The Effects of Spinal or Mesencephalic Transections on Sleep
|Spinal Transection||Mesencephalic Transection|
|Total PS (min)||101 ± 12||81 ± 10||95 ± 2||54 ± 5|
|Number of PS Phases||45.5 ± 10||38.1 ± 5.4||45.8 ± 2.0||27.3 ± 3.1|
|Avg. PS episode dur. (min)||1.9 ± 0.1||1.7 ± 0.1||1.6 ± 0.1||1.8 ± 0.2|
|Total SWS (min)||639 ± 25||613 ± 27||673 ± 21||449 ± 30 *|
|Total Wake (min)||683 ± 33||731 ± 37||664 ± 18||931 ± 35 *|
Note: Values are given as the means ± S.E.M. The data from the spinal and mesencephalic rats represent the mean values from two control recordings (C) and two post-transection recordings (T) obtained from the seventh to ninth day post-lesion. SWS and wakefulness following mesencephalic transection were defined as non-PS episodes with cortical desynchronization or slow wave activity, respectively. The PS episode duration is defined as the duration of PS uninterrupted by either SWS or wakefulness. Any two PS episodes separated by less than four minutes in duration comprise a single PS phase. * Indicates a significant difference with respect to the corresponding pre-lesioned control using paired t-tests; *p<0.001. C, control; T, post-transection.
* Indicates significant difference with respect to the prelesioned controls using paired t-test; *p<0.01, **p<0.001.
|Spinal Transection||Mesencephalic Transection|
|Total Erections||18.1 ± 4.0||0.4 ± 0.2||15.3 ± 1.7||1.7 ± 0.6|
|Percent Phases with Erection||36.5 ± 5.6||0.8 ± 0.5||31.2 ± 2.5||7.0 ± 3.0|
|Slow Wave Sleep|
|Total Erections||0.6 ± 0.3||4.1 ± 2.0||0.9 ± 0.3||3.8 ± 0.8|
|Total Erections||46.5 ± 3.0||36.9 ± 7.0||52.8 ± 5.3||40.2 ± 7.4|
Note: Values are given as the means ± S.E.M. The data from the spinal and mesencephalic rats represent the mean values from two control recordings (C) and two post-transection recordings (T) obtained from the seventh to ninth day post-lesion. * Indicates a significant difference with respect to the corresponding pre-lesioned control using paired t-tests; *p<0.05, **p<0.01, ***p<0.001. C, control; T, post-transection.
Although sleep-wake parameters remained unchanged post-lesion (Table 1), we observed a dramatic and significant reduction in PS erectile activity. The most striking effect was a significant reduction in the number of erections per hour of PS (Fig. 1). Similar reductions in PS erectile activity were found with respect to the total number of erections during PS, as well as the percentage of PS phases associated with an erection (Table 2).
Unlike the disruption in PS-related erections, the quantity of waking-state erections were only minimally decreased following spinal transection (Table 2 ). These erectile events during wakefulness, however, were related almost exclusively to movement of the animal such as during drinking, eating and, especially, during movement of the animal using the forelimbs, causing the perineurn to brush against the surface of the cage. When the animal was relatively motionless during quiet wakefulness or sleep, however, erectile events were rarely observed and appeared only sporadically during these behavioral states. During sleep, for example, post-hoc comparisons following a one-way ANOVA revealed that the number of erections per hour following spinal transection was not significantly different between PS and SWS.
Two of the eight rats used for this study did not survive mesencephalic transection and are not included in the following data analysis. Mesencephalic transections of the remaining six rats were located at the same rostro-caudal level resulting in transection of the rostral superior colliculus and red nucleus (Fig. 2). Moreover, we found minimal rostro-caudal extension of the transection (Fig. 2). Exceptions included partial degeneration of both rostral and caudal portions of the transected red nucleus, as well as some hemorrhaging restricted to midline structures in several rats that often involved the oculomotor and Edinger-Westphal nuclei, but leaving the midline raphe nuclei intact in all cases. In all rats the medial geniculate nucleus, lateral extremes of both the substantia nigra and cerebral peduncles, and dorsal portions of the superior colliculi were not transected (Fig. 3). It should be noted that the completeness of the transections varied among rats. For example, the Periaqueductal Gray (PAG) remained largely intact in two cases (T 11 and J 11), and the lateral extent of the transection was not symmetrical in one rat (M 11) in that a large portion of the left central tegmental field, substantia nigra and cerebral peduncle was not transected (Fig. 3). These individual differences are highlighted when necessary when physiological effects differed.
The approximate laterality (L) of the sections with respect to the midline are indicated in the figure according to the atlas of Paxinos and Watson. (28)
Lesions from sagittal sections were analyzed and reconstructed on a frontal plane, corresponding to P 5.8 on the Paxinos and Watson (28) atlas. SC, superior colliculus; PAG, periaqueductal gray; MG, medial geniculate nucleus; R, red nucleus; SNR, substantia nigra pars reticulata; IP, interpeduncular nucleus; cp, cerebral peduncle; 3, oculomotor nucleus. Scale at the bottom of the figure is in mm. The drawing of the coronal section was adapted from ref. (28)
Periodic monitoring of rectal temperature demonstrated that all rats were able to maintain a normal body temperature of approximately 37.0°C following transection. In addition, all rats maintained the micturition reflex immediately following the mesencephalic lesion. The only significant feature regarding an analysis of micturition was a 70 percent reduction in the total number of micturition events with more than a doubling in the duration of each event in these rats.
Following mesencephalic transection, PS phenomena were still present in all rats as seen by the rhythmic occurrence of muscle atonia and rapid eye movements (Fig. 4). These PS episodes were usually, though not always, associated with a cortical desynchronization. PS was generally absent for the first 24 hours following the transection, reappeared after two to three days, and the total daily quantity of PS stabilized within four to five days post-transection (Fig. 5). The total daily quantity of PS, however, was significantly reduced one week following transection (Table 1). This decrease in total PS was related to a reduction in the number of PS phases and not in the average PS episode duration (Table 1).
Cortical slow wave activity, or SWS, markedly increased for the first two to four days post-transection (Fig. 5) as previously demonstrated in the literature in cerveau isolé preparations. (8,27) Cortical desynchronization then reappeared and dominated the polygraphic recordings one week post-lesion. Indeed, following an initial recovery period, there were both a significant increase in wakefulness and a decrease in SWS one week post-transection with respect to control levels (Table 1).
A: An example of a transition from SWS to PS (see arrow) prior to mesencephalic transection and a typical erectile event during PS occurring approximately thirty seconds later in the same PS episode. PS-related erectile events involve a characteristic increase in baseline CSP pressure and, with BS muscle bursts, dramatic pressure peaks. B: PS episode following a mesencephalic transection in the same rat. Note the simultaneous occurrence of muscle atonia and rapid eye movements which define the PS episode. The arrows mark the beginning and end of the PS episode. Activity seen in the neck EMG during the PS episode is EKG artifact on an otherwise atonic baseline. The CSP pressure during PS post-transection typically remained stable at the flaccid baseline level with an absence of BS muscle bursts.
Although PS remained qualitatively intact, PS-related erections were disrupted following mesencephalic transections (Fig. 4). PS erectile activity was significantly reduced from an average of 9.8 erections per hour pre-lesion to less than two erections per hour of PS (Fig. 1), even after survival periods as long as 24 days post-transection. As shown in Table 2, this reduction in PS-related erectile activity post-transection also involved a highly significant decrease in the total number of PS-related erections and the percentage of PS phases exhibiting an erectile event.
The quantity of PS-related erections remaining following mesencephalic transection appeared to be related to the completeness of the transection. For example, in rat M I I the lateral extent of the transection was not bilaterally symmetrical (Fig. 3), leaving a large portion of the left rostral mesencephalon intact. This rat exhibited the most PS-related erectile activity following transection with an average of 5.0 erections per hour of PS more than one week post-lesion. In contrast, no PS-related erectile events were observed during the entire 12-day, post-transection, recording period of rat Al I in which the mesencephalic transection was relatively complete (Fig. 3).
We also found a significant decrease in the number of erections per hour of wakefulness (Fig. 1). This per hour reduction was related to the significant increase in total wakefulness following the transection since the total number of erections during this state did not significantly decrease (Table 2). On the other hand, SWS was associated with a small but significant increase in the number of erections per hour after transection.
Control, or pre-transection, values from six mesencephalic rats and three spinal rats were pooled together for an analysis of penile reflexes. The ability to induce a reflexive erection was dependent on the level of the transection. For example, whereas 100% of all reflex tests were successful in eliciting erectile events following spinal transection, only 50% of all reflex tests following mesencephalic transection were successful in eliciting penile reflexes, identical to the control condition. Moreover, the latency to reflex induction was significantly decreased following spinal transection, whereas it was significantly increased post-mesencephalic lesion (Table 3). The number of erectile events, or clusters, tended to decrease following mesencephalic transection, but this reduction did not reach statistical significance at the p<0.05 level. Finally, neither spinal nor mesencephalic transection had a significant effect on the total number of glans erections (E1, E2, or E3), flips (F1, F2), or the inter cluster interval (Table 3).
|Latency (sec)||Total Glans Erections||Total Flips||Number of Clusters||lCl (sec)|
|Control||293 ± 58||24.0 ± 2.7||10.2 ± 2.3||11.1 ± 1.4||108 ± 16.7|
|Spinal||80 31||20.0±3.1||11.0±2.5||8.2±1.3||151 ±22.7|
|Mesencephalic||725 135||13.0 ± 6.1||8.5 ± 3.8||5.3 ± 2.2||127 ± 8.3|
Note: Values represent the means ± S.E.M. Statistical analyses were performed using a one-way ANOVA El E2 and E3 events were combined as total glans erections, whereas F1 and F2 vents were combined as total flips since no sig- nificant differences in the number of these events were observed between groups. * indicates significant difference with respect to controls using a post-hoc Tukey test; *p<0.05, **p<0.01. ICI: intercluster interval.
Following mesencephalic transection, wakefulness, and SWS were defined as non-PS episodes associated with cortical desynchronization or slow wave activity, respectively.
This is the first report examining the neural mechanisms of penile erections across sleep-wake states. Using a new animal model for sleep-related erection research, we transected either the spinal cord or rostral mesencephalon in the rat to definitively elucidate the effects of paraplegia on sleep-related erections and to localize neural structures responsible for PS erectile generation. In addition, reflexive erections were performed in both spinal and mesencephalic rats to evaluate the descending control of penile reflexes in the two groups.
Spinal Erectile Mechanisms
Published data regarding the effects of paraplegia on human sleep-related erections are minimal, inconclusive, and contradictory. Although sleep-related erections are commonly thought to be disrupted by spinal cord injury as originally reported in two early abstracts, (17,18) Halstead and colleagues (7) later report in a full publication that human PS-related erections may persist following "complete" upper motor neuron spinal lesions. Any significant PS erectile activity remaining following complete upper spinal cord transection would suggest that the spinal generator controlling erections is activated or modulated by either a humoral component during PS or by simple peripheral reflex activation associated with phasic PS-related muscle twitching above the level of the transection. These latter data only confuse our understanding of sleep-related erection neurophysiology and require clarification or further exploration.
Our results demonstrate that complete midthoracic spinal transections virtually eliminated PS-related erections. Moreover, any remaining erectile activity observed during sleep was found to occur at random during SWS and PS, suggesting intermittent reflex activation of the spinal erection generator during sleep. Other sleep-wake parameters, in contrast, remained unchanged post-lesion. These data clearly demonstrate that an intact spinal cord above the midthoracic level is essential for the integrity of PS erectile activity. It remains to be determined, however, if PS erectile activity would be disrupted following spinal transections below the thoracolumbar level, leaving sympathetic innervation from the brain to the penis intact.
Although PS-related erections were adversely affected after spinal transection, spinal erectile mechanisms remained intact in all rats. Indeed, ex-copula reflexes following spinal transection demonstrated that erectile events were more easily elicited with shorter latencies relative to controls. These results confirm previous data demonstrating that reflexive erections are facilitated following spinal transection (31) or spinal block, (29) suggesting that spinal transection removes a tonic descending erectile inhibition.
In contrast to the decrease in PS-related erections, the total number of waking-state erections, as well as the number of erections per hour of wakefulness, did not significantly decrease following spinal transection. The variability in the waking-state erection data, however, markedly increased post-lesion. Moreover, these waking-state erections following spinal transection were associated almost exclusively with movement of the animal, especially when the perineurn would brush against the surface of the cage. We hypothesize that these waking-state erections were reflexively induced post-lesion. When the animal was relatively motionless during quiet wakefulness or sleep, erectile events were rarely observed after spinal transection.
Brainstem transections have served as an effective technique to localize neural structures responsible for PS generation. (14,43,44) Such data demonstrate that the pons and rostral medulla contain all of the neural elements necessary to generate the classic tonic and phasic phenomena characteristic of PS, (14) including cortical activation, rapid eye movements, muscle atonia, and PGO spike activity. A fundamental question is whether the brainstern also is sufficient to generate PS erectile activity.
Penile erections during PS were found to be dramatically and significantly decreased following mesencephalic transection even though PS was found to persist post-lesion. Moreover, the amount of PS erectile activity remaining post-lesion was related to the completeness of the transection, in that the most complete transections eliminated PS-related erections. These findings suggest that although brainstern PS mechanisms persist post-lesion, as seen by the rhythmic occurrence of rapid eye movements and muscle atonia, neural structures rostral to the transection (i.e., in the forebrain) are necessary for the production of PS-related erections.
Unlike its clear documentation in the cat, (14) PS has never been definitively demonstrated in the rat post-mesencephalic transection. (5,8,27) We observed a rhythmic occurrence of PS in all mesencephalic transected rats as seen by the simultaneous appearance of rapid eye movements and muscle atonia. PS was generally absent for the first 2-3 days post-transection, but stabilized in total daily quantiby day 4-5. Our greater success in observing PS in cerveau isolé rats may be secondary to the significantly longer survival periods in this study (average of 12.8 days vs 2-4 days in other reports). In addition, our transections were located at the mesencephalic-diencephalic junction which is more rostral than transections attempted in rats in previous reports. This more rostral location of the transections may potentially have resulted in less damage to midline pontine structures important for PS generation. Finally, although the transections were not complete as demonstrated histologically (Fig. 3) and polygraphically by the frequent onset of cortical desynchronization during PS, the transections were extensive enough to severely disrupt forebrain PS erectile mechanisms. It should also be noted that PS persisted even in rats with the most extensive transections when PS often was associated with cortical slowave activity.
Although PS remained qualitatively intact, we observed a significant reduction in the total daily quantity of PS one week post-transection. Since our previous results suggest that the probability of observing an erectile event during PS diminishes with decreasing PS episode duration, (38) it was essential to determine if the disruption in PS erectile activity was secondary to a simple reduction in the duration of PS episodes. Further analyses revealed, however, that the reduced total daily quantity of PS was characterized by a decrease in the total number of PS episodes without affecting the average, uninterrupted, PS episode duration.
Reflex erection tests following mesencephalic transection revealed that the latencies to induction were increased, and the percentage of tests eliciting an erectile event were similar to controls. These findings suggest that a descending inhibition not only remains intact but perhaps may be enhanced in a decerebrate preparation. Indeed, recent investigations have identified a major source of erectile inhibition within the brainstem in the rostral, juxtafacial, paragigantocellular nucleus (nPGi). (21,22) In the anesthetized rat, transection of the brainstern caudal to the nPGi removes a descending inhibition of penile reflexes comparable to spinal transection, whereas the inhibition remains intact following transections rostra] to this nucleus. (22) Moreover, bilateral electrolytic or cytotoxic lesions of the mPGi appear to facilitate reflexive erectile events. (21,22) Our reflex erection data demonstrating an intact descending inhibition following transection of the rostral mesencephalon supports the finding of a major inhibitory source penile erections within the brainstem.
Although our results suggest that the brainstem may be sufficient for the normal generation of PS-related erections, the data do not exclude a potential role of the brainstem in the direct descending control of PS erectile activity. For example, the nPGi appears to involve serotonin as a neurotransmitter in tonic descending erectile inhibition. (23) Given that brainstem serotonergic neurons cease firing during PS (PS-off neurons), one may speculate that a decrease in serotonergic inhibition from the brainstem could play a role in the production of PS erectile activity.
Following early transection experiments which localized the brainstem as the generator of PS, it has been assumed that the brainstem is the direct source of all PS-related phenomena. Recent experiments in the cat, however, demonstrate that mesencephalic transection eliminates phasic alterations of systemic arterial blood pressure which typically occur during PS in the intact animal (16). Together with our data, we hypothesize that certain PS phenomena, such as penile erections and PS-related cardiovascular changes, require forebrain structures. It remains to be determined, however, whether the forebrain mechanisms of both penile erections and cardiovascular control are related.
These experiments suggest that neural elements rostral to the mesencephalon are essential for the maintenance and integrity of PS-related erections, providing major directions for future research regarding the localization of neural structures responsible for PS erectile control. Unfortunately, it is presently difficult to speculate where this forebrain control may be derived. Numerous forebrain structures have been implicated in the control of either penile erections or sexual behavior and, therefore, are potential candidates in erectile mechanisms. These structures include the medial preoptic area (1,24,39) bed nucleus of the stria terminalis, (42) hippocampus, (2) amygdala, (20) and paraventricular nucleus. (25) The involvement of these structures in sleep-related erection physiology is unknown. We hypothesize that the preoptic, basal forebrain area may be a major candidate in sleep-related erectile control given its role in both sleep generation (12) and reproductive mechanisms. (1,6)
It remains to be determined how forebrain structures may influence brainstem or spinal erectile mechanisms during PS. We hypothesize that the facilitatory role of the forebrain in PS erectile control may include, at least in part, an inhibition of tonic brainstem inhibitory mechanisms. Indeed, our reflex erection data suggest that a descending erectile inhibition may be enhanced following mesencephalic transection. Such a disinhibitory role of the forebrain in erectile control has been suggested by others. (24)
Additional research is required to localize forebrain structures important for PS erectile generation and to elucidate the descending control of sleep-related erectile mechanisms. All available surgical or medical options for the treatment of erectile failure, including Sildenafil (Viagra), target the end organ level. Further understanding of higher central erectile mechanisms, Particularly forebrain PS erectile control, may revitalize and redirect sleep-related erection testing as a clinical tool for the evaluation of impotence and provide new treatment alternatives for erectile dysfunction.
We would like to thank Dr. Helmut S. Schmidt, Dr. Jian-Sheng Lin, and Dr. Jeffrey Brown for their advice and critical review of the manuscript. Research supported in part by INSERM U52, CNRS URA 1195, Claude Bernard University and the Medical College of Ohio at Toledo.