| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Departments of Biomedical Engineering and Neuroscience, The Johns Hopkins University School of Medicine, Baltimore, Maryland 21205-2196
 |
ABSTRACT |
|---|
 Abstract
 Introduction
Methods
Results
Discussion
References
|
|---|
Chau, David, Namjin Kim, and Lawrence P. Schramm. Sympathetically correlated activity of dorsal horn neurons in spinally transected rats. J. Neurophysiol. 77: 2966-2974, 1997. In mammals with an intact neuraxis, most sympathetic nerve activity is generated by brain stem systems. Therefore these systems have attracted much more attention than spinal systems that generate excitatory inputs to sympathetic preganglionic neurons. The purpose of this study was to determine whether, within hours of C1 spinal cord transection, spinal dorsal horn neurons (DHNs) play a role in generating sympathetic nerve activity. Experiments were conducted in chloralose-anesthetized rats. We recorded renal sympathetic nerve activity (RSNA) in the left renal nerve, and we recorded the activity of neurons located in the left dorsal horn at T2, T8, T10, T13, and L2. We also recorded the activity of neurons in the right dorsal horn at T10. The somatic fields and cutaneous modalities of most neurons were determined. Spike-triggered averaging was used to determine relationships between the ongoing activity of DHNs and ongoing RSNA. In the left dorsal horn, bursts of ongoing activity of 16% of DHNs at T8 and 43% of DHNs at T10 were positively correlated with bursts of ongoing RSNA at latencies of 59 ± 8 (SE) ms. At no other level on the left side, nor in the T10 segment on the right side, was the activity of DHNs correlated with RSNA. DHNs with activity correlated with RSNA were located only in dorsal horn laminae III-V. Deeper laminae were not investigated in these experiments. The activity of all sympathetically correlated DHNs exhibited bursts of action potentials with interspike intervals of <10 ms. All but one of the sympathetically correlated DHNs exhibited wide-dynamic-range modalities. The modalities of sympathetically uncorrelated neurons were more heterogeneous. Brief (5-10 s) noxious cutaneous stimulation of mid- and lower thoracic dermatomes on the left side excited all sympathetically correlated DHNs and simultaneously increased RSNA. The excitatory cutaneous fields of sympathetically correlated neurons were circumscribed by the excitatory fields for RSNA. The excitatory cutaneous fields of some sympathetically uncorrelated DHNs extended beyond the excitatory fields for RSNA. Noxious cutaneous stimulation of the extremities on the left side that decreased RSNA simultaneously decreased the activity of all sympathetically correlated DHNs. These data provide electrophysiological evidence that, in spinally transected rats, a population of DHNs may generate or convey excitatory input to renal sympathetic preganglionic neurons.
Supraspinal systems generate most sympathetic activity in spinally intact mammals (Alexander 1946 Surgery
Adult male Sprague-Dawley rats (Taconic Farms and Charles River), each weighing 250-350 g, were used in these experiments. All procedures for these experiments were approved by the Animal Care and Use Committee of the Johns Hopkins University School of Medicine (protocol 94M242). Pretreatment with atropine (0.05 mg/kg sc) reduced nasal and tracheal mucus secretion. Anesthesia was induced by ether inhalation and continued by
Extracellular and renal nerve recording
The extracellular recordings were made with single-barrel carbon fiber microelectrodes (impedance 2-4 M Somatic stimulation
Innocuous somatic stimulation (light brush with a cotton applicator) and noxious somatic stimulation (pinch with toothed forceps, maintained for 5-10 s) were delivered to sites delineated by the reference grid (Fig. 6). A response in either the RSNA or the single-neuron activity was considered excitatory or inhibitory if it represented an increase or decrease, respectively, of Classification of neurons' afferent modalities
Neurons were classified as low threshold (LT) if they were excited by brush and not further affected by pinch, wide dynamic range (WDR) if they were excited by brush and excited more by pinch, and high threshold (HT) if they were excited only by pinch. Six DHNs were inhibited by either innocuous or noxious stimulation of their (normally excitatory) primary dermatomes, or they responded to neither brush nor pinch. We classified these neurons as "other".
Spike-triggered averaging
Spike-triggered averages of ongoing RSNA were calculated for all recorded neurons. On detection of a neuron's action potential, an epoch of renal nerve activity (rectified and filtered, time constant = 0.04 s) was extracted from the continuous renal nerve recording. The epoch began 300 ms before the occurrence of the action potential and lasted 700 ms after the action potential. Further unit detection was disabled for
Interspike interval analysis
Interspike interval histograms (ISIHs, maximum interval 1 s, bin size 10 ms) were generated for all neurons. The histograms were computed using either 4 min of recordings of ongoing neuronal discharges or 1,000 action potentials, whichever occurred first. The degree of skewness (Sokal and Rohlf 1969 Histology
At the end of experiments, rats were perfused transcardially with buffered saline, followed by 10% buffered formaldehyde. The relevant spinal cord segments were removed and stored in sucrose-formaldehyde solution (30% sucrose in 10% phosphate-buffered formaldehyde, pH 7.4) for 2-5 days. Transverse 40-µm sections were cut on a sliding microtome, mounted on gelatin-coated glass slides, and air dried. The sites of electrolytic lesions were identified microscopically.
Data presentation and statistical analysis
Data are expressed as means ± SE. All statistical analyses employed either a one-tailed Fisher's exact test or a Sample of DHNs
In 56 spinally transected rats we simultaneously recorded the ongoing activity of the left renal sympathetic nerve and the activity of single DHNs, ipsilateral and contralateral to the renal nerve recording (Table 1). Spike-triggered averages were computed for all DHNs listed in Table 1. Afferent modalities of DHNs were based on the neurons' responses to stimulation of their estimated primary dermatomes (METHODS). The fields from which unitary responses could be elicited by noxious stimuli were completely surveyed for a subset of these neurons (Table 1, Somatic fields fully surveyed). The excitatory and inhibitory somatic fields of this subset of DHNs were later compared with fields that, when stimulated by pinch, increased or decreased RSNA (Fig. 6A).
After spinal cord transection, the ongoing activity of only those DHNs recorded in segments containing renal sympathetic preganglionic neurons (Schramm et al. 1993) was correlated to ongoing RSNA
When DHN-triggered averages were compared with averages triggered by frequency-matched dummy neurons, it was clear that the action potentials of a subset of these neurons regularly preceded bursts of RSNA (Fig. 1A). The spike-triggered averages of a much larger subset of DHNs (Fig. 1B) exhibited no relationship between the firing of the spinal neurons and RSNA. A correlation index was used to measure the degree of correlation between the activity of DHNs and RSNA (METHODS). A population of DHNs with an average correlation index of ~0.9 existed at each spinal level investigated (Fig. 2). Spike-triggered averages generated from nonoverlapping segments of the records from these neurons rarely exhibited a consistent peak at any latency. In the ipsilateral T8 and T10 segments, however, we observed DHNs with substantially higher correlation indexes. More neurons in the ipsilateral T10 segment than in T8 exhibited these higher indexes. Spike-triggered averages generated from nonoverlapping segments of the records from these neurons invariably exhibited a consistent peak ~60 ms after action potentials of the DHNs.
Fields from which excitatory and inhibitory renal sympathetic nerve responses were evoked by noxious cutaneous stimulation were topographically organized
In all rats, pinch of the left flank, back, and abdominal regions increased RSNA (Figs. 5A, top and 6A, left). The magnitude of this increase in sympathetic activity was positively related to the applied pressure. Pinch of areas on the rostral and superior boundaries of the laparotomy elicited the greatest responses. Progressively smaller responses were elicited by pinch at increasing distance from the incision. Pinch of the left shoulder and forelimb and the left hip and hindlimb elicited transient reductions in RSNA (Figs. 5B, top and 6A, right). Pinch of contralateral dermatomes occasionally decreased RSNA. The largest reductions in RSNA elicited from the contralateral side followed pinch of the right abdomen and flank.
Fields from which excitatory and inhibitory responses in the activity of DHNs evoked by noxious cutaneous stimulation were topographically organized
Most DHNs were excited when their primary dermatomes were stimulated (Fig. 6, B-F, left). These neurons were often inhibited when regions adjacent to or distal to their primary dermatomes were stimulated (Fig. 6, B-F, right). The cutaneous receptive fields of six DHNs were either aberrant or nonexistent, and these DHNs were assigned to the category of "others" (Fig. 3). Of these, one T2 DHN (Fig. 6B, right) and one L2 DHN (Fig. 6F, right) were inhibited by pinch in areas corresponding to the excitatory fields of all other neurons from the same spinal segment. One T2 neuron was excited by pinch of its primary dermatome, but it was inhibited by brush of exactly the same area. Cutaneous somatic fields could not be found for one ipsilateral T10 DHN and two L2 DHNs.
Our major observations are 1) that the activity of DHNs in a limited range of thoracic segments is strongly correlated with bursts of simultaneously recorded RSNA and 2) that somatic stimuli that alter RSNA similarly alter the activity of these DHNs. Both of these observations support the hypothesis that a subset of DHNs consists of excitatory synaptic antecedents to renal sympathetic preganglionic neurons.
Correlations between ongoing activity in DHNs and ongoing RSNA
We began our search for sympathetically correlated neurons in the left T10 segment for two reasons. First, renal injections of pseudorabies virus (Schramm et al. 1993 Relationships between reflex responses in renal sympathetic activity and reflex responses of DHNs
Our observation that noxious mechanical stimulation of the caudal flank, back, and abdomen of spinally transected rats increased RSNA confirms previous data from this and other laboratories (Kimura et al. 1995
INTRODUCTION
Abstract
Introduction
Methods
Results
Discussion
References
). Simultaneous recordings from brain stem neurons and sympathetic nerves have provided the most comprehensive neurophysiological characterization of these supraspinal systems. Employing the techniques of spike-triggered averaging, spectral analysis, and linear coherence, neurons have been identified in the brain stem with ongoing activity that is highly correlated to rhythms of simultaneously recorded, ongoing, peripheral sympathetic activity (Barman and Gebber 1981; Barman et al. 1995; Gebber et al. 1995; Zhong et al. 1993, 1995). These sympathetically correlated neurons are generally considered to be putative "presympathetic neurons," 1) members of networks that generate or inhibit sympathetic activity or 2) neurons projecting brain stem-generated, sympathoexcitatory or sympathoinhibitory information to spinal levels.
). Spinal sympathetic systems, however, deserve detailed investigation, for they may generate significant levels of sympathetic nerve activity in intact animals (Taylor and Weaver 1993), and, more importantly, they are responsible for all of the sympathetic nerve activity that is generated caudal to serious spinal cord injuries (Mathias and Frankel 1983).
; McLachlan and Hirst 1980). Therefore sympathetic activity must be generated by the synaptic antecedents of preganglionic neurons. Logically, after spinal cord transection these antecedents reside only in the spinal cord. We anatomically identified candidates for spinal presympathetic neurons in experiments in which we injected pseudorabies virus into the kidneys of rats (Schramm et al. 1993). Dorsal horn neurons (DHNs) became infected via retrograde transport of the virus across their synapses either on renal sympathetic preganglionic neurons or on other synaptic antecedents of renal sympathetic preganglionic neurons (Schramm et al. 1993).
METHODS
Abstract
Introduction
Methods
Results
Discussion
References
-chloralose (100 mg/kg iv) delivered via the right femoral vein. Rats were shaved and marked with a reference grid for mapping responses to somatic stimulation (Fig. 6). The right femoral artery was cannulated for measuring arterial pressure. The trachea was intubated for artificial respiration. Rats were mounted in a stereotaxic frame and paralyzed with gallamine triethiodide (Flaxedil, 40 mg/kg iv). Rats recovered from the effects of Flaxedil every 30-60 min, permitting an assessment of the depth of anesthesia. Anesthetic (-chloralose) was supplemented (25 mg/kg) as necessary to keep rats at a surgical plane throughout the experiments. Body temperature was monitored with a rectal probe and maintained at 37°C with a heating pad and heat lamp.
View larger version (54K):
[in a new window]
FIG. 6.
Cutaneous fields for noxious stimulation (METHODS) that excited (A, left) and inhibited (A, right) RSNA or excited (B-F, left) and inhibited (B-F, right) dorsal horn neurons in the spinal segments indicated. Note that fields for sympathetically correlated and uncorrelated ipsilateral T10 dorsal horn neurons are shown separately (C and D). For clarity, only representative fields are illustrated.
90 mmHg. A second laminectomy was performed to expose the spinal cord at the level of the neuronal recording (i.e., at either T2, T8, T10, T13, or L2). The exposed spinal cord was covered by a pool of warm mineral oil. A bilateral pneumothorax was performed to reduce respiratory movements. The spinal cord was stabilized by clamping two processes: one just rostral to the level of the neuronal recording and the other at sacral levels. The left kidney was approached via a left flank laparotomy. After the kidney was retracted, the adrenal gland and fat covering the psoas and the paraspinal muscles were deflected away from the renal nerve, which was usually located at the juncture of the aorta and the renal artery. The nerve was then dissected from the surrounding tissues, placed on a bipolar hook electrode, and covered with warm mineral oil.
) connected to a high-impedance probe (Grass HIP5). The resulting signal was filtered (300-3,000 Hz half-power cutoff frequencies) and amplified 50,000 times (GRASS P5 AC amplifier). The single neurons' action potentials were discriminated by a dual-window discriminator (BAK Electronics). Neurons selected for recording were spontaneously active. We are confident that these neurons represented DHNs, rather than sympathetic preganglionic neurons or the axons of primary afferents, for the following reasons. First, we rarely recorded at depths at which sympathetic preganglionic neurons are found. Second, the intraburst interspike intervals of neurons located at even our most ventral recording sites were much shorter than those reported for sympathetic preganglionic neurons in rats (Gilbey et al. 1986) or cats (Gebber and McCall 1976). Finally, the durations of these neurons' action potentials ranged from 1 to 1.5 ms, substantially longer than the 0.3- to 0.7-ms duration of action potentials we recorded from primary afferents in Lissauer's tract, the lateral funiculus, or the dorsal columns. In most electrode tracks, the first neuron encountered was selected for recording, after which the recording site was marked with an electrolytic lesion (60-90 µm diam) by delivery of an anodal current through the recording electrode (15 µA for 10-15 s). In the remaining electrode tracks, all neurons encountered in a single track were recorded, and a lesion was made at the deepest recording site. For these tracks, more superficial recording sites were reconstructed from this lesion based on readings from the microdrive.
10% from the prestimulus control level.
1 s to avoid the overlapping of RSNA epochs. At least 300 epochs were collected and averaged for each neuron. We calculated a control or "dummy" average for the same RSNA, using as a trigger an electronically generated pseudorandom signal with a frequency approximately equal to that of the recorded neuron.
2 (Fig. 2) and spike-triggered averages generated from nonoverlapping subsections of their recordings had similar peaks at similar latencies.
View larger version (20K):
[in a new window]
FIG. 2.
Incidence of correlation indexes (METHODS) as a function of spinal segment. Note that only segments T8 and T10 contained neurons with correlation indexes >2. Ipsi, ipsilateral; cont, contralateral.
) and the mode of each ISIH were computed as measures of the degree to which neurons' action potentials occurred in bursts.
2 test. Values of P 
0.05 were considered significant.
RESULTS
Abstract
Introduction
Methods
Results
Discussion
References
View this table:
TABLE 1.
Numbers of dorsal horn neurons analyzed
View larger version (32K):
[in a new window]
FIG. 1.
Representative recordings from sympathetically correlated (A) and sympathetically uncorrelated (B) dorsal horn neurons. Top: discriminator output showing action potential occurrences. Middle: renal sympathetic nerve activity (RSNA) recorded simultaneously with dorsal horn neuron recordings. Bottom: bold traces, spike-triggered averages of RSNA triggered with action potentials of dorsal horn neurons (METHODS); fine traces, spike-triggered averages of RSNA triggered with "action potentials" of "dummy neurons." Occurrences of triggers: 0 s. Vertical scales for middle and bottom reflect amplification of 10,000 times (METHODS). Correlation indexes for these representative correlated and uncorrelated neurons: 2.5 and 0.6, respectively.
2 and that also exhibited peaks in their spike-triggered averages at repeatable latencies in nonoverlapping segments of their records were considered correlated. Using these criteria, 43% of the neurons recorded in the ipsilateral T10 segment were sympathetically correlated, compared with 16% of the neurons recorded in the T8 segment. The average latency between the action potentials of correlated DHNs and the peak of their spike-triggered averages of RSNA was 59 ± 8 (SE) ms. Because histologically identified, sympathetically correlated neurons were widely distributed across dorsal horn laminae, their numbers were too small to permit statistical comparisons of their laminar distributions (Fig. 3). However, the relative incidence of correlated neurons appeared to increase in the deeper laminae of ipsilateral T10.
View larger version (17K):
[in a new window]
FIG. 3.
Laminar distributions of histologically recovered dorsal horn neurons at 3 spinal levels. Ipsilateral neurons were located ipsilateral to the left renal nerve. See text for criteria for "correlated" neurons. LT, low threshold; WDR, wide dynamic range; HT, high threshold; "others," none of the above (METHODS). Outlines modified from Figs. 68 and 69 of Paxinos and Watson (1982) .
2 test for independence).
View larger version (22K):
[in a new window]
FIG. 4.
Interspike interval (ISI) histograms of a bursting dorsal horn neuron (top) and a nonbursting dorsal horn neuron (bottom). Bin size: 10 ms. Number of action potentials for bursting and nonbursting neurons: 600 and 900, respectively.
View larger version (19K):
[in a new window]
FIG. 5.
Representative renal sympathetic nerve responses (top) and dorsal horn neuronal responses (bottom) to noxious cutaneous stimulation of the left flank (A) and left hip (B). Responses in A and B were recorded from same neuron and sympathetic nerve. Stimuli were maintained for durations indicated by horizontal bars. Vertical scales for top reflect amplification of 10,000 times (METHODS).
DISCUSSION
Abstract
Introduction
Methods
Results
Discussion
References
) indicated that this segment contained the largest number of renal sympathetic preganglionic neurons, and the distribution of renally infected DHNs correlated strongly with that of renal preganglionic neurons. Second, microinjection of D,L-homocysteic acid into the intermediolateral column at T10 in rats produces the largest increases in RSNA (Taylor and Weaver 1992). In many previous attempts, we failed to find renal sympathetically correlated DHNs at segments rostral to T8 and caudal to T13. These failures were explained when we reinvestigated a wide, longitudinal range of segments in the current study. We found that, rostral and caudal to T10, the incidence of sympathetically correlated DHNs diminished rapidly. At T8, a segment that contains a substantial number of renal sympathetic preganglionic neurons and presumptive renal interneurons (Schramm et al. 1993), the activity of only 16% of DHNs was correlated with RSNA. At T13, a segment that contains a small number of renal sympathetic preganglionic neurons (Schramm et al. 1993) and presumptive renal interneurons, we observed no sympathetically correlated DHNs.
found that contralateral injections of D,L-homocysteic acid produced few increases in RSNA. Second, many fewer DHNs are infected contralateral to renal pseudorabies virus injections (Schramm et al. 1993). Third, Cabot et al. (1994) observed no retrograde, transynaptic transport of cholera toxin from sympathetic preganglionic neurons to neurons on the contralateral side of the spinal cord. All of these observations are consistent with the hypothesis that, in spinally transected rats, the DHNs that are excitatory antecedents to renal sympathetic preganglionic neurons are both longitudinally and laterally restricted.
; Osborn et al. 1987; Taylor and Weaver 1993). The potential sources of ongoing renal sympathetic activity after spinal transection in rats have been thoroughly reviewed by Taylor and Weaver (1993). Ongoing activity could be attributed to drive provided by primary afferents. Alternatively, the source of ongoing sympathetic activity could be spinal networks that are independent of afferent drive. After spinal transection, extensive dorsal rhizotomy decreases renal sympathetic activity by 25%, suggesting a role for tonic afferent drive in generating renal sympathetic activity (Taylor and Weaver 1993). On the other hand, the fact that 75% of RSNA (and 100% of mesenteric nerve sympathetic activity) survives these rhizotomies supports the existence of endogenous spinal sympathetic generators. Although the present experiments are the first to suggest a role for DHNs in either the generation or transmission of ongoing excitatory drive to sympathetic preganglionic neurons, we are not able to conclude which of these roles, generation or transmission, is manifested by the observed correlations. It should be noted, however, that some portion of the ongoing RSNA and some of the ongoing activity in DHNs observed in these acutely spinalized rats may have been generated by the surgery and anesthesia necessary to perform our experiments. Maiorov et al. (1997) have recently shown that, in the absence of anesthesia, acute surgery, or other somatic or visceral stimuli, chronically transected rats exhibit relatively little ongoing RSNA. The role of DHNs in generating this small, residual RSNA after chronic spinal transection is conjectural. On the basis of the data from present experiments, however, it is plausible that DHNs play a role in generating the temporal patterns of stimulus-driven RSNA in the chronic spinal mammal.
; Poree and Schramm 1992). In our earlier experiments, we showed for the first time that noxious thermal stimulation of these regions not only increased RSNA, but synchronously increased the activity of caudal thoracic DHNs (Poree and Schramm 1992). On the basis of those results, we tentatively hypothesized that thoracic DHNs may mediate homotopic somatosympathetic and viscerosympathetic reflex responses to noxious stimuli by increasing excitatory input to sympathetic preganglionic neurons.
). That separate propriospinal mechanisms might mediate inhibition of DHNs and inhibition of sympathetic preganglionic neurons is possible. It is more likely, however, that noxious stimulation of ipsilateral fore- and hindlimb dermatomes and the contralateral flank reduces the activity of many lower thoracic DHNs. Reduction of the activity of those DHNs that are excitatory synaptic antecedents to renal sympathetic preganglionic neurons would be expected to reduce RSNA.
; Weaver et al. 1983), most authors emphasize the segmental character of sympathetic reflexes after spinal cord transection (see Kimura et al. 1995 for review). The reduction of RSNA by noxious stimulation of the extremities and the inhibition of most DHNs at T2, T10, and L2 by heterosegmental, noxious stimulation are important, however, because they demonstrate the existence of functional, intersegmental regulation of the activity of DHNs and sympathetic activity after spinal transection.
found that, after C1 spinal cord transection in rats, electrical or chemical stimulation of the dorsolateral surface of the cervical spinal cord substantially reduced RSNA, very likely by exciting part of a descending brain stem inhibitory pathway or a propriospinal inhibitory pathway. Poree and Schramm (1992), recording simultaneously from thoracic DHNs and the renal sympathetic nerves, showed that electrical and chemical cervical stimulation, similar to that described by Schramm and Livingstone, synchronously reduced ongoing DHN activity and ongoing RSNA. Further, cervical stimulation reduced responses in both RSNA and DHNs to noxious stimuli applied to cutaneous excitatory fields on the left flank.
| |
ACKNOWLEDGEMENTS |
|---|
We gratefully acknowledge the assistance provided by J. Black, J. Lai, and C. Ma in the analysis of data.
This research was supported by National Heart, Lung, and Blood Institute Grant HL-16315.
| |
FOOTNOTES |
|---|
Address for reprint requests: L. P. Schramm, The Johns Hopkins University School of Medicine, 720 North Rutland Ave., 606 Traylor Bldg., Baltimore, MD 21205-2196.
Received 13 December 1996; accepted in final form 5 February 1997.
| |
REFERENCES |
|---|
|
Abstract
Introduction
Methods
Results
Discussion
References
|
|---|
This article has been cited by other articles:
|
|
 |
|
  B. Pan, M. R. Zahner, E. Kulikowicz, and L. P. Schramm Effects of corticospinal tract stimulation on renal sympathetic nerve activity in rats with intact and chronically lesioned spinal cords Am J Physiol Regulatory Integrative Comp Physiol, July 1, 2007; 293(1): R178 - R184. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 D. Chau, D. G. Johns, and L. P. Schramm Ongoing and Stimulus-Evoked Activity of Sympathetically Correlated Neurons in the Intermediate Zone and Dorsal Horn of Acutely Spinalized Rats J Neurophysiol, May 1, 2000; 83(5): 2699 - 2707. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| Visit Other APS Journals Online |
