offset peaks in spike-triggered averages are most simply interpreted as excitatory connections (see Discussion), this result was consistent with the hypothesis that increased premotor drive originating in the brainstem contributes to long-term facilitation of respiratory motoneurone activity.
Four averages triggered by Böt-VRG neurones, which decreased their activity with long-term facilitation, had central or offset troughs. Two of these cells were classified as I-Decr, one was an I-Aug, and the fourth an E-Aug. A central trough in an average triggered by one of the I-Decr neurones is shown in Fig. 2D. Functional implications are considered in Discussion.
Fourteen of twenty-one neurones in raphe obscurus with control firing rates less than 4 Hz had significant long-term increases in activity. A significant proportion of neurones with firing rates under 4 Hz in raphe have been characterized as serotonergic (Woch et al. 1996). Five of the fourteen had short time scale features in phrenic spike-triggered averages, including three central peaks, one with an offset trough with a positive time lag, and one average with multiple peaks and troughs.
Correlated spike trains
Cross-correlation histograms were calculated using the entire recorded spike trains for each of a total of 1272 pairs; 153 had significant features (detectability index, DI > 2.0, a subset of features summarized in Morris et al. 1996a,b). When the data were segmented into two control periods and one post-stimulus interval, there were sufficiently many spikes in those samples to calculate DIs for each period for 93 pairs. The changes in synchrony (i.e. short time scale correlations as measured by DIs, see Methods) after long-term facilitation were greater than those between the equivalent time periods in the control phase, more often than chance (sign test, P < 0.05). This result suggested that long-term facilitation was associated with altered effective connectivity among neurones in the sampled brainstem regions.
Increased synchrony after induction of long-term facilitation was detected between pairs of inspiratory neurones that included putative premotor neurones. A greater increase in detectability indices after long-term facilitation than between control periods (sign test, P < 0.05) was measured in 6 of 7 pairs (85.7 %) of inspiratory VRG neurones with cross-correlogram peaks and at least one offset peak in the corresponding spike-triggered averages of phrenic activity. Cross-correlograms of one of the six pairs of VRG inspiratory neurones are shown in Fig. 3A. The top correlogram was calculated from data recorded during the control period (700 s); no central peak was detected. The middle correlogram from data recorded after induction of long-term facilitation had a central peak (DI = 2.9), also apparent in the central bin of the bottom correlogram calculated with larger bin widths (also 700 s).
 |
View larger version [in this window] [in a new window] |
|
|
A, top, cross-correlogram from a pair of caudal VRG inspiratory neurones calculated from spike train samples acquired during the prestimulus control period had no significant feature; 3453 reference and 11 720 target spikes; 0.5 ms bins. Centre and bottom, correlograms (0.5 ms and 5.5 ms bins, respectively) computed with spikes from the same neurones recorded during expression of the respiratory memory; 4332 reference and 9495 target spikes. A central peak was observed (centre: DI = 2.9); it persisted as a bin with high counts in the bottom correlogram (DI = 7.3). B, a narrow offset peak with a positive 0.5 ms time lag documents the increased firing probability of a caudal VRG inspiratory neurone following spikes in a rostral VRG neurone. The top two cross-correlograms were calculated from two samples of spike data recorded prior to induction of long-term facilitation (top, DI = 10.7, 8267 reference and 4892 target spikes; middle, DI = 10.6, 8216 reference and 4569 target spikes). The bottom correlogram was calculated from a 300 s sample recorded after long-term facilitation induction (DI = 12.4, 8170 reference and 5892 target spikes). C, the left panel shows the surprise matrix derived from a normalized JPSTH calculated from interleaved samples of spike data from the neurones represented in B. The data consisted of 1760 concatenated control and long-term facilitation data segments, each 200 ms in duration, arranged as described in Fig. 1. The arrows indicate pre- to post-induction border. Note that the upper left and lower right quadrants represent correlations of spike trains that are actually disjoint many minutes in time. The increased number of significant bins along the diagonal in the upper right quadrant indicates an increase in effective connectivity after long-term facilitation. The right panel shows a smoothed (Gaussian of 2 bins) histogram that is a tally of corresponding paradiagonal bins in the surprise matrix; the pair of neurones discharged more synchronously after induction of the respiratory memory. Calculation based on 12 960 rostral VRG neurone spikes, 10 546 caudal VRG cell spikes.
| ||
Other correlational analyses
There were two subsets of data that presented additional opportunities for further analysis. The criteria for their selection were: (a) respiratory modulated firing patterns of the constituent neurones; (b) significant changes in firing rate during carotid chemoreceptor stimulation correlated with altered respiratory efferent activity; (c) persistent firing rate changes with long-term facilitation; (d) evidence of effective connectivity appropriate to contribute to long-term facilitation; and (e) greater changes in effective connectivity after induction of long-term facilitation than during different control periods.
Results from the analysis of one pair of Böt-VRG neurones are shown in Fig. 3B. The trigger neurone for these cross-correlograms was an I-Decr neurone recorded in the rostral VRG (region of the pre-Bötzinger complex). The target cell was an I-Aug neurone in the caudal VRG. Averages of phrenic motoneurone activity triggered by spikes in each of the two neurones included offset peaks with positive lags superimposed on broader peaks. These correlations, documented previously (Fig. 7 of Morris et al. 1996b), were consistent with paucisynaptic excitatory connectivity with the phrenic motoneurone pool, given that the averaging procedure measures changes due to direct and indirect parallel premotor pathways (Discussion).
The top two correlograms in Fig. 3B, calculated from control data samples, both document a transient increase in the firing probability of the caudal neurone following spikes in the rostral cell with DI values of 10.7 and 10.6, respectively. The peak in the correlogram from the spike data recorded after induction of long-term facilitation had a DI of 12.4. The percentage changes in detectability indices that were greater with long-term facilitation than during control periods in groups of neurone pairs could not be used to establish changes in synchrony of specific individual pairs because we lacked data as to what percentage change in the detectability index constituted significance for a particular feature. Therefore, we adapted the joint peristimulus time histogram (JPSTH; Methods) to test the significance of changes in specific cross-correlogram features and to compare effective connectivity before and after long-term facilitation.
The JPSTH calculated for interleaved control and post-long-term facilitation spike data for the pair of neurones represented in Fig. 3B detected an increase in firing synchrony with long-term facilitation (Fig. 3C). Bins along the diagonal of the surprise matrix with significantly high counts (P < 0.01) were clustered in the upper right quadrant, which contains data for the period following induction of long-term facilitation. The firing rate of the reference neurone decreased with long-term facilitation (from 27.6 ± 2.5 to 24.2 ± 2.7 spikes s-1, vertical firing rate histogram, Fig. 3C), whilst the firing rate of the target neurone increased (15.9 ± 2.3 to 17.4 ± 2.7 spikes s-1, horizontal firing rate histogram). The corresponding Gaussian-smoothed diagonal histogram of surprise values in the right panel also shows that the synchrony was greater with long-term facilitation.
The dynamic correlational properties of seven raphe neurones were also evaluated in detail. Changes in effective connectivity associated with long-term facilitation within this raphe neuronal assembly were detected with two complementary approaches. Modified JPSTH analyses of effective connectivity among five of the neurones, represented in the firing rate histograms of Fig. 1A and designated R1 to R5 for convenience, revealed changes associated with long-term facilitation. Surprise matrices derived from normalized JPSTHs calculated from three sets of concatenated spike data identified increased firing synchrony in two pairs of neurones (Fig. 4A and B), and a reduction in short time scale asynchrony in a third. This latter change was represented in the surprise matrix as a reduction in the number of contiguous paradiagonal bins with significantly low counts (Fig. 4C). This result was consistent with differences in ordinary cross-correlograms for this pair calculated from spike samples recorded before and after induction of the facilitation (Fig. 4E).
 |
View larger version [in this window] [in a new window] |
|
|
A, B and C, surprise matrices of JPSTHs calculated from 1340 interleaved 1 s spike train slices, each 0.5 s from before followed by 0.5 s after induction of long-term facilitation (see Methods). Only significantly high (A and B) or low (C) bin counts are displayed. A, left, surprise matrix with significantly high (P < 0.01) bin counts clustered along the upper right quadrant diagonal. Right, corresponding high bin counts in the right half of the Gaussian-smoothed (4 bins) histogram of diagonal counts in the surprise matrix. These features indicate increased firing synchrony of two raphe neurones after induction of long-term facilitation; the corresponding cross-correlogram central peak is shown in the upper right. Control segments: 4360 reference (R2), 585 target spikes (R1); long-term facilitation segments: 7718 reference, 1592 target spikes. B, surprise matrix and related data for raphe neurones R3 and R2 have greater paradiagonal clustering of significant bins (P < 0.01) following induction of long-term facilitation, indicative of increased synchrony. Control segments: 13 878 reference (R3), 4360 target spikes (R2); long-term facilitation segments: 15 701 reference, 4309 target spikes. C, surprise matrix with significantly low (P < 0.05) paradiagonal bin counts indicative of reduced asynchrony of raphe neurones R4 and R5 during expression of the respiratory memory. Control segments: 2299 reference (R4), 13 868 target spikes (R5); long-term facilitation segments: 3379 reference, 9693 target spikes. D, correlation feature map indicates primary cross-correlogram features for a subset of correlated raphe neurones as described in the text. Each rectangle on the right shows the identifier and firing rate change with long-term facilitation of the represented neurone. Symbols on the line arising from each rectangle (from the bottom) represent the primary correlogram features (see key) found in cross-correlograms calculated using the corresponding neuron as the reference cell and the neurone represented just above the symbol as the target cell. E, ordinary cross-correlograms for the neurones represented in C. Note the reduced 'anti-correlated' firing of the two neurones (central trough) after induction of long-term facilitation (dotted line histogram) relative to control (continuous line). F, average of phrenic nerve activity triggered by 20 040 spikes in the tonic E-Dec neurone designated R2 represented in A, B and D. The arrow indicates a central peak; the broad slope to the left reflects the inspiratory to expiratory phase transition.
| ||
Ordinary cross-correlation analysis had previously detected non-random timing relationships represented in the peaks and troughs summarized in the correlation feature map (Fig. 4D). These five neurones all had respiratory modulated firing rates. Cell R4 was inspiratory modulated; the others were classified as expiratory neurones. Central peaks in averages of phrenic nerve activity triggered by R2 (Fig. 4F) and R3 (not shown) suggested that those neurones and phrenic motoneurones were influenced by functionally antecedent shared inputs. An offset trough and an offset peak (see Fig. 2 of Morris et al. 1996a) were consistent with gain modulation of phrenic motoneurone activity by neurones R4 and R5, respectively.
The gravity method and related pattern detection procedures also detected changes in the non-random temporal relationships among the raphe neurones following induction of long-term facilitation. Whilst the JPSTH measured the average synchronous activity of pairs of neurones, the gravitational representation screened the multiple spike trains, as a group, for non-random timing relationships visualized as variations in the aggregation velocities of particles corresponding to each neurone. The method can detect dynamic fluctuations in synchrony that may not be apparent in measures of averages such as the cross-correlogram and the JPSTH. Each of five samples was constructed by concatenation of two data segments, each 1 min in duration, from before and after long-term facilitation induction. An interval equal to 4 times the charge decay time constant used in the gravity calculation was inserted between the two concatenated segments so that spikes on opposite sides of the concatenation boundary did not influence particle clustering.
In each sample, particles representing six of the neurones aggregated significantly, whilst the distance between particle 4, corresponding to neurone R4, and all other particles increased. Labelled circles in the last frame of an animated projection from one data set show final particle positions (Fig. 5A). Trails can be followed back to initial positions in the projected space. Because of information loss in such projections from the N space, the distance between each pair of particles was plotted as a function of time and evaluated for significance. The graph of all particle pair distances shows both the condensation and repulsion. All pairs for which the distance increased included particle 4; the other pairs aggregated significantly (Fig. 5B).
 |
View larger version [in this window] [in a new window] |
|
|
A, final frame of an animated projection of the particle trajectories from the N space to a plane from gravity analysis of 7 simultaneously monitored raphe neurone spike trains; labelled circles show final particle positions. Sample included 62 s of control activity concatenated with 62 s of data recorded after induction of long-term facilitation. An interval equal to 4 times the charge decay time constant used in the gravity calculation was added to the end of the control segment to prevent spikes on opposite sides of the concatenation boundary from influencing particle aggregation. Calculation parameters: acceptor charge decay forward; effector charge decay backward; decay time constant 8.0 ms. Numbers of spikes: 1, 120; 2, 848; 3, 2823; 4, 548; 5, 998; 6, 2029; 7, 1328. Particles 1-5 represent spikes in neurons R1-R5 in Fig. 4. B, particle distance as a function of time plot shows distances between all pairs represented in A. Labelled subset of distance plots represents all pairs that included the particle representing neuron R4 (designated 4 in this group); these pairs were characterized by repulsion - and the resultant increased interparticle distance. The dashed line marks the concatenation of control and long-term facilitation spike data. C, two recurring patterns of synchrony (designated a and b) displayed as two sets of aggregation velocity vectors with common origins at their times of occurrence. Pattern a (plotted in grey) occurred only before long-term facilitation, pattern b only after.
| ||
To identify excessively repeating patterns of correlated activity, the records of interparticle distance were divided into 1000 increments and the average aggregation velocity during each of these 120 ms intervals was calculated. The resulting array was screened using template matching algorithms. Patterns that had more matches in the real data set than in 100 randomized control data sets were reported as significant (see Methods). The average number of matching patterns in the five samples before long-term facilitation was 18.8 ± 3.6 (S.D.) There were 56.8 ± 10.0 (S.D.) significant patterns that repeated during both control and long-term facilitation, whilst 11.8 ± 3.9 (S.D.) patterns were found only after long-term facilitation induction.
Patterns that recurred with a frequency greater than that of any of the shifted data sets were displayed as sets of vectors, each with a common origin and at the time of occurrence. Vector length indicated the aggregation velocity; vector direction served solely to identify the neurone pair represented. Figure 5C shows two sets of patterns found in the sample represented in Fig. 5A and B. Pattern a recurred only before long-term facilitation; pattern b was detected during expression of the respiratory memory.
Figure 6A shows the times of occurrence of all recurring patterns in one of the five samples tested. The vertical lines in each row of the plot indicate the times of occurrence of a particular repeated pattern of synchrony; rows were sorted by the time of the first occurrence of each set of patterns. The patterns below the dotted line occurred only after induction of long-term facilitation. As an additional control for this study, the original aggregation velocity data set for each pair of particles generated from each of the five samples was first shuffled randomly and then searched for patterns that repeated more often than chance at the P < 0.01 confidence level. In contrast to the original data (e.g. Fig. 6A), no more than three patterns in any pair-wise shuffled data set passed the test of significance. Results from one such trial show the three patterns in the pair-wise shuffled data set that recurred more often than in 100 additional shuffled data sets (Fig. 6B).
 |
View larger version [in this window] [in a new window] |
|
|
A, each row of lines represents one set of recurring spark patterns that repeated more often than in any of 100 shuffled samples from the same data set; a subset of these patterns were detailed in Fig. 5. Seven patterns recurred exclusively during long-term facilitation (below dotted line). B, a test for the likelihood of false positives detected only three sets of patterns that recurred in a pair-wise shuffled sample more often than in 100 additional shuffled samples.
| ||
|
DISCUSSION |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionReferences |
|---|
The results of this study show that the firing rates and correlational properties of neurones recorded in multiple respiratory-related medullary sites change following induction of a respiratory memory by repeated stimulation of the carotid chemoreceptors. The detected changes were appropriate to contribute to the expression of long-term facilitation (Fig. 7A). The persistent increase in integrated phrenic motoneurone activity was accompanied by altered activity in inspiratory neurones recorded in the caudal VRG and DRG, sites with high concentrations of premotor neurones. Offset peaks in the spike-triggered averages of multiunit efferent phrenic activity (Fig. 2C) were consistent with a premotor function of inspiratory neurones with increased firing rates. The averaging procedure measures changes due to direct and indirect pathways. Given the parallel nature of the premotor system (Davies et al. 1985), the observed offset peaks may have reflected the contribution of unobserved neurones firing in partial synchrony with the particular neurone used to trigger the average. Troughs observed in some averages triggered by neurones that decreased activity with long-term facilitation (Fig. 7A) were suggestive of functional inhibitory connections or shared connections of opposite sign. These results suggest that disinhibition of premotor or phrenic motoneurones also plays a role in long-term facilitation.
 |
View larger version [in this window] [in a new window] |
|
|
A, support for the hypothesis that distributed brainstem processes contribute to the expression of the respiratory memory include: (1) observed increases in the firing rates of putative DRG and VRG premotor neurones and modulatory raphe neurones (up arrows in representative 'cell bodies'), (2) increased synchrony or effective connectivity (circled arrows), and (3) increases in shared input of opposite sign or decreased premotor inhibition from I-Decr neurones (see Fig. 2D). B, some inferred sites of altered effective connectivity within a raphe correlational assembly. C, graphical representation of a possible mechanism for the 'ratcheting' of long-term facilitation with transient repeated episodes of peripheral chemoreceptor stimulation. Some potential sites of modulation are shown in A. See text for further discussion.
| ||
Long-term facilitation was also associated with changes in the impulse synchrony of inspiratory neurone pairs characterized by central cross-correlogram peaks and paradiagonal features in the surprise matrix derived from the joint peristimulus time histogram (JPSTH). More coincident impulses in premotor neurones may provide increased drive to phrenic motoneurones through summation of their synaptic influences (Fig. 7A). Increased effective connectivity was measured between rostral and caudal VRG neurones (Fig. 3B and C), extending previous observations (Morris et al. 1996b).
The firing rates and synchrony of simultaneously recorded raphe neurones also changed following induction of long-term facilitation. The non-random timing relationships detected with ordinary cross-correlation analysis, JPSTHs and the gravity method support the hypothesis that raphe medullary neuronal assemblies are reconfigured following the induction of long-term facilitation. Pattern detection methods revealed transient correlational assemblies, including distinct configurations observed after induction of the facilitation.
Some inferred relationships among the raphe neurones represented in Fig. 4 are shown in Fig. 7B. The increased firing synchrony of neurone pairs R2-R1 and R3-R2 following induction of long-term facilitation (Fig. 4A and B) was consistent with enhanced efficacy of a shared input. Although modulation of a shared inhibition could generate such altered spike timing relationships, changes in mutual excitation would also be consistent with the observed increases in firing rates of the three neurones. Given the asymmetry of the correlogram peaks, excitatory interactions between the pairs, such as R3 driving R2, are also plausible. The observed increase in R3 firing rate and/or increased synaptic efficacy would then have contributed to the increase in R2 firing rate.
A parsimonious interpretation of the significantly low bin counts in the surprise matrix calculated for neurones R4 and R5 (Fig. 4C) and the corresponding trough in the ordinary cross-correlogram (Fig. 4E) includes a shared input with opposite actions on the two neurones. A decrease in inhibition of R4 and a concomitant shared decrease in excitation of R5 could have contributed to the persistent firing rate changes observed in those spike trains.
The measured changes in effective connectivity and firing rate and the inferred relationships are consistent with previous work that has implicated serotonergic neurones in the expression of long-term facilitation (Millhorn et al. 1980; Holtman et al. 1986). Many of the raphe neurones with increased activity had low firing rates. A significant proportion of neurones with firing rates less than 4 Hz have been characterized as serotonergic in raphe obscurus (Woch et al. 1996) and other raphe nuclei (Mason, 1997). Available data are consistent with the hypothesis that raphe neurones have parallel modulatory influences (Fig. 7A and B) in the medulla (Woch et al. 1996; Lindsey et al. 1998) and spinal cord (Holtman et al. 1986; Mitchell et al. 1992; Hayashi et al. 1993; Lindsay & Feldman, 1993). Although serotonin has been associated with blocking of long-term potentiation at some brain sites (Villani & Johnston, 1993), it has been implicated in the long-term facilitation of phrenic motor activity (Millhorn et al. 1980). Modulatory transmitters, such as serotonin, could act on synaptic mechanisms (e.g. Michael et al. 1998; Beaumont & Zucker, 2000) or alter the excitability of the premotor neurones directly (Wallen et al. 1989; Berger et al. 1992).
The efficiency of multi-array recording technology permitted the evaluation of the concurrent responses of relatively many neurones before and after the induction of long-term facilitation, a process that 'saturates' and could only be performed once or twice per animal. Induction of respiratory long-term facilitation requires repeated brief, not extended, stimulation (Dwinell et al. 1997; Turner & Mitchell, 1997; Baker & Mitchell, 2000) similar to the stimulus protocols necessary for long-term potentiation induction (Lu et al. 1999). Other stimulus parameters can lead to short-term facilitation or short-term depression of respiration (Powell et al. 1998). The response profiles of concurrently monitored raphe neurones to carotid chemoreceptor stimulation vary; the firing rates of some increase transiently at the onset of carotid chemoreceptor stimulation, and decline as a delayed increase in others develops (Morris et al. 1996a).
These distinct response properties and related short time scale correlations are consistent with a 'ratchet' model (Fig. 7C). Neurones with the transient response contribute to the generation of the facilitated state during periods of high firing rate. Inhibitory actions of the delayed responders could play a role in limiting the amount of potentiation induced with each stimulus. This hypothesis is consistent with the more general concept that mid-line respiratory-related neuronal assemblies use recurrent inhibitory connections in an internal equilibrium-seeking system that acts to stabilize and regulate the gain of respiratory motor output (Morris et al. 1996a).
The present results do not preclude roles for neurones at other sites in the induction or expression of long-term facilitation (Gozal et al. 1999). Other research has suggested that persistent motor neurone membrane changes are important for the maintenance of long-term facilitation (Fuller et al. 2000a). The neurone responses measured here are from recording segments pre- and post-stimulation that were of the order of several minutes in duration. It is probable that the contribution to long-term facilitation of these adaptive neural networks continues to change over the expression of a memory shown to persist for hours.
The facilitation considered in this study constitutes only one of several responses of the respiratory network to hypoxia. Available data suggest that apnoea and gasping associated with prolonged hypoxia also involve distributed mechanisms (Richter et al. 1999; Solomon et al. 2000). Preceding bouts of episodic hypoxia may also influence these processes (Gozal & Gozal, 1999).
|
REFERENCES |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionReferences |
|---|
| AERTSEN A. M. H. J. & GERSTEIN, G. L. (1985). Evaluation of neuronal connectivity: sensitivity of cross-correlation. Brain Research 340, 341-354 | [Medline] |
| AERTSEN A. M. H. J., GERSTEIN, G. L., HABIB, M. K. & PALM, G. (1989). Dynamics of neuronal firing correlation: Modulation of "effective connectivity". Journal of Neurophysiology 61, 900-917 | [Medline] |
| BACH K. B. & MITCHELL, G. S. (1996). Hypoxia-induced long-term facilitation of respiratory activity is serotonin dependent. Respiration Physiology 104, 251-260 | [Medline] |
| BAKER T. L. & MITCHELL, G. S. (2000). Episodic but not continuous hypoxia elicits long-term facilitation of phrenic motor output in rats. Journal of Physiology 529, 215-219 | [Abstract/Full Text] |
| BEAUMONT V. & ZUCKER, R. S. (2000). Enhancement of synaptic transmission by cyclic AMP modulation of presynaptic Ih channels. Nature Neuroscience 3, 133-141 | [Medline] |
| BERGER A. J., BAYLISS, D. A. & VIANA, F. (1992). Modulation of neonatal rat hypoglossal motoneuron excitability by serotonin. Neuroscience Letters 143, 164-168 | [Medline] |
| BIANCHI A. L., DENAVIT-SAUBIE, M. & CHAMPAGNAT, J. (1995). Central control of breathing in mammals: neuronal circuitry, membrane properties, and neurotransmitters. Physiological Reviews 75, 1-45 | [Medline] |
| COHEN M. I., PIERCEY, M. F., GOOTMAN, P. M. & WOLOTSKY, P. (1974). Synaptic connections between medullary inspiratory neurons and phrenic motoneurons as revealed by cross-correlation. Brain Research 81, 319-324 | [Medline] |
| CONNELLY C. A., ELLENBERGER, H. H. & FELDMAN, J. L. (1989). Are there serotonergic projections from the raphe and retrotrapeziod nuclei to the ventral respiratory group in the rat? Neuroscience Letters 105, 34-40 | [Medline] |
| DAVIES J. G. MCF., KIRKWOOD, P. A. & SEARS, T. A. (1985). The distribution of monosynaptic connexions from inspiratory bulbospinal neurons to inspiratory motoneurones in the cat. Journal of Physiology 368, 63-87 | [Abstract] |
| DWINELL M. R., JANSSEN, P. L. & BISGARD, G. E. (1997). Lack of long-term facilitation of ventilation after exposure to hypoxia in goats. Respiration Physiology 108, 1-9 | [Medline] |
| FULLER D. D., BACH, K. B., BAKER, T. L., KINKEAD, R. & MITCHELL, G. S. (2000a). Long term facilitation of phrenic motor output. Respiration Physiology 121, 135-146 | [Medline] |
| FULLER D. D., ZABKA, A. G., BAKER, T. L., FREGOSI, R. F. & MITCHELL, G. S. (2000b). Phrenic long-term facilitation requires 5-HT2 receptor activation during but not following episodic hypoxia. Society for Neuroscience Abstracts 26, 556 | |
| GERSTEIN G. L. & AERTSEN, A. M. H. J. (1985). Representation of cooperative firing activity among simultaneously recorded neurons. Journal of Neurophysiology 54, 1513-1528 | [Medline] |
| GERSTEIN G. L., PERKEL, D. J. & DAYHOFF, J. E. (1985). Cooperative firing activity in simultaneously recorded populations of neurons: detection and measurement. Journal of Neuroscience 5, 881-889 | [Abstract] |
| GOZAL D. & GOZAL, E. (1999). Episodic hypoxia enhances late hypoxic ventilation in developing rat: putative role of neuronal NO synthase. American Journal of Physiology 276, R17-22 | [Medline] |
| GOZAL D., XUE, Y. D. & SIMAKAJORNBOON, N. (1999). Hypoxia induces c-Fos protein expression in NMDA but not AMPA glutamate receptor labeled neurons within the nucleus tractus solitarii of the conscious rat. Neuroscience Letters 262, 93-96 | [Medline] |
| HAYASHI F., COLES, S. K., BACH, K. B., MITCHELL, G. S. & MCCRIMMON, D. R. (1993). Time-dependent phrenic nerve responses to carotid afferent activation: intact vs. decerebellate rats. American Journal of Physiology 265, R811-819 | [Medline] |
| HOLTMAN J. R. JR, DICK, T. E. & BERGER, A. J. (1986). Involvement of serotonin in the excitation of phrenic motoneurons evoked by stimulation of the raphe obscurus. Journal of Neuroscience 6, 1185-1193 | [Abstract] |
| JACOBS B. L. & AZMITIA, E. C. (1992). Structure and function of the brain-serotonin system. Physiological Reviews 72, 165-229 | [Medline] |
| KORTEN J. B. & HADDAD, G. G. (1989). Respiratory waveform pattern recognition using digital techniques. Computers in Biology and Medicine 19, 207-217 | [Medline] |
| LINDSAY A. D. & FELDMAN, J. L. (1993). Modulation of respiratory activity of neonatal rat phrenic motoneurones by serotonin. Journal of Physiology 461, 213-233 | [Abstract] |
| LINDSEY B. G., ARATA, A., MORRIS, K. F., HERNANDEZ, Y. M. & SHANNON, R. (1998). Medullary raphe neurones and baroreceptor modulation of the respiratory motor pattern in the cat. Journal of Physiology 512, 863-882 | [Abstract/Full Text] |
| LINDSEY B. G., HERNANDEZ, Y. M., MORRIS, K. F., SHANON, R. & GERSTEIN, G. L. (1992). Respiratory related neural assemblies in the brainstem midline. Journal of Neurophysiology 67, 905-922 | [Medline] |
| LINDSEY B. G., MORRIS, K. F., SHANNON, R. & GERSTEIN, G. L. (1997). Repeated patterns of distributed synchrony in neuronal assemblies. Journal of Neurophysiology 78, 1714-1719. | [Abstract/Full Text] |
| LU Y. F., KANDEL, E. R. & HAWKINS, R. D. (1999). Nitric oxide signaling contributes to late-phase LTP and CREB phosphorylation in the hippocampus. Journal of Neuroscience 19, 10250-10261 | [Abstract/Full Text] |
| MASON P. (1997). Physiological identification of pontomedullary serotonergic neurons in the rat. Journal of Neurophysiology 77, 1087-1098 | [Abstract/Full Text] |
| MICHAEL D., MARTIN, K. C., SEGER, R., NING, M. M., BASTON, R. & KANDEL, E. R. (1998). Repeated pulses of serotonin required for long-term facilitation activate mitogen-activated protein kinase in sensory neurons of Aplysia. Proceedings of the National Academy of Sciences of the USA 95, 1864-1869 | [Abstract/Full Text] |
| MILLHORN D. E., ELDRIDGE, F. L. & WALDROP, T. G. (1980). Prolonged stimulation of respiration by endogenous central serotonin. Respiration Physiology 42, 171-188 | [Medline] |
| MITCHELL G. S., SLOAN, H. E., JIANG, C., MILETIC, V., HAYASHI, F. & LIPSKI, J. (1992). 5-Hydroxytryptophan (5-HTP) augments spontaneous and evoked phrenic motoneuron discharge in spinalized rats. Neuroscience Letters 141, 75-78. | [Medline] |
| MONTEAU R., MORIN, D., HENNEQUIN, S. & HILAIRE, G. (1990). Differential effects of serotonin on respiratory activity of hypoglossal and cervical motoneurons: an in vitro study on the newborn rat. Neuroscience Letters 111, 127-132 | [Medline] |
| MORRIS K. F., ARATA, A., SHANNON, R. & LINDSEY, B. G. (1993). Concurrent, distributed alterations in brainstem neural assemblies during induction of long-term enhancement of respiratory activity in vivo by carotid chemoreceptor stimulation. Society for Neuroscience Abstracts 19, 574.4 | |
| MORRIS K. F., ARATA, A., SHANNON, R. & LINDSEY, B. G. (1996a). Long-term facilitation of phrenic nerve activity in cats: responses and short time scale correlations of medullary neurones. Journal of Physiology 490, 463-480 | [Abstract] |
| MORRIS K. F., ARATA, A., SHANNON, R. & LINDSEY, B. G. (1996b). Inspiratory drive and phase duration during carotid chemoreceptor stimulation in the cat: medullary neurone correlations. Journal of Physiology 491, 241-259 | [Abstract] |
| MORRIS K. F., SHANNON, R. & LINDSEY, B. G. (1996c). Evidence for change of respiratory phase variance and changes in synchrony of raphe neuron assemblies during long term facilitation. The Physiologist 39, 186 | |
| MORRISON S. F. & GEBBER, G. L. (1984). Raphe neurons with sympathetic-related activity: baroreceptor responses and spinal connections. American Journal of Physiology 246, R338-348 | [Medline] |
| OREM J. & DICK, T. (1983). Consistency and signal strength of respiratory neuronal activity. Journal of Neurophysiology 50, 1098-1107 | [Medline] |
| PALKOVITS M., BROWNSTEIN, M. & SAAVEDRA, J. M. (1974). Serotonin content of the brain stem nuclei in the rat. Brain Research 80, 237-249 | [Medline] |
| PERKEL D. H., GERSTEIN, G. L. & MOORE, G. P. (1967). Neuronal spike trains and stochastic point processes II. Simultaneous spike trains. Biophysical Journal 7, 419-440. | [Medline] |
| POWELL F. L., MILSOM, W. K. & MITCHELL, G. S. (1998). Time domains of the hypoxic ventilatory response. Respiration Physiology 112, 123-134 | [Medline] |
| REKLING J. C. & FELDMAN, J. L. (1998). Pre-Bötzinger complex and pacemaker neurons: hypothesized site and kernel for respiratory rhythm generation. Annual Review of Physiology 60, 385-405 | [Abstract/Full Text] |
| RICHTER D. W., SCHMIDT-GARCON, P., PIERREFICHE, O., BISCHOFF, A. M. & LALLEY, P. M. (1999). Neurotransmitters and neuromodulators controlling the hypoxic respiratory response in anaesthetized cats. Journal of Physiology 514, 567-578. | [Abstract/Full Text] |
| SEGERS L. S., SHANNON, R., SAPORTA, S. & LINDSEY, B. G. (1987). Functional associations among simultaneously monitored lateral medullary respiratory neurons in the cat. I: Evidence for excitatory and inhibitory actions of inspiratory neurons. Journal of Neurophysiology 57, 1078-1100 | [Medline] |
| SMITH J. C., MORRISON, D. E., ELLENBERGER, H. H., OTTO, M. R. & FELDMAN, J. L. (1989). Brainstem projections to the major respiratory neuron populations in the medulla of the cat. Journal of Comparative Neurology 281, 69-96 | [Medline] |
| SOLOMON I. C., EDELMAN, N. H. & NEUBAUER, J. A. (2000). Pre-Botzinger complex functions as a central hypoxia chemosensor for respiration in vivo. Journal of Neurophysiology 83, 2854-2868 | [Abstract/Full Text] |
| TURNER D. L. & MITCHELL, G. S., (1997). Long-term facilitation of ventilation following repeated hypoxic episodes in awake goats. Journal of Physiology 499, 543-550 | [Abstract] |
| VILLANI F. & JOHNSTON, D. (1993). Serotonin inhibits induction of long-term potentiation at commissural synapses in hippocampus. Brain Research 606, 304-308 | [Medline] |
| WALLEN P., BUCHANAN, J. T., GRILLNER, S., HILL, R. H., CHRISTENSON, J. & HOKFELT, T. (1989). Effects of 5-hydroxytryptamine on the afterhyperpolarization, spike frequency regulation, and oscillatory membrane properties in lamprey spinal cord neurons. Journal of Neurophysiology 61, 759-768 | [Medline] |
| WOCH G., DAVIES, R. O., PACK, A. I. & KUBIN, L. (1996). Behaviour of raphe cells projecting to the dorsomedial medulla during carbachol-induced atonia in the cat. Journal of Physiology 490, 745-758. | [Abstract] |
Acknowledgements
This work was supported by National Institute of Neural Disorders and Stroke grant NS19814. The authors thank J. Gilliland, C. Orsini, K. D. Morris, D. Baekey and R. McGowan for excellent technical support.
Corresponding author
K. F. Morris: Department of Physiology and Biophysics, University of South Florida Health Sciences Center, 12901 Bruce B. Downs Boulevard, Tampa, FL 33612-4799, USA.
Email: kmorris{at}hsc.usf.edu
This article has been cited by other articles:
|
|
 |
|
  M. McGuire, C. Liu, Y. Cao, and L. Ling Formation and maintenance of ventilatory long-term facilitation require NMDA but not non-NMDA receptors in awake rats J Appl Physiol, September 1, 2008; 105(3): 942 - 950. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 C. Julien, A. Bairam, and V. Joseph Chronic intermittent hypoxia reduces ventilatory long-term facilitation and enhances apnea frequency in newborn rats Am J Physiol Regulatory Integrative Comp Physiol, April 1, 2008; 294(4): R1356 - R1366. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 J. E. Dayhoff Computational properties of networks of synchronous groups of spiking neurons. Neural Comput., September 1, 2007; 19(9): 2433 - 2467. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 T. E. Dick, Y.-H. Hsieh, N. Wang, and N. Prabhakar Acute intermittent hypoxia increases both phrenic and sympathetic nerve activities in the rat Exp Physiol, January 1, 2007; 92(1): 87 - 97. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 M. McGuire, Y. Zhang, D. P. White, and L. Ling Phrenic long-term facilitation requires NMDA receptors in the phrenic motonucleus in rats J. Physiol., September 1, 2005; 567(2): 599 - 611. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
J. Duffin Functional organization of respiratory neurones: a brief review of current questions and speculations Exp Physiol, September 1, 2004; 89(5): 517 - 529. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
M. McGuire, Y. Zhang, D. P. White, and L. Ling Serotonin receptor subtypes required for ventilatory long-term facilitation and its enhancement after chronic intermittent hypoxia in awake rats Am J Physiol Regulatory Integrative Comp Physiol, February 1, 2004; 286(2): R334 - R341. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
K. F. Morris, D. M. Baekey, S. C. Nuding, T. E. Dick, R. Shannon, and B. G. Lindsey Plasticity in Respiratory Motor Control: Invited Review: Neural network plasticity in respiratory control J Appl Physiol, March 1, 2003; 94(3): 1242 - 1252. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
D. L. Young, F. L. Eldridge, and C.-S. Poon Integration-differentiation and gating of carotid afferent traffic that shapes the respiratory pattern J Appl Physiol, March 1, 2003; 94(3): 1213 - 1229. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
G. S. Mitchell and S. M. Johnson Plasticity in Respiratory Motor Control: Invited Review: Neuroplasticity in respiratory motor control J Appl Physiol, January 1, 2003; 94(1): 358 - 374. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
D. M. Baekey, K. F. Morris, S. C. Nuding, L. S. Segers, B. G. Lindsey, and R. Shannon Medullary raphe neuron activity is altered during fictive cough in the decerebrate cat J Appl Physiol, January 1, 2003; 94(1): 93 - 100. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
R. W. Bavis and G. S. Mitchell Plasticity in Respiratory Motor Control: Selected Contribution: Intermittent hypoxia induces phrenic long-term facilitation in carotid-denervated rats J Appl Physiol, January 1, 2003; 94(1): 399 - 409. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||
