13
Brain Research, 547 (1991) 13-21 © 1991 Elsevier Science Publishers B.V. 0006-8993/91/$03.50 ADONIS 000689939116517Q BRES 16517
Projections of the parvocellular reticular formation to the contralateral mesencephalic trigeminal nucleus in the rat R.E Minkels, P.J.W. Jfich, G.J. Ter Horst and J.D. Van Willigen Department of Neurobiology and Oral Physiology, University of Groningen, Groningen (The Netherlands) (Accepted 6 November 1990)
Key words: Parvocellular reticular nucleus; Mesencephalic trigeminal nucleus; Jaw reflex; Periodontal afferent; Muscle spindle afferent; Rat
Projections of the parvocellular reticular nucleus (PCRt) to the contralateral mesencephalic trigeminal nucleus (Me5) were studied in the rat with neurophysiological and neuroanatomical techniques. Three types of responses (classified by latencies) were recorded extracellularly in the Me5 area after electrical stimulation of the PCRt: (1) R1 fast unitary reactions (latency 0.2-0.5 ms) found throughout the entire caudal Me5 area; (2) R2 slower unitary reactions (latency 0.7-1.2 ms) recorded ventral and lateral to the caudal Me5; and (3) R3 compound potentials (latency 1.0-2.5 ms) recorded within the ventrocaudal Me5. Relayed or synaptic fields were not observed. IntraceUular recordings of identified cell bodies of periodontal afferents, muscle spindle afferents and interneurones revealed no short-term postsynaptic potentials following PCRt stimulation. In some experiments jaw muscle spindle afferent activity was increased by PCRt stimulation and jaw-opening and jaw-closing reflexes were facilitated in the contralateral musculature. Neuroanatomical experiments, using Phaseolus vulgaris leucoagglutinin as an anterograde tracer, showed direct projections with intensive collateralization of PCRt fibres into the Me5 area. Synaptic contacts of PCRt fibres with primary afferent cell bodies were observed in the ventrocaudal parts of the Me5 only. The electrophysiological results are discussed in relation to the neuroanatomical findings.
INTRODUCTION The oral system is involved in several motor functions such as sucking, swallowing, mastication, drinking and deglutition. Movements related to these performances are executed by a system of muscles that is attached to the skull, the mandible, the hyoid bone, the vertebral column, the thoracic cage and the shoulder girdle. The system comprises 6 muscle groups, namely the muscles of mastication, the muscles of the tongue, the suprahyoidal and the infrahyoidal muscles, the muscles of the pharynx and the muscles of the neck. The jaw musculature has a unique bilateral organization since it moves the mandible with two joints in the transverse plane. To achieve elevation or depression of the mandible, groups of muscles on both sides of the head must be activated. This implies a bilateral organization of neural control of the jaw movements. Important structures contributing to the coordination of bilateral muscle activity during oral motor behaviour are the supratrigeminal zone 25 and the medullary reticular formation 17,~8. The mesencephalic trigeminal nucleus (Me5) which is strictly ipsilaterally organized,
contains the primary afferent neurones of the masticatory muscle spindles 12 and low threshold periodontal mechanoreceptors 8. Collaterals of Me5 dental afferents synapse on interneurones of the supratrigeminal nucleus which in turn make contact with neurones of trigeminal nuclei on both sides of the brainstem 25. Moreover collaterals of Me5 periodontal receptors and jaw-closing muscle spindles also reach the parvocellular reticular formation (PCRt) via the bundle of Probst 1'9'31. Neuroanatomical studies have indicated that this PCRt has efferent connections with the trigeminal nucleus oralis, the supratrigeminal area, the Me5, the trigeminal motor nucleus and the facial motor nucleus 25'26~32. The PCRt projections appeared to be bilaterally organized with an ipsilateral dominance. Therefore the PCRt is considered to be a crucial link in the neural circuitry for bilateral trigeminal motor control. Rokx et al. 26 showed that projections from the PCRt to the contralateral Me5 are restricted to the caudal third of this nucleus. As the periodontal afferents are mainly located in this caudal part of the Me53 it is suggested that these afferents in particular are influenced by P C R t input. Jtich and Rokx 2° demonstrated electrophysioiogically ipsilateral projec-
Correspondence: R.F. Minkels, Department of Neurobiology and Oral Physiology, University of Groningen, Bloemsingel 10, 9712 KZ Groningen, The Netherlands.
14 tions of the P C R t to the Me5; by stimulating the P C R t electrically they activated spindle afferent units, facilitated ipsilateral masseter reflexes and r e c o r d e d relayed and direct afferent volleys of P C R t fibres in the Me5. The existence of two populations of P C R t fibres projecting on different regions of the Me5 was indicated. In this p a p e r we will describe experiments in which we explored the contralateral Me5 with microelectrodes and r e c o r d e d contralateral jaw muscle activity during P C R t stimulation. We l o o k e d to see whether: (1) the contralateral Me5 receive projections of different populations of P C R t fibres; (2) the PCRt fibres affect predominantly periodontal afferents in the contralateral Me5 as suggested in the literature; and (3) P C R t stimulation influences reflexes in the contralateral jaw muscles. The results will be discussed in connection to contralateral projections of the PCRt as revealed by tract-tracing techniques using Phaseolus vulgaris leucoagglutinin as an anterograde tracer.
MATERIALS AND METHODS
Electrophysiological experiments The experiments were performed on 27 halothane-anaesthetized rats (male, 250-300 g). A tracheal cannula was inserted and the animal was transferred to a stereotaxic frame. A midline scalp incision was made and a craniotomy was carried out for the introduction of stimulating and recording electrodes. Rectal temperature was maintained at 37 °C. Bipolar glass-coated stainlesssteel electrodes (tip separation 0.7-1.0 mm; impedance 200 kD) were used for electrical stimulation of the PCRt. The stimulus electrodes were connected to the output of a constant current stimulus isolation unit. The stimulus site comprised the area defined by the following coordinates24: AP -1.0 to -2.5 mm, DV 1.0 to 2.0 mm and Lat. 2.0 to 2.4 mm. Extracellular recordings in the Me5 region were made with glass-coated stainless-steel microelectrodes (impedance 1-2 MD). Intracellular recordings from cells in the Me5 area were made with glass microelectrodes filled with 150 mM KCI2'11'z9. Impedance of the electrodes at 30 Hz ranged from 25 to 35 MD. The signals were amplified (1000 x) and filtered (extracellular 5-50 kHz; intracellular 0-10 kHz). All recordings were restricted contralateral to the stimulus site. Muscle spindle afferents in the Me5 were located either by stretching and probing the jaw closer muscles or by stimulating the masseter nerve electrically. Bipolar strap electrodes were used for stimulation of the nerve 19. Periodontal afferents in the Me5 were located by electrical stimulation of the inferior alveolar nerve (IAN). The IAN was stimulated with bipolar enamelled silverwire electrodes (o.d. 110 /~m) with bare ends of 1 mm. One electrode was inserted up to 10 mm into the mandibular canal via the foramen mentale, the other up to 2 ram. In 5 animals the anterior digastric and masseter muscle were fitted with bipolar wire electrodes. The EMGs from these muscles were amplified (1000 x) and filtered (5 Hz-5 kHz). Jaw-opening reflexes were evoked by IAN stimulation; jaw-closing reflexes by applying jerks to the lower jaw21. All data were analyzed on-line (Macintosh II; sampling rate 100 kI-lz) and the results were stored on hard disk for later printing. Extracellular recording and stimulation sites were verified using the Prussian blue marking method. In experiments where pipettes were used, the same marking method was applied, here the glass electrode was replaced by a stainless-steel microelectrode.
Neuroanatomical experiments Iontophoretic injections of Phaseolus vulgaris leucoagglutinin (PHA-L) were made in the PCRt of 6 animals. Glass micropipettes (tip diameter 10-15 ~m) were filled with a solution of 2.5% PHA-L in 0.05 M Tris-buffered saline (TBS, pH 7.4). The tracer was injected by passing an alternating positive current (5/~A; 7 s on-7 s off) through the pipette during 20 rain. After a survival time of 7 days the rats were perfused intraeardially with a solution made up of 2.5% glutaraldehyde, 0.5% paraformaldehyde and 4% sucrose in a 0.05 M phosphate buffer (pH 7.4). The brains were dehydrated for 12 h in a 30% sucrose solution and cut on a cryostat in 40 ~m sections. The sections were incubated with rabbit anti-PHA-L (Dakopatts, 1:2000, 48 h), goat anti-rabbit IgG (Sigma, 1:150, 14 h) and rabbit peroxidase-anti-peroxidase (Dakopatts, 1:800, 4h). Diaminobenzidine (DAB, 0.04%) was used to visualize the peroxidase. The sections were mounted onto gelatin-coated slides, air dried, counterstained with Cresyl violet, dehydrated, coverslipped and studied by light microscope. RESULTS
Electrophysiological experiments Field potentials in the Me5 area evoked by PCRt stimulation. E a c h e x p e r i m e n t c o m m e n c e d with positioning of the stimulus electrodes. This was achieved by recording multi-unit spindle afferent responses from Me5 collaterals in the P C R t with the cathode of the stimulus electrode. Best spindle responses were r e c o r d e d in the P C R t at a d e p t h of 8.5 m m and at a laterality of 2.2 m m at A P - 1 . 5 to - 2 . 0 ram. A f t e r having d e t e r m i n e d the stimulus site, the recording e l e c t r o d e was placed stereotaxically in the Me5 area. Subsequently the Me5 area was e x p l o r e d for responses to P C R t stimulation. Extracellular recordings revealed different types o f n e g a t i v e - p o s i t i v e field potentials in response to the stimulus. T h e s e responses were classified in 3 categories on the basis of their latencies. The earliest response d e t e c t e d after the stimulus artifact a p p e a r e d as a sharp n e g a t i v e - p o s i t i v e deflection (R1, Fig. 1B). The latency of the negative c o m p o n e n t of this response ranged between 0.2 and 0.5 ms, giving a conduction velocity (CV) of 12-30 m/s. T h e distance b e t w e e n the P C R t and the contralateral Me5 (necessary for calculation of the CV) was m e a s u r e d from o u r n e u r o a n a t o m i c a l data as described later in this section. R1 responses showed no change to high stimulus frequencies and were able to follow up to 600 Hz. R1 responses were only r e c o r d e d in the caudal part of Me5 and, outside the Me5, in the supratrigeminal zone and in the area lateral of Me5 (Fig. 3). A second group of responses (R2, Fig. 1A, B) r e c o r d e d in the contralateral Me5 a r e a a p p e a r e d as sharp well-defined n e g a t i v e - p o s i t i v e waves at latencies of 0.7-1.2 ms giving a C V of 5.0-8.5 m/s. A frequency following test showed no variability of timing at stimulus frequencies up to 600 Hz. T h e R2 deflexions were p r o m i n e n t in the supratrigeminal a r e a and in the area
15
~/~~,~V~J~
30I~A
~
43 ~LA
~ ~
2
0
0
i
i
i
!
i
0
1
2
3
4
Fig. 1. Extracellular recordings of unitary R1 and R2 responses in the contralateral mesencephalic trigeminal area due to stimulation of the parvocellular reticular nucleus (at 2 I-Iz, 100/~s; recording site indicated in Fig. 3). A: recording of an R2 response at threshold stimulation (26/~A). Small amplitude compound potentials of the R1 (left arrow) and R2 (right arrow in the subthreshold sweep) are already developed at this stimulus strength. B: recording of an R1 and two successive R2 responses at threshold stimulus for the second R2 response (38 gA; same recording site as in A). A threshold value of 32/~A was found for the R1 response.
I
/
F
\
/
/ /
1O0 p.V
i
i
i
i
i
0
1
2
3
4
ms
Fig. 2. Field potentials of the R3 response recorded within the contralateral mesencephalic trigeminal nucleus (recording site indicated in Fig. 3) following stimulation of the parvocellular reticular nucleus with increasing stimulus strel~,th (at 2 Hz, 100 ks). The stimulus intensity is indicated at the right hand side for each sweep. Maximum amplitude for R3 is reached at a stimulus strength of 60 pA as can be seen from the bottom trace. At a stimulus intensity of 60/~A a small amplitude R2 response was also recorded together with a compound R1 wave which developed with increasing stimulus strength.
I IJ
\
f
j
i
\
A
60 p,A
I
~x ( / M o 5 ( \~ / I
~
l~V
ms
51 I~A
I
PnC
I ~
~.
/
PY
II
I
~-~=R2
~=RI+R3
[~7~] = R I + R 2
[--'-I = R 1
~=RI+R2+R3
Fig. 3. A: diagram of a transverse section of the brainstem showing the contralateral mesencephalic trigeminal area (inset) from which responses were recorded following stimulation of the parvocellular reticular nucleus. B: detailed diagram of the Me5 area showing an overall view of the different sites from which R1, R2, and R3 responses were recorded. The filled circle and asterisk show the recording sites of the responses shown in Figs. 1 and 2, respectively. Cb, cerebellum; LC, locus coeruleus; mcp, middle cerebellar peduncle; Me5, mesencephalic trigeminal nucleus; Mo5, trigeminal motor nucleus; py, pyramidal tract; PnC, pontine reticular nucleus caudalis; PrS, principal sensory trigeminal nucleus; sS, sensory root of trigeminal nerve; SCP, superior cerebellar peduncle.
16 lateral and ventromedial of Me5. Moving the recording electrode dorsally out of the supratrigeminal area into the
400'
A
DI: 115I
S I: 88
E
200
0
400 to
E
200
400"
C
t/)
I DI: 139
CL
S I: 76
E
200"
1 s
0
\
/
Me5 itself resulted in a decrease of the amplitude of the R2 responses (in case of compound potentials). Unitary R2 reactions within the Me5 had small amplitudes and were rarely observed (Fig. 2, bottom trace). The site from which R2 responses were recorded is indicated in Fig. 3. Fig. 2 shows an example of the third category of responses (R3) recorded within the contralateral Me5 area. With stimuli of two times threshold this group appeared as negative-positive waves of relative long duration (about 1.5 ms). The shortest measured latency of the negative component was 1.0 ms. The amplitude of the R3 responses depended on the stimulus strength, indicating that the reactions can be regarded as compound responses. Distinct unitary spikes could not be recorded with the type of metal electrodes used. At stimulus frequencies above 100 Hz the R3 response became smaller in time and disappeared at frequencies above 500 Hz. Large amplitude R3 responses as shown in Fig. 2 were only recorded within the ventrocaudal part of Me5 (Fig. 3). Often negative-positive small amplitude wavelets appearing at latencies of 1.2-2.5 ms were recorded outside the Me5 in the supratrigeminal zone and the area laterally of Me5. Because these wavelets responded as R3 waves to changes in stimulus strength and frequency and appeared within the same time domain as the large R3 waves did, we considered them as small R3 responses. We never observed R1, R2 or R3 responses in the rostral part of the contralateral Me5 nucleus. Effects o f P C R t stimulation on Me5 neuron activity. The action of high-frequency PCRt stimulation (100 Hz) was studied on jaw muscle spindle afferents activated by ramp-and-hold stretches. We have subjected 24 units to this test (no attempts were made to discriminate between primary and secondary afferents). Eight units responded with an increase in activity; 16 units did not respond. Inhibition of spindle afferent activity was never observed. Fig. 4 shows an example of a masseter unit that responded to PCRt stimulation with an increase in the dynamic and static index during ramp-and-hold stretches 6. All recorded spindle units could not be fired by PCRt stimulation alone; neither by single PCRt stimulus pulses nor by high frequency PCRt stimulation.
Fig. 4. Ramp-and-hold histograms of a masseter spindle afferent in the contralateral mesencephalic trigeminal nucleus before (A), during (B) and immediately after (C) stimulation of the parvoccllular reticular nucleus (at 100 Hz, 50~s and 30#A). Each histogram (bin width 30 ms) is the average of 10 ramp-and-hold stretches. This unit responded to PCRt stimulation with an increase in the dynamic (DI) and static index (SI). The movement of the lower jaw (max. amplitude 4 mm) is shown at the bottom of the figure.
17
j• l
A[20 mV =
I 2 mV
__ --_-- . . . . . . . . . . . . . . . . .
. . . . . . .
i
|
|
|
|
!
0
5
10
15
20
25
ms
B
20 mV
0
5
10
15
20
25
ms
I 2 mV 0
5
10
|
|
!
15
20
25
ms
Fig. 5. Intracellular recordings of identified cells in the mesencephalic trigeminal nucleus. A: masscter spindle unit identified by muscle stretching and muscle probing. B: periodontal afferent identified by stimulation of the inferior alveolar nerve (20/~A, 100 /~s). C: interneuron located in the ventral border of Me5 that responded to IAN stimulation (20 /~A, 100 /~s). Upper traces: identificationof the cell, lower traces: response to stimulation of the parvocellular reticular formation with single pulses (50/~A, 50 ks).
Spindle unit activity evoked by single pulses applied to the masseter nerve was also not affected by high frequency PCRt stimulation. As spindle units could not be driven by PCRt stimulation alone we interpret the observed increase in spindle activity during the rampand-hold stretches as a result of fusimotor action induced by PCRt stimulation. In addition to these experiments we have recorded from spindle afferent units intracellulady (22 units). The spindle afferents were localized in the Me5 by stimulation of the masseter nerve and by muscle probing, and the effect of PCRt stimulation on the membrane potential was studied. We never observed any IPSPs or EPSPs in response to either single or short train (up to 5 pulses) PCRt stimulus pulses (Fig. 5A). Identical effects (Fig. 5B) were observed in membrane potentials of periodontal afferents (16 units recorded) and of interneurones (responding to IAN stimulation, 5 units recorded) in the Me5 and the supratrigeminal zone (Fig. 5C). Both periodontal and interneurone membrane potentials were not affected by the PCRt stimulus. Additionally it was found that periodontal activity evoked by IAN stimulation was also not affected by high frequency (100 Hz) PCRt stimulation. Effect of PCRt stimulation on jaw reflexes. During stimulation of the PCRt (at 100 Hz) the jaw-closing as well as the jaw-opening reflex in the relating contralateral muscles were facilitated progressively with increasing stimulus strength. Stimulation of IAN at subthreshold levels produced a small jaw-opening reflex when applied simultaneously with the PCRt stimulus (Fig. 6A). Stimulation of IAN at suprathreshold level resulted in a further increase in opening reflex activity (Fig. 6B). Similar results were obtained for jaw-closing reflexes (Fig. 6C). Neuroanatomical experiments. The anterograde tracing studies revealed that the PCRt projected predominantly onto the ventrocaudal part of the contralateral Me5 nucleus. The fibres left the PCRt mainly in a ventromedial direction, crossed the midline just dorsal to the pyramidal tract and ran dorsolaterally to the Me5 area. The course of these fibres is illustrated in Fig. 7A. Labelled fibres were also traced to more rostral parts of the brain. These fibres travelled through the Me5 area and innervated cerebellar, mesencephalic, thalamic and forebrain structures. A few synaptic structures (varicosities) were observed in the Gi and PnC. Intensive labeling was found in the PCRt, the Mo5, the supratrigeminal area and the medial part of the facial motor nucleus. A detailed outline of the intensive bouton labeling in the contralateral caudal Me5 area is illustrated in Fig. 7B.
18 DISCUSSION
A control
50 I.tA 100 IsV i
i
5
10
!
15 ms
mm
B control
L
50 btA 500 IsV
I
I
!
!
0
5
10
15
ms
C ":--'-"- control
~
40 ~A
I |
i
0
5
10
|
i
i
15
20
25
100 pV ms
The extracellular recordings showed that the R2 responses were distinct and had large amplitudes in the area ventrally, ventrolaterally and ventromedially of the caudal Me5. Small R2 responses however were recorded within the ventrocaudal part of the nucleus. R3 waves on the other hand showed opposite features; large amplitudes within the ventrocaudal Me5 and small amplitudes outside. The finding that the R3 waves were only recorded as compound potentials of relative long duration and that they could not easily follow high frequency stimulation indicate the existence of a strong collateralization of small fibres within the Me5. We regard the R3 potentials as being recorded from collaterals of axons from which the R2 waves originate. These axons then approach the Me5 at its caudal-ventral edge and ramify on entering the nucleus. The findings that latencies of the R3 waves were higher than the latencies for the R2 potentials and even slightly overlapped are in line with this view. As R1 responses were detected inside as well as outside the Me5, we consider them as being recorded from PCRt axons travelling through the Me5 area to more rostral brain areas. Our physiological observations are supported by the neuroanatomical results. Here we found labeled PCRt fibres approaching the contralateral Me5 at its ventral border and terminating in its ventrocaudal part. It is conceivable that we recorded the R2 and R3 potentials from these fibres. The R1 responses then might be recorded from the fibres running through the Me5 area to the cerebellum, thalamus and forebrain. Therefore we tentatively conclude that the contralateral Me5 receives projections from only one population of PCRt fibres. The morphology and central course of mesencephalic trigeminal neurones innervating the jaw muscle spindles and periodontal ligament is well documented 1'9"27'28. These authors indicate that sensory information from the periphery will arrive at the level of the trigeminal motor nucleus first, then reach the trigeminal nucleus oralis followed by the supratrigeminal nucleus and finally arrives at the Me5. By following this central course via collaterals of Me5 cells, action potentials evoked in the
Fig. 6. Effect of high frequency stimulation of the parvocellular reticular nucleus on jaw reflexes in the contralateral musculature. A: activity in the anterior digastric muscle in response to a subthreshold stimulus applied to the inferior alveolar nerve (10 gA, 100 ps). B: reflex EMG activity in the anterior digastric muscle evoked by IAN stimulation (100 gA, 100 #s). C: jaw-closing reflex activity in the masseter muscle evoked by a twitch applied to the lower jaw. Upper traces: control; lower traces: the same reflexes but now during high frequency stimulation of the PCRt (100 Hz, 50 #s; stimulus strength for each record indicated at the right).
19
SCO
(;
,.v-?
::i. ,-~
f
Brain Research, 547 (1991) 13-21 © 1991 Elsevier Science Publishers B.V. 0006-8993/91/$03.50 ADONIS 000689939116517Q BRES 16517
Projections of the parvocellular reticular formation to the contralateral mesencephalic trigeminal nucleus in the rat R.E Minkels, P.J.W. Jfich, G.J. Ter Horst and J.D. Van Willigen Department of Neurobiology and Oral Physiology, University of Groningen, Groningen (The Netherlands) (Accepted 6 November 1990)
Key words: Parvocellular reticular nucleus; Mesencephalic trigeminal nucleus; Jaw reflex; Periodontal afferent; Muscle spindle afferent; Rat
Projections of the parvocellular reticular nucleus (PCRt) to the contralateral mesencephalic trigeminal nucleus (Me5) were studied in the rat with neurophysiological and neuroanatomical techniques. Three types of responses (classified by latencies) were recorded extracellularly in the Me5 area after electrical stimulation of the PCRt: (1) R1 fast unitary reactions (latency 0.2-0.5 ms) found throughout the entire caudal Me5 area; (2) R2 slower unitary reactions (latency 0.7-1.2 ms) recorded ventral and lateral to the caudal Me5; and (3) R3 compound potentials (latency 1.0-2.5 ms) recorded within the ventrocaudal Me5. Relayed or synaptic fields were not observed. IntraceUular recordings of identified cell bodies of periodontal afferents, muscle spindle afferents and interneurones revealed no short-term postsynaptic potentials following PCRt stimulation. In some experiments jaw muscle spindle afferent activity was increased by PCRt stimulation and jaw-opening and jaw-closing reflexes were facilitated in the contralateral musculature. Neuroanatomical experiments, using Phaseolus vulgaris leucoagglutinin as an anterograde tracer, showed direct projections with intensive collateralization of PCRt fibres into the Me5 area. Synaptic contacts of PCRt fibres with primary afferent cell bodies were observed in the ventrocaudal parts of the Me5 only. The electrophysiological results are discussed in relation to the neuroanatomical findings.
INTRODUCTION The oral system is involved in several motor functions such as sucking, swallowing, mastication, drinking and deglutition. Movements related to these performances are executed by a system of muscles that is attached to the skull, the mandible, the hyoid bone, the vertebral column, the thoracic cage and the shoulder girdle. The system comprises 6 muscle groups, namely the muscles of mastication, the muscles of the tongue, the suprahyoidal and the infrahyoidal muscles, the muscles of the pharynx and the muscles of the neck. The jaw musculature has a unique bilateral organization since it moves the mandible with two joints in the transverse plane. To achieve elevation or depression of the mandible, groups of muscles on both sides of the head must be activated. This implies a bilateral organization of neural control of the jaw movements. Important structures contributing to the coordination of bilateral muscle activity during oral motor behaviour are the supratrigeminal zone 25 and the medullary reticular formation 17,~8. The mesencephalic trigeminal nucleus (Me5) which is strictly ipsilaterally organized,
contains the primary afferent neurones of the masticatory muscle spindles 12 and low threshold periodontal mechanoreceptors 8. Collaterals of Me5 dental afferents synapse on interneurones of the supratrigeminal nucleus which in turn make contact with neurones of trigeminal nuclei on both sides of the brainstem 25. Moreover collaterals of Me5 periodontal receptors and jaw-closing muscle spindles also reach the parvocellular reticular formation (PCRt) via the bundle of Probst 1'9'31. Neuroanatomical studies have indicated that this PCRt has efferent connections with the trigeminal nucleus oralis, the supratrigeminal area, the Me5, the trigeminal motor nucleus and the facial motor nucleus 25'26~32. The PCRt projections appeared to be bilaterally organized with an ipsilateral dominance. Therefore the PCRt is considered to be a crucial link in the neural circuitry for bilateral trigeminal motor control. Rokx et al. 26 showed that projections from the PCRt to the contralateral Me5 are restricted to the caudal third of this nucleus. As the periodontal afferents are mainly located in this caudal part of the Me53 it is suggested that these afferents in particular are influenced by P C R t input. Jtich and Rokx 2° demonstrated electrophysioiogically ipsilateral projec-
Correspondence: R.F. Minkels, Department of Neurobiology and Oral Physiology, University of Groningen, Bloemsingel 10, 9712 KZ Groningen, The Netherlands.
14 tions of the P C R t to the Me5; by stimulating the P C R t electrically they activated spindle afferent units, facilitated ipsilateral masseter reflexes and r e c o r d e d relayed and direct afferent volleys of P C R t fibres in the Me5. The existence of two populations of P C R t fibres projecting on different regions of the Me5 was indicated. In this p a p e r we will describe experiments in which we explored the contralateral Me5 with microelectrodes and r e c o r d e d contralateral jaw muscle activity during P C R t stimulation. We l o o k e d to see whether: (1) the contralateral Me5 receive projections of different populations of P C R t fibres; (2) the PCRt fibres affect predominantly periodontal afferents in the contralateral Me5 as suggested in the literature; and (3) P C R t stimulation influences reflexes in the contralateral jaw muscles. The results will be discussed in connection to contralateral projections of the PCRt as revealed by tract-tracing techniques using Phaseolus vulgaris leucoagglutinin as an anterograde tracer.
MATERIALS AND METHODS
Electrophysiological experiments The experiments were performed on 27 halothane-anaesthetized rats (male, 250-300 g). A tracheal cannula was inserted and the animal was transferred to a stereotaxic frame. A midline scalp incision was made and a craniotomy was carried out for the introduction of stimulating and recording electrodes. Rectal temperature was maintained at 37 °C. Bipolar glass-coated stainlesssteel electrodes (tip separation 0.7-1.0 mm; impedance 200 kD) were used for electrical stimulation of the PCRt. The stimulus electrodes were connected to the output of a constant current stimulus isolation unit. The stimulus site comprised the area defined by the following coordinates24: AP -1.0 to -2.5 mm, DV 1.0 to 2.0 mm and Lat. 2.0 to 2.4 mm. Extracellular recordings in the Me5 region were made with glass-coated stainless-steel microelectrodes (impedance 1-2 MD). Intracellular recordings from cells in the Me5 area were made with glass microelectrodes filled with 150 mM KCI2'11'z9. Impedance of the electrodes at 30 Hz ranged from 25 to 35 MD. The signals were amplified (1000 x) and filtered (extracellular 5-50 kHz; intracellular 0-10 kHz). All recordings were restricted contralateral to the stimulus site. Muscle spindle afferents in the Me5 were located either by stretching and probing the jaw closer muscles or by stimulating the masseter nerve electrically. Bipolar strap electrodes were used for stimulation of the nerve 19. Periodontal afferents in the Me5 were located by electrical stimulation of the inferior alveolar nerve (IAN). The IAN was stimulated with bipolar enamelled silverwire electrodes (o.d. 110 /~m) with bare ends of 1 mm. One electrode was inserted up to 10 mm into the mandibular canal via the foramen mentale, the other up to 2 ram. In 5 animals the anterior digastric and masseter muscle were fitted with bipolar wire electrodes. The EMGs from these muscles were amplified (1000 x) and filtered (5 Hz-5 kHz). Jaw-opening reflexes were evoked by IAN stimulation; jaw-closing reflexes by applying jerks to the lower jaw21. All data were analyzed on-line (Macintosh II; sampling rate 100 kI-lz) and the results were stored on hard disk for later printing. Extracellular recording and stimulation sites were verified using the Prussian blue marking method. In experiments where pipettes were used, the same marking method was applied, here the glass electrode was replaced by a stainless-steel microelectrode.
Neuroanatomical experiments Iontophoretic injections of Phaseolus vulgaris leucoagglutinin (PHA-L) were made in the PCRt of 6 animals. Glass micropipettes (tip diameter 10-15 ~m) were filled with a solution of 2.5% PHA-L in 0.05 M Tris-buffered saline (TBS, pH 7.4). The tracer was injected by passing an alternating positive current (5/~A; 7 s on-7 s off) through the pipette during 20 rain. After a survival time of 7 days the rats were perfused intraeardially with a solution made up of 2.5% glutaraldehyde, 0.5% paraformaldehyde and 4% sucrose in a 0.05 M phosphate buffer (pH 7.4). The brains were dehydrated for 12 h in a 30% sucrose solution and cut on a cryostat in 40 ~m sections. The sections were incubated with rabbit anti-PHA-L (Dakopatts, 1:2000, 48 h), goat anti-rabbit IgG (Sigma, 1:150, 14 h) and rabbit peroxidase-anti-peroxidase (Dakopatts, 1:800, 4h). Diaminobenzidine (DAB, 0.04%) was used to visualize the peroxidase. The sections were mounted onto gelatin-coated slides, air dried, counterstained with Cresyl violet, dehydrated, coverslipped and studied by light microscope. RESULTS
Electrophysiological experiments Field potentials in the Me5 area evoked by PCRt stimulation. E a c h e x p e r i m e n t c o m m e n c e d with positioning of the stimulus electrodes. This was achieved by recording multi-unit spindle afferent responses from Me5 collaterals in the P C R t with the cathode of the stimulus electrode. Best spindle responses were r e c o r d e d in the P C R t at a d e p t h of 8.5 m m and at a laterality of 2.2 m m at A P - 1 . 5 to - 2 . 0 ram. A f t e r having d e t e r m i n e d the stimulus site, the recording e l e c t r o d e was placed stereotaxically in the Me5 area. Subsequently the Me5 area was e x p l o r e d for responses to P C R t stimulation. Extracellular recordings revealed different types o f n e g a t i v e - p o s i t i v e field potentials in response to the stimulus. T h e s e responses were classified in 3 categories on the basis of their latencies. The earliest response d e t e c t e d after the stimulus artifact a p p e a r e d as a sharp n e g a t i v e - p o s i t i v e deflection (R1, Fig. 1B). The latency of the negative c o m p o n e n t of this response ranged between 0.2 and 0.5 ms, giving a conduction velocity (CV) of 12-30 m/s. T h e distance b e t w e e n the P C R t and the contralateral Me5 (necessary for calculation of the CV) was m e a s u r e d from o u r n e u r o a n a t o m i c a l data as described later in this section. R1 responses showed no change to high stimulus frequencies and were able to follow up to 600 Hz. R1 responses were only r e c o r d e d in the caudal part of Me5 and, outside the Me5, in the supratrigeminal zone and in the area lateral of Me5 (Fig. 3). A second group of responses (R2, Fig. 1A, B) r e c o r d e d in the contralateral Me5 a r e a a p p e a r e d as sharp well-defined n e g a t i v e - p o s i t i v e waves at latencies of 0.7-1.2 ms giving a C V of 5.0-8.5 m/s. A frequency following test showed no variability of timing at stimulus frequencies up to 600 Hz. T h e R2 deflexions were p r o m i n e n t in the supratrigeminal a r e a and in the area
15
~/~~,~V~J~
30I~A
~
43 ~LA
~ ~
2
0
0
i
i
i
!
i
0
1
2
3
4
Fig. 1. Extracellular recordings of unitary R1 and R2 responses in the contralateral mesencephalic trigeminal area due to stimulation of the parvocellular reticular nucleus (at 2 I-Iz, 100/~s; recording site indicated in Fig. 3). A: recording of an R2 response at threshold stimulation (26/~A). Small amplitude compound potentials of the R1 (left arrow) and R2 (right arrow in the subthreshold sweep) are already developed at this stimulus strength. B: recording of an R1 and two successive R2 responses at threshold stimulus for the second R2 response (38 gA; same recording site as in A). A threshold value of 32/~A was found for the R1 response.
I
/
F
\
/
/ /
1O0 p.V
i
i
i
i
i
0
1
2
3
4
ms
Fig. 2. Field potentials of the R3 response recorded within the contralateral mesencephalic trigeminal nucleus (recording site indicated in Fig. 3) following stimulation of the parvocellular reticular nucleus with increasing stimulus strel~,th (at 2 Hz, 100 ks). The stimulus intensity is indicated at the right hand side for each sweep. Maximum amplitude for R3 is reached at a stimulus strength of 60 pA as can be seen from the bottom trace. At a stimulus intensity of 60/~A a small amplitude R2 response was also recorded together with a compound R1 wave which developed with increasing stimulus strength.
I IJ
\
f
j
i
\
A
60 p,A
I
~x ( / M o 5 ( \~ / I
~
l~V
ms
51 I~A
I
PnC
I ~
~.
/
PY
II
I
~-~=R2
~=RI+R3
[~7~] = R I + R 2
[--'-I = R 1
~=RI+R2+R3
Fig. 3. A: diagram of a transverse section of the brainstem showing the contralateral mesencephalic trigeminal area (inset) from which responses were recorded following stimulation of the parvocellular reticular nucleus. B: detailed diagram of the Me5 area showing an overall view of the different sites from which R1, R2, and R3 responses were recorded. The filled circle and asterisk show the recording sites of the responses shown in Figs. 1 and 2, respectively. Cb, cerebellum; LC, locus coeruleus; mcp, middle cerebellar peduncle; Me5, mesencephalic trigeminal nucleus; Mo5, trigeminal motor nucleus; py, pyramidal tract; PnC, pontine reticular nucleus caudalis; PrS, principal sensory trigeminal nucleus; sS, sensory root of trigeminal nerve; SCP, superior cerebellar peduncle.
16 lateral and ventromedial of Me5. Moving the recording electrode dorsally out of the supratrigeminal area into the
400'
A
DI: 115I
S I: 88
E
200
0
400 to
E
200
400"
C
t/)
I DI: 139
CL
S I: 76
E
200"
1 s
0
\
/
Me5 itself resulted in a decrease of the amplitude of the R2 responses (in case of compound potentials). Unitary R2 reactions within the Me5 had small amplitudes and were rarely observed (Fig. 2, bottom trace). The site from which R2 responses were recorded is indicated in Fig. 3. Fig. 2 shows an example of the third category of responses (R3) recorded within the contralateral Me5 area. With stimuli of two times threshold this group appeared as negative-positive waves of relative long duration (about 1.5 ms). The shortest measured latency of the negative component was 1.0 ms. The amplitude of the R3 responses depended on the stimulus strength, indicating that the reactions can be regarded as compound responses. Distinct unitary spikes could not be recorded with the type of metal electrodes used. At stimulus frequencies above 100 Hz the R3 response became smaller in time and disappeared at frequencies above 500 Hz. Large amplitude R3 responses as shown in Fig. 2 were only recorded within the ventrocaudal part of Me5 (Fig. 3). Often negative-positive small amplitude wavelets appearing at latencies of 1.2-2.5 ms were recorded outside the Me5 in the supratrigeminal zone and the area laterally of Me5. Because these wavelets responded as R3 waves to changes in stimulus strength and frequency and appeared within the same time domain as the large R3 waves did, we considered them as small R3 responses. We never observed R1, R2 or R3 responses in the rostral part of the contralateral Me5 nucleus. Effects o f P C R t stimulation on Me5 neuron activity. The action of high-frequency PCRt stimulation (100 Hz) was studied on jaw muscle spindle afferents activated by ramp-and-hold stretches. We have subjected 24 units to this test (no attempts were made to discriminate between primary and secondary afferents). Eight units responded with an increase in activity; 16 units did not respond. Inhibition of spindle afferent activity was never observed. Fig. 4 shows an example of a masseter unit that responded to PCRt stimulation with an increase in the dynamic and static index during ramp-and-hold stretches 6. All recorded spindle units could not be fired by PCRt stimulation alone; neither by single PCRt stimulus pulses nor by high frequency PCRt stimulation.
Fig. 4. Ramp-and-hold histograms of a masseter spindle afferent in the contralateral mesencephalic trigeminal nucleus before (A), during (B) and immediately after (C) stimulation of the parvoccllular reticular nucleus (at 100 Hz, 50~s and 30#A). Each histogram (bin width 30 ms) is the average of 10 ramp-and-hold stretches. This unit responded to PCRt stimulation with an increase in the dynamic (DI) and static index (SI). The movement of the lower jaw (max. amplitude 4 mm) is shown at the bottom of the figure.
17
j• l
A[20 mV =
I 2 mV
__ --_-- . . . . . . . . . . . . . . . . .
. . . . . . .
i
|
|
|
|
!
0
5
10
15
20
25
ms
B
20 mV
0
5
10
15
20
25
ms
I 2 mV 0
5
10
|
|
!
15
20
25
ms
Fig. 5. Intracellular recordings of identified cells in the mesencephalic trigeminal nucleus. A: masscter spindle unit identified by muscle stretching and muscle probing. B: periodontal afferent identified by stimulation of the inferior alveolar nerve (20/~A, 100 /~s). C: interneuron located in the ventral border of Me5 that responded to IAN stimulation (20 /~A, 100 /~s). Upper traces: identificationof the cell, lower traces: response to stimulation of the parvocellular reticular formation with single pulses (50/~A, 50 ks).
Spindle unit activity evoked by single pulses applied to the masseter nerve was also not affected by high frequency PCRt stimulation. As spindle units could not be driven by PCRt stimulation alone we interpret the observed increase in spindle activity during the rampand-hold stretches as a result of fusimotor action induced by PCRt stimulation. In addition to these experiments we have recorded from spindle afferent units intracellulady (22 units). The spindle afferents were localized in the Me5 by stimulation of the masseter nerve and by muscle probing, and the effect of PCRt stimulation on the membrane potential was studied. We never observed any IPSPs or EPSPs in response to either single or short train (up to 5 pulses) PCRt stimulus pulses (Fig. 5A). Identical effects (Fig. 5B) were observed in membrane potentials of periodontal afferents (16 units recorded) and of interneurones (responding to IAN stimulation, 5 units recorded) in the Me5 and the supratrigeminal zone (Fig. 5C). Both periodontal and interneurone membrane potentials were not affected by the PCRt stimulus. Additionally it was found that periodontal activity evoked by IAN stimulation was also not affected by high frequency (100 Hz) PCRt stimulation. Effect of PCRt stimulation on jaw reflexes. During stimulation of the PCRt (at 100 Hz) the jaw-closing as well as the jaw-opening reflex in the relating contralateral muscles were facilitated progressively with increasing stimulus strength. Stimulation of IAN at subthreshold levels produced a small jaw-opening reflex when applied simultaneously with the PCRt stimulus (Fig. 6A). Stimulation of IAN at suprathreshold level resulted in a further increase in opening reflex activity (Fig. 6B). Similar results were obtained for jaw-closing reflexes (Fig. 6C). Neuroanatomical experiments. The anterograde tracing studies revealed that the PCRt projected predominantly onto the ventrocaudal part of the contralateral Me5 nucleus. The fibres left the PCRt mainly in a ventromedial direction, crossed the midline just dorsal to the pyramidal tract and ran dorsolaterally to the Me5 area. The course of these fibres is illustrated in Fig. 7A. Labelled fibres were also traced to more rostral parts of the brain. These fibres travelled through the Me5 area and innervated cerebellar, mesencephalic, thalamic and forebrain structures. A few synaptic structures (varicosities) were observed in the Gi and PnC. Intensive labeling was found in the PCRt, the Mo5, the supratrigeminal area and the medial part of the facial motor nucleus. A detailed outline of the intensive bouton labeling in the contralateral caudal Me5 area is illustrated in Fig. 7B.
18 DISCUSSION
A control
50 I.tA 100 IsV i
i
5
10
!
15 ms
mm
B control
L
50 btA 500 IsV
I
I
!
!
0
5
10
15
ms
C ":--'-"- control
~
40 ~A
I |
i
0
5
10
|
i
i
15
20
25
100 pV ms
The extracellular recordings showed that the R2 responses were distinct and had large amplitudes in the area ventrally, ventrolaterally and ventromedially of the caudal Me5. Small R2 responses however were recorded within the ventrocaudal part of the nucleus. R3 waves on the other hand showed opposite features; large amplitudes within the ventrocaudal Me5 and small amplitudes outside. The finding that the R3 waves were only recorded as compound potentials of relative long duration and that they could not easily follow high frequency stimulation indicate the existence of a strong collateralization of small fibres within the Me5. We regard the R3 potentials as being recorded from collaterals of axons from which the R2 waves originate. These axons then approach the Me5 at its caudal-ventral edge and ramify on entering the nucleus. The findings that latencies of the R3 waves were higher than the latencies for the R2 potentials and even slightly overlapped are in line with this view. As R1 responses were detected inside as well as outside the Me5, we consider them as being recorded from PCRt axons travelling through the Me5 area to more rostral brain areas. Our physiological observations are supported by the neuroanatomical results. Here we found labeled PCRt fibres approaching the contralateral Me5 at its ventral border and terminating in its ventrocaudal part. It is conceivable that we recorded the R2 and R3 potentials from these fibres. The R1 responses then might be recorded from the fibres running through the Me5 area to the cerebellum, thalamus and forebrain. Therefore we tentatively conclude that the contralateral Me5 receives projections from only one population of PCRt fibres. The morphology and central course of mesencephalic trigeminal neurones innervating the jaw muscle spindles and periodontal ligament is well documented 1'9"27'28. These authors indicate that sensory information from the periphery will arrive at the level of the trigeminal motor nucleus first, then reach the trigeminal nucleus oralis followed by the supratrigeminal nucleus and finally arrives at the Me5. By following this central course via collaterals of Me5 cells, action potentials evoked in the
Fig. 6. Effect of high frequency stimulation of the parvocellular reticular nucleus on jaw reflexes in the contralateral musculature. A: activity in the anterior digastric muscle in response to a subthreshold stimulus applied to the inferior alveolar nerve (10 gA, 100 ps). B: reflex EMG activity in the anterior digastric muscle evoked by IAN stimulation (100 gA, 100 #s). C: jaw-closing reflex activity in the masseter muscle evoked by a twitch applied to the lower jaw. Upper traces: control; lower traces: the same reflexes but now during high frequency stimulation of the PCRt (100 Hz, 50 #s; stimulus strength for each record indicated at the right).
19
SCO
(;
,.v-?
::i. ,-~
f
