Skip to main content
NCBI home page
The Journal of Physiology logoLink to The Journal of Physiology
. 2000 Mar 1;523(Pt 2):449–457. doi: 10.1111/j.1469-7793.2000.00449.x

Coherent rhythmic discharges in sympathetic nerves supplying thermoregulatory circulations in the rat

Julia E Smith 1, Michael P Gilbey 1
PMCID: PMC2269801  PMID: 10699088

Abstract

  1. In anaesthetised rats, activity recorded from sympathetic postganglionic neurones innervating the tail circulation has characteristic rhythmicity (0·4-1·2 Hz). At the population level this rhythmicity can be seen as a peak (T-peak) in autospectra of sympathetic activity recorded from ventral collector nerves (VCNs).

  2. Here we investigated whether nerves supplying thermoregulatory circulations share common rhythmic discharges at T-peak frequency. Activity was recorded from nerve pairs consisting of left ventral collector nerve (LVCN) and one of the following: right ventral collector nerve (RVCN), left dorsal collector nerve (DCN), left saphenous nerve (SN) or left renal nerve (RN).

  3. During central apnoea, T-peak frequencies in RVCN autospectra were similar to those of simultaneously recorded LVCN and these activities were coherent. Similar observations were made for nerve pairs involving LVCN-DCN and LVCN-SN. In contrast, autospectra of RN activity did not contain T-peaks.

  4. In comparison to the peaks in autospectra of RN activity, when the frequency of rhythmic phrenic nerve activity was manipulated T-peaks in VCN, DCN and SN autospectra did not show obligatory 1:1 locking.

  5. We conclude that T-peaks are a robust feature of autospectra of sympathetic discharges supplying thermoregulatory circulation but not those influencing the kidney. The high coherence demonstrated between the T-peak discharges is consistent with the view that common/coupled oscillators located within the CNS influence cutaneous vasoconstrictor sympathetic activity.

Under conditions of heightened sympathetic drive, discharges recorded from human cutaneous nerves are often rhythmic (Delius et al. 1972; Bini et al. 1980). We have demonstrated similarly that sympathetic discharges recorded from postganglionic neurones (PGNs) innervating the caudal ventral artery (CVA) of the rat tail (thermoregulatory) circulation show rhythmic discharges (T-rhythm; see Chang et al. 1999b). The discharges of these PGNs appear to be influenced by multiple oscillators that have a frequency range of 0.4-1.2 Hz (Chang et al. 1999b). When the activity of a population of ventral collector nerve (VCN) PGNs supplying the tail is recorded and subjected to spectral analysis, the rhythmic discharges are revealed as a prominent peak in the autospectrum (T-peak; Chang et al. 1999b). Therefore, the T-peak can be used as a marker for the T-rhythm.

The rhythmic discharges recorded from PGNs influencing the tail circulation can be entrained or reset by various inputs (e.g. central respiratory drive, and lung inflation related, cardiovascular mechanoreceptor and somatic afferents; Johnson & Gilbey, 1998; Chang et al. 1999a,b; Staras et al. 1999). As the strength and timing (with respect to the oscillator cycle) of such inputs determines their effect on an oscillator population (e.g. see Glass & Mackey, 1988; Kelso, 1995), it has been suggested that neuronal oscillators, through dynamic and graded coupling, may provide one mechanism for generating complex patterns of sympathetic nervous response (see Chang et al. 1999b).

In order to gain further insight into the nervous control of cutaneous circulations, in this study we investigated whether nerves that innervate thermoregulatory circulations share common rhythmic discharges at T-peak frequency. Correlated activity would indicate that the rhythmic discharges arise from common or coupled sources (Kenney et al. 1991; Gebber et al. 1995). Consequently, we made paired recordings of sympathetic discharges from left VCNs (LVCNs) and one of the following: right VCN (RVCN), left dorsal collector nerve (DCN) or left saphenous nerve (excluding skeletal muscle branches; SN). Collector nerves and SNs innervate the vasculature of the tail and distal portions of hindlimb and foot, respectively (Greene, 1955; Jänig, 1985; Chang et al. 1999b). As sympathetic activity recorded from these nerves is switched off during hyperthermia (Johnson & Gilbey, 1994; authors' unpublished observations), the activities are considered to be representative of those influencing thermoregulatory circulations (see Key & Wigfield, 1994). Renal nerve (RN) activity was used as a visceral (non-thermoregulatory) comparison where activity is increased during hyperthermia (Kenney et al. 1995).

We found a robust T-peak in the discharges of VCNs, DCNs and SNs but not RNs. Activities at T-peak were always significantly coherent in nerve pairs consisting of LVCN and RVCN, DCN or SN, indicating that common/coupled oscillators within the CNS drive their activities. Some of this work has been published as abstracts (Smith & Gilbey, 1998b, 1999).

METHODS

Anaesthesia and animal maintenance

Experiments were performed on 31 male Sprague-Dawley rats (230–300 g) under Project and Personal licences issued by the Home Office. Animals were anaesthetised initially with sodium pentobarbitone (60 mg kg−1i.p.) and supplemented with α-chloralose (5–10 mg i.v.) as required. The depth of anaesthesia was monitored continuously and was judged from: (i) stability of heart rate, blood pressure, phrenic nerve activity (or respiratory related movements); (ii) pupil size; and (iii) palpebral and paw pinch reflexes. Animals were paralysed (gallamine triethiodide, 16 mg kg−1 h−1i.v.) during periods of data collection. During such periods, the depth of anaesthesia was assessed from the stability of blood pressure and phrenic nerve activity. At the end of experiments animals were killed by an overdose of sodium pentobarbitone (i.v.).

The right femoral artery and vein were cannulated to monitor arterial blood pressure and to administer drugs, respectively. The trachea was cannulated. The animals were then vagotomised, given a bilateral pneumothorax and ventilated artificially (rate, 1.2-2 Hz; O2-enriched room air; positive pressure applied to expiratory line). Peak expiratory CO2 was monitored continuously and arterial blood samples (75 μl) were taken after each 300 s data collection period. Oesophageal temperature was monitored and maintained at 37.0 ± 0.5°C. The bladder was cannulated to allow free passage of urine. Phrenic nerve activity was recorded as an indicator of central respiratory drive. For further details see Johnson & Gilbey (1996).

Recordings of sympathetic activity

VCNs and DCNs contain both sympathetic and somatic motor axons. Thus, to enable recordings of only sympathetic activity from these nerves, animals had their cauda equina transected (vertebrae L5-L6) to decentralise the somatic motor innervation of the tail whilst leaving the sympathetic supplies intact (see Smith et al. 1998). SN sympathetic activity was recorded from cutaneous branches isolated below the level of the knee. The kidney was exposed by a retroperitoneal approach and a RN identified at the root of the renal artery and traced to the kidney. The sympathetic efferent nature of nerve activity was confirmed by abolition following ganglionic blockade with chlorisondamine chloride (Ecolid/SU3088; a gift from CIBA-Geigy Corporation, USA; 3 mg kg−1i.v.) or trimetaphan cansylate (Arfonad, Roche Products Ltd, UK; 50 mg kg−1i.v.).

VCN, DCN, SN and RN nerve activities were recorded differentially using conventional bipolar electrodes (see Chang et al. 1999b). Nerves were either covered with liquid paraffin or embedded in dental impression material (President light body, Coltène/Whaledent Ltd, UK).

Data capture

All nerve activities were recorded through high impedance headstages (NL 100, Neurolog; Digitimer Ltd, UK), preamplified (10000–20000 times; NL 104, Neurolog), filtered (300–1000 Hz; NL 25, Neurolog) and amplified (×10). All nerve activities were then rectified and smoothed (time constant, 100 ms; NL 703, Neurolog). Processing nerve recordings in this manner generates an envelope of sympathetic activity with the advantage of removing movement-related artefacts (see Chang et al. 1999b).

All nerve discharges, blood pressure, tracheal pressure and TTL (transistor-transistor logic) pulses were monitored continuously on an IBM compatible computer (Viglen, UK) using a 1401 plus interface and Spike2 software (Cambridge Electronic Design, UK) and digital storage oscilloscopes (VG-6023, Hitachi, Japan). Nerve and pressure signals were also digitised (11800 samples s−1 channel−1; VR100-B; Instrutech, NY, USA) and recorded on videotape for off-line analysis.

Data analysis

The presence of rhythmic sympathetic discharges recorded from whole nerves and the degree of correlation between paired nerve recordings were assessed using frequency domain analysis. Three hundred seconds of integrated whole nerve activity was sampled at 100 Hz using the Spike2 computer software. Such data files were converted into text files and analysed using Matlab computer software (Maths Works). Fast Fourier transformation (FFT; size 2048, 50 % overlap) was performed on 286.72 s of integrated whole nerve activity. This divided the data sets into 28 half-overlapped 20.48 s long subsections with 2048 data points in each, giving a frequency resolution of 0.049 Hz. DC components and linear trends were then removed from the subsections. An autospectrum was averaged from these 28 subsections according to the Welch Method (see Chatfield, 1996).

Coherence spectra were used to reveal linear correlation between sympathetic activities at LVCN T-peak frequency recorded from nerve pairs. A coherence value significantly different from zero indicates that discharges at that particular frequency are linearly related. Coherence spectra were averaged from the same 28 half-overlapped subsections used to generate autospectra. The squared coherence coefficient (referred to as coherence value) at each frequency was estimated by normalising the cross-spectrum between two nerve activities (see Chatfield, 1996).

We used a 95 % confidence level threshold value of 0.1 for non-zero coherence. This was generated using the equation:

graphic file with name tjp0523-0449-mu1.jpg

(Rosenberg et al. 1989), where S is the threshold level for non-zero coherence, P is probability (i.e. 0.05) and L is the number of sections used to calculate coherence.

Definition of a peak in an autospectrum

A peak was allocated to an autospectrum where the envelope of spectral density that the peak was associated with increased by more than 50 % over six bins (∼0.3 Hz) above levels outside T-rhythm frequency range (0.4-1.2 Hz; see Chang et al. 1999b).

The frequency resolution of auto- and coherence spectra used in this study was 0.049 Hz. The frequency of a peak on a spectrum was assessed (using the Matlab software) from a cursor positioned by an operator. The frequency reading (to two decimal places) was recorded (see Fig. 1). These estimates of peak frequencies were used rather than relating the read frequency to a particular 0.049 Hz bin. However, the limits of frequency resolution were considered when T-peak frequencies in autospectra from nerve pairs were compared (see Fig. 2).

Figure 1. Neurograms, autospectra and coherence spectra of nerve activities recorded during central apnoea.

Figure 1

Open in a new tab

Rhythmic sympathetic burst discharges at T-peak frequency were a prominent feature of smoothed and rectified neurograms of sympathetic activity recorded from RVCNs (Aa), DCNs (Ab) and SNs (Ac) during central apnoea. Frequency domain analysis of the same recordings in A revealed a prominent T-peak (star) in autospectra (Ba-c). RVCN, DCN and SN T-peak frequencies were similar to those recorded simultaneously from LVCNs (Ca, b and c, respectively). Paired LVCN-RVCN, LVCN-DCN and LVCN-SN activities displayed significant coherence (i.e. coherence values were > 0.1, see text) at LVCN T-peak frequency (Da, b and c, respectively). In contrast, rhythmic burst discharges were not such a robust feature of sympathetic activity recorded from RNs during central apnoea (Ad). Moreover, paired LVCN-RN activities did not display significant coherence at LVCN T-peak frequency (Dd).

Figure 2. Plots of T-peak frequencies from simultaneously recorded nerve pairs.

Figure 2

Open in a new tab

Nerve pairs: LVCN-RVCN (^, 4 animals), LVCN-DCN (□, 4 animals) and LVCN-SN (▵, 5 animals). Each point represents data collected from different preparations during central apnoea. Since the frequency resolution of the autospectra was 0.049 Hz, T-peak frequencies on paired nerve recordings were considered to be similar if the difference between them was < 0.05 Hz (area between interrupted lines).

Allocation of dominant nerve burst discharge frequency

The dominant nerve discharge frequency was allocated normally to the frequency with the highest peak. However, when the frequency of the highest peak coincided with the 1st harmonic frequency (i.e. twice the fundamental frequency) of phrenic nerve burst discharge frequency and there was another peak which was > 50 % of its size (8/94 ‘sympathetic nerve’ autospectra), dominant nerve discharge frequency was allocated to the 2nd highest peak. This criterion was applied to ensure that dominant nerve discharge frequency was not allocated inappropriately to a harmonic frequency of rhythmic phrenic nerve activity. (Application of frequency domain analysis to nerve activity (which is not a sinusoidal waveform) generates spectral density at harmonic frequencies of rhythmic nerve discharges; Chatfield, 1996.)

Linear regression analysis

Linear regression analysis was performed to test whether dominant nerve burst discharge frequency displayed a linear 1:1 relationship with phrenic nerve burst discharge frequency. Least squares linear regression analysis was performed using Microcal Origin 3.54 (Microcal Software Inc.). The slopes of regression lines were compared to 1 using Student's t test (Glantz, 1996).

Calculation of mean arterial blood pressure

Mean arterial blood pressure (MAP) was calculated using the following equation:

graphic file with name tjp0523-0449-mu2.jpg

Diastolic and systolic blood pressure values were averaged from six values taken from the data set at 60 s intervals. The median mean arterial blood pressure of the 62 data sets used in this study was 84 mmHg (range, 70–106 mmHg).

RESULTS

Autospectra and coherence during central apnoea

We first compared autospectra of nerve pairs and their coherence during central apnoea (indicated by the absence of rhythmic phrenic activity). T-peaks were present in the autospectra of discharges of VCNs, DCNs and SNs, but not in those of RNs. In this study T-peaks were observed in the frequency range 0.64-1.03 Hz (median, 0.88 Hz; 44 nerve recordings from 31 animals).

LVCNs and RVCNs

During central apnoea (arterial O2 levels, 100–440 mmHg; CO2 levels, 17–36 mmHg; and arterial pH range, 7.4-7.6), discharges of variable amplitude were evident in LVCN sympathetic activity, like those observed in RVCN activity (Fig. 1). Spectral analysis revealed a prominent T-peak (median frequency, 0.86 Hz; range, 0.73-0.94 Hz; 6 animals). In 4/6 animals, LVCN and RVCN activities were recorded concurrently. The autospectra generated from one such paired recording are shown in Fig. 1. In all cases, T-peak frequency recorded from the LVCN (0.73-0.94 Hz; 4 animals) was similar to T-peak frequency recorded simultaneously from the RVCN (0.78-0.92 Hz) (see Fig 1 and Fig 2). LVCN→RVCN coherence at LVCN T-peak frequency was always significant (Fig. 3). A typical LVCN→RVCN coherence spectrum is shown in Fig. 1Da. In this and all other cases coherence was abolished by blockade of ganglionic transmission.

Figure 3. Coherence at T-peak frequency in different nerve pairs.

Figure 3

Open in a new tab

Coherence values for paired nerve recordings at LVCN T-peak frequency during central apnoea. Each point represents data collected from different preparations during central apnoea. #, 3 superimposed RN points. 95 % CL = 95 % confidence level for coherence values to be significantly different from zero.

LVCNs and DCNs

Similar to activity recorded from VCNs, discharges of variable amplitude were evident in DCN recordings and spectral analysis revealed a T-peak in all autospectra (Fig. 1; median DCN T-peak frequency, 0.86 Hz; range, 0.77-1.03 Hz; 9 animals). In 4/9 animals, activity was recorded concurrently from LVCNs and DCNs. T-peak frequency was similar in autospectra of both nerve activities (Figs 1 and 2). LVCN→DCN coherence at LVCN T-peak frequency was always significant (Figs 1 and 3).

LVCNs and SNs

As with VCN and DCN recordings, discharges of variable amplitude were evident in SN neurograms and T-peaks were present in SN autospectra (Figs 1 and 2; median SN T-peak frequency, 0.91 Hz; range, 0.82-1.02 Hz; 11 animals). In 5/11 animals, sympathetic activity was recorded simultaneously from LVCNs and SNs. These paired LVCN-SN recordings displayed significant coherence at LVCN T-peak frequency (Figs 1 and 3).

Coherence at frequencies other than LVCN T-peak frequency

Paired LVCN-RVCN, LVCN-DCN and LVCN-SN recordings could also display significant coherence at frequencies other than LVCN T-peak frequency (see Fig. 1D). Such coherence can be generated by harmonics of T-peak frequency and by other activity in the nerve recordings. However, for the purposes of the present study, the coherence generated at T-peak frequency is of greatest interest as: (i) most of the power in VCN, DCN and SN autospectra is in the T-peak (see Fig. 1B and C); and (ii) only the coherence at LVCN T-peak frequency can be used to support the hypothesis that the T-rhythm arises centrally.

LVCNs and RNs

In contrast to simultaneously recorded LVCN activity, autospectra generated from RN sympathetic activity did not display a peak (n= 5 animals; Fig. 1). Furthermore, LVCN→RN coherence at LVCN T-peak frequency was not significantly different from zero in 4/5 cases (Figs 1 and 3). In the one experiment in which significant coherence was present at LVCN T-peak frequency, it occurred for only one bin (0.049 Hz), thus casting doubt on its significance: contrast with LVCN→RVCN, LVCN→DCN and LVCN→SN coherence spectra where coherence was sustained (Fig. 1).

Effect of manipulating rhythmic phrenic nerve burst discharge frequency on autospectra of sympathetic activity

Rhythmic phrenic nerve burst discharge frequency (from central apnoea to ∼1.0 Hz) was altered by manipulating the fraction of inspired O2 and artificial ventilation frequency. In this manner arterial levels of O2 (104–440 mmHg) and CO2 (19–45 mmHg) were manipulated. It has been shown that, with blood gases in these ranges, T-rhythm oscillators are not entrained faithfully to central respiratory drive (Smith & Gilbey, 1998a; Chang et al. 1999b).

During periods of rhythmic phrenic nerve activity, peaks at the same frequency (and often its harmonic frequencies) appeared in VCN, SN and DCN autospectra. Such peaks could coexist with peaks at other frequencies within the T-rhythm frequency range (Fig. 4). In autospectra where there was more than one peak, T-peak frequency was assigned to the highest peak (unless this frequency coincided with the 1st harmonic of rhythmic phrenic nerve burst discharge frequency; see Methods).

Figure 4. Autospectra for a paired LVCN-SN recording in which rhythmic phrenic nerve activity was manipulated.

Figure 4

Open in a new tab

A-D, autospectra generated during central apnoea and when phrenic nerve burst discharge frequency was 0.49, 0.63 and 0.73 Hz, respectively. During periods of phrenic nerve burst discharge, peaks at frequencies related to phrenic nerve burst discharge frequency (diamonds: fundamental and harmonics) appeared in SN and LVCN autospectra. T-peak frequency (star) was always allocated to the frequency of the highest peak, unless it coincided with the 1st harmonic of phrenic nerve burst discharge frequency (see text). SN T-peak frequency was the same as phrenic nerve burst discharge frequency when the latter was 0.73 Hz (D) but, like LVCN T-peak frequency, did not follow phrenic nerve burst discharge frequency when the latter was reduced (A-C). PN, phrenic nerve.

From the example shown in Fig. 4, it can be seen that when the frequency of phrenic nerve burst discharge was shifted the dominant peak in LVCN and SN autospectra failed to follow faithfully, although there are indications of intermittent 1:1 lock (see Chang et al. 1999b). The T-peak frequencies of paired LVCN-SN and LVCN-DCN recordings are plotted against phrenic nerve burst discharge frequency in Fig. 5. Linear regression analysis failed to demonstrate a 1:1 co-variance between the frequencies of rhythmic phrenic nerve burst discharge and T-peak frequencies in VCN, DCN and SN autospectra, i.e. the slopes of the regression lines were significantly different from 1 (P < 0.001).

Figure 5. Regression analysis on relationships between peak frequencies in phrenic and sympathetic autospectra.

Figure 5

Open in a new tab

A-C, T-peak frequencies of LVCN (^), DCN (□), and SN (▵) activities did not display a positive 1:1 linear relationship with phrenic nerve burst discharge frequency whereas RN sympathetic discharge frequency (▾) did (C). Linear regression lines: A, LVCN (dotted line): c= 0.82 ± 0.04 Hz, m= 0.02 ± 0.08, n= 19, P < 0.001; SN (continuous line): c= 0.91 ± 0.04 Hz, m=−0.19 ± 0.08, n= 19, P < 0.001. B, LVCN (dotted line): c= 0.82 ± 0.03 Hz, m= 0.001 ± 0.06, n= 13, P < 0.001; DCN (continuous line): c= 0.86 ± 0.04 Hz, m=−0.10 ± 0.08, n= 13, P < 0.001. C, LVCN (dotted line): c= 0.76 ± 0.04 Hz, m= 0.10 ± 0.07, n= 17, P < 0.001; RN (continuous line): c= 0.03 ± 0.03 Hz, m= 0.96 ± 0.05, n= 12, P= 0.206. c is the intercept of the line on the y-axis, m is the slope and n is the number of data points. Values are means ±s.e.m.

In contrast to autospectra from simultaneously recorded LVCN activity, peaks in RN autospectra were always at rhythmic phrenic nerve burst discharge frequency and its harmonic frequencies (Fig. 6). T-peak and RN dominant discharge frequencies from paired LVCN-RN recordings are plotted against phrenic nerve burst discharge frequency in Fig. 5C. Linear regression analysis demonstrated that, unlike LVCN T-peak frequency, the dominant frequency of RN sympathetic discharges displayed a significant positive 1:1 linear relationship with phrenic nerve burst discharge frequency (intercept = 0.03 ± 0.03 Hz, slope = 0.96 ± 0.05, number of data points = 12, P= 0.206).

Figure 6. Autospectra for a paired LVCN-RN recording in which rhythmic phrenic nerve activity was manipulated.

Figure 6

Open in a new tab

A-D, autospectra generated during central apnoea and when phrenic nerve burst discharge frequency was 0.49, 0.62 and 0.79 Hz, respectively. In contrast to LVCN T-peak frequency (star), the peaks in RN autospectra were always related to phrenic nerve burst discharge frequency (diamonds). PN, phrenic nerve.

DISCUSSION

In this study we have demonstrated that the T-peak, observed previously in the autospectra of sympathetic discharges recorded from VCNs (Chang et al. 1999b), is also a robust feature of the autospectra of sympathetic discharges recorded from DCNs and SNs that likewise influence thermoregulatory circulations. Paired recordings involving these nerves also revealed coherent activities at T-peak frequency. In marked contrast, autospectra of RN activity did not show T-peaks.

On the source of T-peak discharges

In the absence of central respiratory drive, the activities of nerve pairs (LVCN-RVCN, LVCN-DCN and LVCN-SN), at T-peak frequency, were correlated linearly (i.e. displayed significant coherence). Therefore it appears that, in part, a common neuronal substrate or coupled neuronal substrates must underlie the discharges of these nerve pairs. Coherent discharges at T-peak frequency may be generated by common or coupled sources themselves, or through ‘filtering’ or by entrainment at a particular ratio of co-ordination by neurones downstream from ‘master oscillator(s)’ (Selverston & Moulins, 1985; Kocsis et al. 1990).

Pre- and postganglionic neurones do not appear to be appropriately coupled to form the networks required to explain the coherent activity we observed in nerve pairs. First, PGNs projecting into the SN and VCN arise from different ganglia (lumbar and sacral ganglia, respectively: Sittiracha et al. 1987; Baron et al. 1988). Second, approximately 95 % of PGNs projecting into LVCN and RVCN arise from ipsilateral ganglia (Sittiracha et al. 1987). Third, although interganglionic pathways have been suggested (Wolff et al. 1993) there are no reports of similar connections between sacral and lumbar ganglia. Fourth, LVCNs and RVCNs appear to be influenced only by ipsilateral sympathetic preganglionic neurones as responses were evoked in VCNs only following electrical stimulation on the ipsilateral side of the spinal cord (Smith & Gilbey, 1998a). Fifth, different populations of preganglionic sympathetic neurones regulate VCNs and SNs as evidenced by their response to noxious pinch demonstrating ‘local sign’ (Jänig, 1985; Häbler et al. 1994; M. P. Gilbey, unpublished observations). Lastly, although coupling between ipsilateral sympathetic neurones has been demonstrated, there is no evidence for coupling between sympathetic preganglionic neurones located on opposite sides of the spinal cord (Logan et al. 1996; Spanswick et al. 1998; Nolan et al. 1999). Therefore it is probable that antecedent neuronal networks to sympathetic preganglionic neurones form the substrate underlying the generation of coherent activity. Preganglionic neurones and/or PGNs, however, might be involved in the generation of discharges in the T-rhythm range in response to common/coupled rhythmic inputs from a master oscillator(s). Indeed, the known properties of sympathetic preganglionic neurones (Dembowsky et al. 1986; Yoshimura et al. 1987; Lewis & Coote, 1990; Gilbey, 1997; Nolan et al. 1999) and putative paravertebral vasoconstrictor PGNs (Cassell et al. 1986) are consistent with this idea: both can generate self-limiting bursts of action potentials. In this manner the intrinsic burst cycle of these neurones could be entrained by master oscillator(s) to produce T-rhythm discharges. An alternative possibility is that T-rhythm oscillators are located at antecedent neurones to preganglionic neurones, the rhythm being either transmitted faithfully or refined at pre- and postganglionic levels.

Relationship between T-peak discharge and central respiratory drive frequencies

T-peak frequencies, in VCN, DCN and SN autospectra, did not show a 1:1 linear relationship with phrenic nerve burst discharge frequency. This is in agreement with observations that T-rhythm oscillators (under similar experimental conditions) did not maintain a stable entrainment to central respiratory drive (see Smith & Gilbey, 1998a; Chang et al. 1999b). Chang et al. (1999b) demonstrated that, within the arterial range of CO2 used in the present study (19–45 mmHg), the T-peak in VCN autospectra showed frequency drifting between ‘free run’ and phrenic nerve burst discharge frequency. This same phenomenon is indicated in this study: peaks often occurred in autospectra of VCNs, DCNs and SNs at both free run T-rhythm frequency and phrenic nerve burst discharge frequencies. This is indicative of dynamic coupling of T-rhythm oscillators to central respiratory drive (Chang et al. 1999b).

Relationship between central respiratory drive and discharges in RNs

During periods of rhythmic phrenic nerve activity, the peaks in RN autospectra always coincided with rhythmic phrenic discharge frequency and its harmonic frequencies. Thus, there was no indication of dynamic coupling between RN and central respiratory drive. Such a relationship with rhythmic phrenic nerve burst discharge frequency is consistent with RN activity being patterned directly by, rather than dynamically coupled to, central respiratory networks (e.g. Bachoo & Polosa, 1987; Haselton & Guyenet, 1989; Richter & Spyer, 1990; Spyer, 1996).

On the T-peak as a signature of sympathetic discharges controlling thermoregulatory circulations

In agreement with our observations on RNs, other workers have observed a similar distribution of power in the autospectra of activities recorded from RNs (Taylor & Schramm, 1987; Kenney et al. 1995, 1998) and other visceral nerves including splanchnic, splenic and lumbar sympathetic chain (Gebber et al. 1989; Allen et al. 1993; Kenney et al. 1998). In addition, Häbler et al. (1999) have observed on occasion, in PGNs dissected from VCNs, a discharge pattern resembling T-rhythm discharges. Therefore, it is possible that the T-rhythm is peculiar to sympathetic discharges controlling thermoregulatory circulations. As appealing as this idea is, under our experimental conditions the sympathetic drive to thermoregulatory circulations (VCNs, DCNs and SNs) is probably near maximal whereas that to RNs is relatively low. Consequently, it is possible that T-rhythm discharges could be an emergent characteristic of RN sympathetic discharges during periods of heightened sympathetic drive to the kidney.

In conclusion, this study has shown that, in the anaesthetised rat, T-peaks are a robust feature of autospectra of sympathetic discharges supplying thermoregulatory circulations but not those influencing the kidney. This raises the possibility that the T-peak is a signature of cutaneous vasoconstrictor activity. As high coherence was demonstrated between T-peak discharges of LVCN-RVCN, LVCN-DCN and LVCN-SN nerve pairs, we suggest a central location for the oscillators. We do not, however, exclude the possibility that preganglionic and/or postganglionic neurones generate discharges in the T-rhythm range dependent upon common/coupled inputs from antecedent master oscillators.

Acknowledgments

J.E.S. was supported by British Heart Foundation grant FS/96009. We thank Hong-Shui Chang for purpose writing the Matlab scripts and Bruce Cotsell for his technical support.

References

  1. Allen AM, Adams JM, Guyenet PG. Role of the spinal cord in generating the 2- to 6-Hz rhythm in rat sympathetic outflow. American Journal of Physiology. 1993;264:R938–945. doi: 10.1152/ajpregu.1993.264.5.R938. [DOI] [PubMed] [Google Scholar]
  2. Bachoo M, Polosa C. Lack of evidence of coupled oscillator mechanisms in the generation of sympathetic rhythms. In: Ciriello J, Calaresu FR, Renaud LP, Polosa C, editors. Organization of the Autonomic Nervous System. Vol. 31. New York: Alan R. Liss; 1987. pp. 189–202. [Google Scholar]
  3. Baron R, Jänig W, Kollmann W. Sympathetic and afferent somata projecting in hind limb nerves and the anatomical organisation of the lumbar sympathetic nervous system of the rat. Journal of Comparative Neurology. 1988;275:460–468. doi: 10.1002/cne.902750310. [DOI] [PubMed] [Google Scholar]
  4. Bini G, Hagbarth K-E, Hynninen P, Wallin BG. Thermoregulatory and rhythm-generating mechanisms governing the sudomotor and vasoconstrictor outflow in human cutaneous nerves. The Journal of Physiology. 1980;306:537–552. doi: 10.1113/jphysiol.1980.sp013413. [DOI] [PMC free article] [PubMed] [Google Scholar]
  5. Cassell JF, Clark AL, Mclachlan EM. Characteristics of phasic and tonic sympathetic ganglion cells of the guinea-pig. The Journal of Physiology. 1986;372:457–483. doi: 10.1113/jphysiol.1986.sp016020. [DOI] [PMC free article] [PubMed] [Google Scholar]
  6. Chang H-S, Staras K, Gilbey MP. Complex dynamics of periodically driven sympathetically driven oscillators. Society for Neuroscience Abstracts. 1999a;25:759.7. [Google Scholar]
  7. Chang H-S, Staras K, Smith JE, Gilbey MP. Sympathetic neuronal oscillators are capable of dynamic synchronisation. Journal of Neuroscience. 1999b;19:3183–3197. doi: 10.1523/JNEUROSCI.19-08-03183.1999. [DOI] [PMC free article] [PubMed] [Google Scholar]
  8. Chatfield C. The Analysis of Time Series. London: Thomson Science and Professional; 1996. [Google Scholar]
  9. Delius W, Hagbarth K-E, Hongell A, Wallin BG. Manoeuvres affecting sympathetic outflow in human skin nerves. Acta Physiologica Scandinavica. 1972;84:177–186. doi: 10.1111/j.1748-1716.1972.tb05168.x. [DOI] [PubMed] [Google Scholar]
  10. Dembowsky K, Czachurski J, Seller H. Three types of sympathetic preganglionic neurones with different electrophysiological properties are identified by intracellular recordings in the cat. Pflügers Archiv. 1986;406:112–120. doi: 10.1007/BF00586671. [DOI] [PubMed] [Google Scholar]
  11. Gebber GL, Barman SM, Zviman M. Sympathetic activity remains synchronized in presence of a glutamate antagonist. American Journal of Physiology. 1989;256:R722–732. doi: 10.1152/ajpregu.1989.256.3.R722. [DOI] [PubMed] [Google Scholar]
  12. Gebber GL, Zhong S, Barman SM. Synchronization of cardiac-related discharges of sympathetic nerves with inputs from widely separated spinal segments. American Journal of Physiology. 1995;268:R1472–1483. doi: 10.1152/ajpregu.1995.268.6.R1472. [DOI] [PubMed] [Google Scholar]
  13. Gilbey MP. Fundamental aspects of the control of sympathetic preganglionic neuronal discharge. In: Jordan D, editor. Central Nervous Control of Autonomic Function. The Netherlands: Harwood Academic Publishers; 1997. pp. 1–28. [Google Scholar]
  14. Glantz SA. Primer of Biostatistics. New York: McGraw Hill; 1996. [Google Scholar]
  15. Glass L, Mackey MC. From Clocks to Chaos: The Rhythms of Life. West Sussex: Prince Town University Press; 1988. [Google Scholar]
  16. Greene EC. Anatomy of the Rat. New York: Hafner Publishing Co.; 1955. [Google Scholar]
  17. Häbler H-J, Bartsch T, Jänig W. Rhythmicity in single fiber postganglionic activity supplying the rat tail. Journal of Neurophysiology. 1999;81:2026–2036. doi: 10.1152/jn.1999.81.5.2026. [DOI] [PubMed] [Google Scholar]
  18. Häbler H-J, Jänig W, Krummel M, Peters OA. Reflex patterns in postganglionic neurons supplying skin and skeletal muscle of the rat hind limb. Journal of Neurophysiology. 1994;72:2222–2236. doi: 10.1152/jn.1994.72.5.2222. [DOI] [PubMed] [Google Scholar]
  19. Haselton JR, Guyenet PG. Central respiratory modulation of medullary sympathoexcitatory neurons in rat. American Journal of Physiology. 1989;256:R739–750. doi: 10.1152/ajpregu.1989.256.3.R739. [DOI] [PubMed] [Google Scholar]
  20. Jänig W. Organization of the lumbar sympathetic outflow to skeletal muscle and the skin of the cat hind limb and tail. Reviews of Physiology, Biochemistry and Pharmacology. 1985;102:119–213. doi: 10.1007/BFb0034086. [DOI] [PubMed] [Google Scholar]
  21. Johnson CD, Gilbey MP. Sympathetic activity recorded from the rat caudal ventral artery in vivo. The Journal of Physiology. 1994;476:437–442. doi: 10.1113/jphysiol.1994.sp020145. [DOI] [PMC free article] [PubMed] [Google Scholar]
  22. Johnson CD, Gilbey MP. On the dominant rhythm in the discharges of single postganglionic sympathetic neurones innervating the rat tail artery. The Journal of Physiology. 1996;497:241–259. doi: 10.1113/jphysiol.1996.sp021764. [DOI] [PMC free article] [PubMed] [Google Scholar]
  23. Johnson CD, Gilbey MP. Effects of aortic nerve stimulation on discharges of sympathetic neurones innervating rat tail artery and vein. American Journal of Physiology. 1998;275:R942–949. doi: 10.1152/ajpregu.1998.275.4.R942. [DOI] [PubMed] [Google Scholar]
  24. Kelso JAS. Dynamic Patterns: the Self-Organization of Brain and Behaviour. Cambridge, MA, USA.: MIT Press; 1995. [Google Scholar]
  25. Kenney MJ, Barman SM, Gebber GL, Zhong S. Differential relationships among discharges of postganglionic sympathetic nerves. American Journal of Physiology. 1991;260:R1159–1167. doi: 10.1152/ajpregu.1991.260.6.R1159. [DOI] [PubMed] [Google Scholar]
  26. Kenney MJ, Barney CC, Hirai T, Gisolfi CV. Sympathetic nerve responses to hyperthermia in the anesthetized rat. Journal of Applied Physiology. 1995;78:881–889. doi: 10.1152/jappl.1995.78.3.881. [DOI] [PubMed] [Google Scholar]
  27. Kenney MJ, Claassen DE, Bishop MR, Fels RJ. Regulation of the sympathetic nerve discharge bursting pattern during heat stress. American Journal of Physiology. 1998;275:R1992–2001. doi: 10.1152/ajpregu.1998.275.6.R1992. [DOI] [PubMed] [Google Scholar]
  28. Key BJ, Wigfield CC. The influence of the ventrolateral medulla on thermoregulatory circulations in the rat. Journal of the Autonomic Nervous System. 1994;48:79–89. doi: 10.1016/0165-1838(94)90162-7. [DOI] [PubMed] [Google Scholar]
  29. Kocsis B, Gebber GL, Barman SM, Kenney MJ. Relationships between activity of sympathetic nerve pairs: phase and coherence. American Journal of Physiology. 1990;259:R549–560. doi: 10.1152/ajpregu.1990.259.3.R549. [DOI] [PubMed] [Google Scholar]
  30. Lewis DI, Coote JH. Excitation and inhibition of rat preganglionic neurones by catecholamines. Brain Research. 1990;530:229–234. doi: 10.1016/0006-8993(90)91287-q. [DOI] [PubMed] [Google Scholar]
  31. Logan SD, Pickering AE, Gibson IC, Nolan MF, Spanswick D. Electronic coupling between rat sympathetic neurones in vitro. The Journal of Physiology. 1996;495:491–502. doi: 10.1113/jphysiol.1996.sp021609. [DOI] [PMC free article] [PubMed] [Google Scholar]
  32. Nolan MF, Logan SD, Spanswick D. Electrophysiological properties of electrical synapses between rat sympathetic preganglionic neurones in vitro. The Journal of Physiology. 1999;519:753–764. doi: 10.1111/j.1469-7793.1999.0753n.x. [DOI] [PMC free article] [PubMed] [Google Scholar]
  33. Richter DW, Spyer KM. Cardiorespiratory control. In: Loewy AD, Spyer KM, editors. Central Regulation of Autonomic Functions. New York: Oxford University Press; 1990. pp. 349–366. [Google Scholar]
  34. Rosenberg JR, Amjad AM, Breeze P, Brillinger DR, Halliday DM. The Fourier approach to the identification of functional coupling between neural spike trains. Progress in Biophysics and Molecular Biology. 1989;53:1–31. doi: 10.1016/0079-6107(89)90004-7. [DOI] [PubMed] [Google Scholar]
  35. Selverston AI, Moulins M. Oscillatory neural networks. Annual Review of Physiology. 1985;47:29–48. doi: 10.1146/annurev.ph.47.030185.000333. [DOI] [PubMed] [Google Scholar]
  36. Sittiracha T, Mclachlan EM, Bell C. The innervation of the caudal artery of the rat. Neuroscience. 1987;21:647–659. doi: 10.1016/0306-4522(87)90150-3. [DOI] [PubMed] [Google Scholar]
  37. Smith JE, Gilbey MP. Rhythmical sympathetic discharges and escape behaviour. Brain Research. 1998a;811:169–172. doi: 10.1016/s0006-8993(98)00974-3. [DOI] [PubMed] [Google Scholar]
  38. Smith JE, Gilbey MP. Rhythmical discharges in the sympathetic supply to rat tail are of central origin. The Journal of Physiology. 1998b;513.P:81P. [Google Scholar]
  39. Smith JE, Gilbey MP. The robust component of the rhythmical discharges recorded from the sympathetic outflow to the tail display high coherence with the sympathetic outflow to the hindlimb skin but not the kidney. The Journal of Physiology. 1999;515.P:143–144P. [Google Scholar]
  40. Smith JE, Jansen ASP, Gilbey MP, Loewy AD. CNS cell groups projecting to sympathetic outflow to tail artery: The major heat loss system of the rat. Brain Research. 1998;786:153–164. doi: 10.1016/s0006-8993(97)01437-6. [DOI] [PubMed] [Google Scholar]
  41. Spanswick D, Renaud LP, Logan SD. Bilaterally evoked monosynaptic EPSPs, NMDA receptors and potentiation in rat sympathetic preganglionic neurones in vitro. The Journal of Physiology. 1998;509:195–209. doi: 10.1111/j.1469-7793.1998.195bo.x. [DOI] [PMC free article] [PubMed] [Google Scholar]
  42. Spyer KM. Central nervous integration of cardiorespiratory control. In: Greger R, Windhorst U, editors. Comprehensive Human Physiology. Vol. 2. Berlin Heidelberg: Springer-Verlag; 1996. pp. 2129–2144. [Google Scholar]
  43. Staras K, Chang H-S, Gilbey MP. Somatic afferent stimulation resets ongoing rhythmical discharges of single postganglionic sympathetic neurones (PGNs) innervating the caudal ventral artery (CVA) of the rat. The Journal of Physiology. 1999;518.P:27P. [Google Scholar]
  44. Taylor RF, Schramm LP. Differential effects of spinal transection on sympathetic nerve activities in rats. American Journal of Physiology. 1987;253:R611–618. doi: 10.1152/ajpregu.1987.253.4.R611. [DOI] [PubMed] [Google Scholar]
  45. Wolff JR, Kasa P, Dobo E, Romgens HJ, Parducz A, Joo F, Wolff A. Distribution of GABA-immunoreactive nerve fibers and cells in the cervical and thoracic paravertebral sympathetic trunk of adult rat: evidence for an ascending feed-forward inhibition system. Journal of Comparative Neurology. 1993;334:281–293. doi: 10.1002/cne.903340209. [DOI] [PubMed] [Google Scholar]
  46. Yoshimura M, Polosa C, Nishi S. Noradrenaline induces rhythmic bursting in sympathetic preganglionic neurons. Brain Research. 1987;420:147–151. doi: 10.1016/0006-8993(87)90250-2. [DOI] [PubMed] [Google Scholar]

Articles from The Journal of Physiology are provided here courtesy of The Physiological Society

RESOURCES