Unraveling a masticatory - oculomotor neural pathway in rat: Implications for a pathophysiological neural circuit in human?

Abstract

Anterograde tracers were injected into the mesencephalic trigeminal nucleus (Vme) in pons, labeled axons and terminals were observed in ipsilateral oculomotor (III) and trochlear (IV) nuclei, as well as in interstitial nucleus of Cajar and Darkschewitsch nucleus (INC/DN), the well-known premotor nuclei to the III/IV, but not in abducens nucleus and central mesencephalic and paramedian pontine reticular formation (CMRF/PPRF). Retrogradely labeled INC/DN neurons do ensue from injection of tracers into the III. Confocal microscopy revealed labeled Vme axonal terminals contact with labeled pre-oculomotor neurons in the INC/DN. In response to electrical stimulation of trigeminal nerve root (TR) jaw muscle branches, which contains peripheral processes of the jaw muscle spindle, extracellular unit discharges were recorded in the ipsilateral III/IV and INC/DN. Electromyography (EMG) was also recorded from superior rectus (SR) and levator palpebrae (LP) following electrical stimulation of the TR. Moreover, stimulation of the TR induced Fos expression in the INC/DN pre-oculomotor neurons, but not in CMRF/PPRF that harbors horizontal eye moving premotor neurons. By injection of retrograde tracers into the III combined with Fos immunostain, double labeled pre-oculomotor neurons were observed in the INC/DN. About 80% of retrogradely labeled III premotor neurons express Fos. These results suggest a neural pathway from the masticatory Vme neurons to the oculomotor system that is probably involved exclusively in vertical and torsional eye movement as well as eyelid retraction. The potential relationship between this pathway and Marcus Gunn Syndrome (MGS), a congenital jaw-winking syndrome, was discussed.

Introduction

Marcus Gunn Syndrome (MGS), reported and described by Marcus Gunn more than 120 years ago, manifests as an abnormal eye movement and prompt eyelid retraction conjugated with jaw movement [1-4]. The mechanism and neural circuits underlying the disease remains an enigma, and there is still no efficacious treatment for this disease [5, 6]. In clinical electromyography (EMG) studies on the MGS cases, distinct co-firing of masticatory and extraocular muscles was recorded when both muscle groups were examined simultaneously [3, 4]. Moreover, stimulation of the pterygoid muscle nerve can elicit ipsilateral eyelid retraction, and section of the nerve from the trigeminal motor root could relieve the eyelid activity [7], suggesting an intrinsic linkage between the masticatory and extrtaocular muscle systems. Scientists who studied the MGS thought that the masticatory-oculomotor reflex observed in the MGS cases was probably a primitive reflex that was extinguished during phylogenetic development and might be released under certain disease conditions, a so-called “release hypothesis”[3, 4, 8]. However, it is unknown whether the neural pathway still exists after the extinguishment of the reflex during development. It is also unknown what would be the neural circuit conducting this “primitive reflex”, nor the mechanism(s) underlying the release of such a “primitive reflex”. Hiscock and Straznicky have reported that the central processes of the mesencephalic trigeminal nucleus (Vme) neurons, those innervate the temporalis, project directly to the oculomotor nucleus (III) and trochlear nucleus (IV) in Xenopus toad [9]. This neuronal circuit may help an amphibian project and focus its eyes on its prey when the animal targets to the prey and opens mouth widely [9]. But it is unknown whether such a neural circuit exists in mammals or even human. We deduce that this neural pathway might be related to MGS pathogenesis if it exists in human. To this end, we conducted current experiments in rat aiming at obtaining experimental evidence in supporting our deduction. Through injection of tracers into the Vme of rat, we found out labeled terminals in the III and IV (III/IV), and in their premotor (pre-oculomotor) neurons [10-13] situated in interstitial nucleus of Cajal and Darkschewitsch nucleus (INC/DN). We also recorded extracellular unit discharges in the III/IV and INC/DN in response to electrical stimulation of trigeminal nerve root (TR) that contains jaw muscle nerve branches. Moreover, we examined EMG recorded in extraocular muscles in response to stimulation of the TR and analyzed Fos expression along the aforementioned neural pathway.

Materials and methods

Animals

Thirty three adult Sprague-Dawley rats (∼300g in body weight), purchased from Charles River Laboratories (Wilmington, MA), were used in this study. All experimental protocols and animal care were carried out in accordance with the National Institutes of Health Guide for the Care of Laboratory Animals in Research. All efforts were made to minimize animal suffering and the number of animals used in this study. Another 10 Sprague-Dawley rats (300-350g) were obtained from Animal Facilities of the Shanxi Provincial Eye Research Institute and Eye Hospital. All procedures of the experiments were compliance with Guidelines for the Care of Laboratory Animals in Research issued by The Chinese Academy of Sciences, approved by the institute and hospital.

Biotinylated Dextran Amine (BDA) anterograde and chloride toxin B (CTB) retrograde tract tracing

The animals were anesthetized with sodium pentobarbital (40 mg/kg, i.p.) and were placed on a stereotactic apparatus. Ten percent BDA (Molecular Probes, USA) in saline was delivered to the Vme by iontophoresis with 10-20 μA positive current, in 250 mS ON/OF model for 10 min. The coordinates used for injection were 6-7 mm posterior to Bregma and 1.3-1.4 mm lateral to Midline [14], with 20° of anterio-posterior inclination of electrode holder. After 5 days, animals were re-anesthetized in a way same as the above mentioned and 2% CTB (Sigma Chemical, USA) in saline was iontophorized into the III controlateral to the Vme injection with 10-20 μA positive current, in 250 ms ON/OF model for 10 min. Coordinates are 6.5-6.8 mm posterior to Bregma and 0.25-0.40 mm lateral to Midline according to “The rat brain stereotaxic atlas” [14]. All injections were performed unilaterally with the III injection in opposite side, because the pre-oculomotor neurons innervate the III/IV motor neurons crosswise in lateral-eyed animal. After another 5 days, the animals were euthanized with sodium pentobarbital (140 mg/ kg, i.p.) and perfused transcardially with saline and 4% paraformaldehyde. The brain was removed and the pons and midbrain were cut into coronal sections (30 mm in thickness). The brain sections were incubated with goat anti-CTB (List Biological Labs, USA), and followed by a cocktail of streptavidin-fluorescein (Alexa Fluor 488, Molecular Probes) and donkey anti-goat Alexa Fluor 568 (Molecular Probes, USA). Then, the sections were mounted and sealed with Vectashield mounting medium (Vector), examined and photographed with Nikon E-600 fluorescent and Bio-Rad 1024 Laser Scan Confocal Microscopes.

Recording of extracellular unit discharges

The animals were anesthetized with urethane (1.25∼1.50 g/kg, i.p.) and experiments were started after disappearance of limb-withdrawal reflex. The anesthesia usually lasted for 6-8 hours and there were no remarkable changes (> 10 mmHg) on blood pressure during electrical stimulation of TR. The TR was surgically exposed to its end against bone and a bipolar silver-wire collar electrode was placed around the trunk of the root, and then secured with Kwik-Cast Silicon Sealant glue (World Precision Instruments Inc., USA) to isolate the electrode from the surrounding tissues. The animal was paralyzed with gallamine triethiodide (30 mg/ kg, i.p.), placed onto a stereotaxic apparatus, and ventilated artificially (2cm³, 1.7 Hz). A parietal craniotomy was performed ipsilateral to the TR stimulation side and the brain exposed was covered with mineral oil (37°C). For extracellular unit discharge recordings, threshold stimulation intensity (T; 0.13 - 0.17 mA) was determined at the beginning of each experiment and 1 to 2.5 T was used for experiments. Core body temperature was maintained at 37°C with thermostatically controlled heating pad.

The recording electrode, filled with 2% Pontamine sky blue in 0.5 M sodium acetate (pH 7.4) and had an impedance of 8-10 MΩ, was advanced into the III, IV and INC/DN ipsilateral to the TR stimulation side. Unit discharges elicited by electrical stimulation of TR were amplified with an Axoclamp2B amplifier, digitized through Digidata 1322A, and recorded on a Dell PC computer equipped with pCLAMP9 software (Molecular Devices, Sunnyvale, CA) for data acquisition and analyses. The recording site was labeled with sky blue at the end of each experiment. The animal was then euthanized and transcardially perfused. The brain was removed and cut into 50μm-thick sections which were counterstained to view the recording site.

Recording of superior rectus and levator palpebrae EMG

In each experiment, animal was anesthetized and the TR was exposed and stimulated using the similar protocols as described above. To record superior rectus and levator palpebrae EMG, an incision was made along the upper eyelid ipsilateral to the TR stimulation. Then, conjunctive membrane was cut and fatty tissue was removed to expose superior rectus (SR) and levator palpebrae (LP). A unipolar stainless electrodes coated with Teflon were placed into the middle of the incision for SR/LP EMG recording. The incision was then closed with instant bio-glue. A reference electrode was placed on the earlap opposite to the TR stimulation and recording site. EMG was induced either by train electrical stimulation (1-5T, 50 Hz, 0.2 ms pulse duration, 500 ms train duration, 2.5 s train intervals) or by pressing down lower jaw. An NTS-2000 EMG-EEG multi-channel amplifier (Pukang Electronic Tech. Ltd, Shanghai, China) was used to record EMG and signals were filtered at 1 to 5 KHz.

Induction of Fos expression in retrogradely labeled pre-oculomotor neurons in the INC/DN following electrical stimulation of TR

Train stimulation (1T, 50Hz, 0.2ms pulse duration, 200ms train duration, 300ms train interval) was applied to the TR for 10 S to induce Fos expression. The animals were euthanized at 30 min, 1 h and 2 h after stimulation. The midbrain was removed and cut into 30μm-thick sections. Theses coronal sections were incubated with rabbit anti-Fos (Santa Cruz Biotech, USA) followed by donkey anti-rabbit-Alexa Fluor 488 (Molecular Probes). Since Fos-like immunoreactivity appeared 30 min after stimulation and peaked at 1 h, therefore 1h-time point was used for double labeling. In double labeling group, animals were anesthetized with sodium pentobarbital (40 mg/kg, i.p.) and placed on a stereotaxic apparatus. Two percent CTB was delivered into the III controlateral to the TR stimulation side. Five days later, animals were re-anesthetized and the TR was exposed for electrical stimulation as described in section 3. The animals were then euthanized 1 h later and perfused transcardially with saline and 4% paraformaldehyde. Brain sections (30 μm in thickness) were cut and incubated with both rabbit anti-Fos and goat anti-CTB, followed by donkey anti-rabbit-Alexa Fluor 488 and anti-goat-Alexa Fluor 568 to reveal double-labeled INC/DN pre-oculomotor neurons. The sections were examined and photographed with Nikon E-600 fluorescent and Bio-Rad 1024 Laser Scan Confocal Microscopes.

Results

Projections of Vme neurons to the III and IV, and to their premotor neurons in the INC/DN

By injection of BDA into the caudal Vme (Figure 1A), labeled Vme neuronal axons and boutons were found in the III and IV (Figure 1B) exclusively ipsilateral to the injection side. The Vme neuronal axons and terminals were also observed in ipsilateral INC/DN area where pre-oculomotor neurons are located (Figure 1D). In the III, the labeled axons and terminals were located predominately in ventral portion of the nucleus with a few scattered axons and terminals in dorso-medial part and few labeling in dorso-lateral division of the nucleus (Figure 1B). By injection of CTB into contralateral III (Figure 1C), retrogradely labeled premotor neurons were seen to be co-localized with the Vme axonal terminals in the INC/DN (Figure 1D). Under confocal microscopy, many Vme neuronal terminals were found to have close contact with pre-oculomotor neurons in INC/DN area (Figure 1E and F). No labeled fiber or terminal was observed in the abducens nucleus (VI) and the primate central mesencephalic and paramedian pontine reticular formations; those harbor horizontal eye-moving motoneurons (the former) and their premotor neurons (the latter). Our results indicate the Vme neurons project to both the III/IV and their premotor neurons innervating vertical-torsional eye movement, but not project to the VI and their premotor neurons those are in charge of horizontal eye movement.

Figure 1.

A, BDA injection site in the Vme. B, BDA labeled fibers and terminals are distributed completely in unilateral III and IV, and coronally the labeling is located mainly in ventral part. C, CTB injection site in the controlateral III. D, Overlapping of BDA labeled Vme neuronal fibers and terminals with CTB-labeled pre-oculomotor neurons in the INC/DN, ipsilateral to the BDA injection and controlateral to the CTB injection. The framed areas were further investigated using confocal microscopy. E and F, confocal images from framed areas in D, showing close contacts (arrowheads) between BDA labeled Vme neuronal terminals and CTB labeled pre-oculomotor neurons in the INC/DN. 3V, 3rd ventricle. III, oculomotor nucleus. Aq, aqueduct. DN, Darkschewitsch nucleus. INC, interstitial nucleus of Cajal. LC, locus ceoruleus. MPB, medial parabrachial nucleus. PAG, periaqueductal gray. scp, superior cerebelar peduncle. Vme, mesencephalic trigeminal nucleus. Vmo, trigeminal motor nucleus. Vp, principal sensory trigeminal nucleus. Scale bars: 200 μm in A-D; 20 μm in E and F.

Orthodromic responses of the INC/DN and the III/IV neurons to the TR stimulation

To further test functional connection of the aforementioned projections, extracellular unit discharges were recorded in the INC/DN and III/IV following the TR stimulation. Twelve single unit discharges in the INC/DN area (Figure 2A) were recorded. Five units responded to the single TR stimulation, the latency was 4.7±2.9 ms, a typical unit response recorded in the INC (Figure 2A) is shown in Figure 2C. Other 7 units exhibited spontaneous discharges and the discharge rates were altered by train stimulation (1 -2.5T, 50 Hz, 200ms train duration, 5-8 trains) applied to the TR (data not shown). These results indicate innervations of jaw muscle afferent Vme neurons onto the INC/DN neurons.

Figure 2.

A and B, sky blue marked recording sites in the INC (A) and the III (B) are indicated by arrowheads. C and D, orthodromic unit responses recorded in the INC (C) and the III (D), in which downward arrowheads indicate the start of electrical stimulation on the trigeminal nerve root which contains Jaw muscle nerve branches. EW, Edinger-Westphal nucleus, fr, fasciculus retroflexus. PaR, pararubral nucleus. pc, posterior commission.

Eight single units were recorded in the III (Figure 2B) and IV. Half of them responded to the single TR stimulation without spontaneous discharge before and after stimulation, the latency was 3.9±1.8 ms. These 8 units showed diversified properties, two of them fired spontaneously but responded only to the train stimulation, two with spontaneous discharge responded to both single and train stimulation. Two units appear to have a second discharge with longer latency after a single stimulation. A unit with such a property recorded in the III (Figure 2B) is shown in Figure 2D. Although the mechanisms underlying the observed diversified properties remain to determined, it may reflect the nature of multi-synaptic activities.

Superior rectus and levator palpebrae EMG evoked by the TR stimulation

Ten animals were used in this study. EMG was successfully recorded in SR/LP of 8 animals and a typical recording is shown in Figure 3A and B. The stimulation intensity was 1-5T of the threshold intensity used in extracellular unit recording and the recording usually lasted for 0.5 to 1 min. In majority of the recordings, the EMG was initially small in magnitude and turned to be larger after train stimulation (50Hz, 0.2ms pulse duration, 500ms train duration, 2.5s train interval) for about 10 - 15 seconds. However, pressing down the lower jaw did not elicit clear EMG (data not show), inferred induction of the EMG requires intensified or enforced stimulation.

Figure 3.

A, schematic diagram illustrating the exposed trigeminal nerve root that contains jaw muscle nerve branches and the localization of a bipolar stimulation electrode, as well as the placement of EMG recording electrodes. B, an exemplary EMG recorded from superior rectus and levator palpebrae (upper line), and the stimulation artifacts of train stimulations of the TR (lower line).

Fos Expression in the Pre-oculomotor neurons in the INC/DN

To exam the nuclei or neuronal groups involved in a functional pathway, Fos expression was frequently used for such examination. Following electrical stimulation of the TR which was similar to the stimulation used for elicitation of orthodromic discharge and SR/LP EMG, Fos-like immunoreactivity was induced in both sides of the INC/DN with stronger signal in the side ipsilateral to the TR stimulation, but not in central mesencephalic and paramedian pontine reticular formation - the brain areas harboring the horizontal eye-move premotor neurons [10]. A clear Fos-like immunofluorescent stain was visualized 30 min after electrical stimulation of the TR and peaked at 60 min (Figure 4C). As a result, we chose 1 hour point to do double labeling. After administration of CTB into the III contralateral to the stimulation (Figure 4A), retrogradely labeled pre-oculomotor neurons were observed in the INC/DN ipsilateral to the stimulation (Figure 4B). Numerous CTB and Fos double labeled neurons were observed in the INC/ DN (Figure 4D). By counting every other section, it was uncovered that about 80% (averaged from 3 animals) of CTB labeled pre-oculomotor neurons express Fos protein.

Figure 4.

A, CTB injection site in the III. B, CTB retrogradely labeled pre-oculomotor neurons in the INC/DN contro-lateral to the injection. C, Fos-like immunocytochemistry stain in the INC/DN, induced by stimulation of the TR. D. Double labeling of the INC/DN pre-oculomotor neurons by both CTB and Fos like immunostain (arrowheads). mlf, medial longitudinal fasciculus. Scale bars: 200 μm in A; 50 μm in B-D.

Discussion

The present studies have demonstrated a neural pathway from the masticatory to oculomotor system via the Vme neurons in the rat. It has been shown that the central axons of the Vme neurons distributed broadly in the unilateral brainstem in the rat and cat [15, 16], while their peripheral axonal terminals innervate the ipsilateral masticatory muscle spindles through the trigeminal nerve branching into different masticatory muscle nerves [17]. These Vme neuronal peripheral axons are activated when the TR, which contains the jaw muscle branches, is electrically stimulated [18-20]. In a previous study carried out in rat, Luo and colleagues [15] showed that some intracellular labeled central axons of the masseter muscle spindle afferents climbed up into the ipsilateral III as demonstrated by intracellular recording and horseradish peroxidase injection upon jaw muscle spindle afferent Vme neurons, indicating the primary afferents of this pathway may be the jaw muscle spindle Vme neurons. Such a potential pathway is supported by our anterograde labeling study in which injection of tracer into the Vme resulted in labeled axons and terminals in the III/IV and INC/DN, and by our electrophysiological study that orthodromic discharges in the III/IV and INC/DN were evoked by TR stimulation, as well as by our extraocular muscle EMG study demonstrating distinct SR/LP EMG can be elicited by TR stimulation.

It has been reported in Xenopus toad that temporal muscle afferent Vme neurons project to the III/IV motoneurons directly [9] and this neuronal circuit may help a toad protrude its eyes to its prey when it opens mouth widely and preys. A rat never preys like a toad, so rat must not have a reflex of staring a prey with a wide-opened mouth as the toad does [9]. However, the results observed in our studies demonstrated the existence of a potential neural pathway in rat that could conduct such a reflex. This finding implies that although this “primitive reflex” is believed to be distinguished during phylogenetic development, the neural pathway might be preserved but does not function under normal condition, or the conduction of this pathway is strongly inhibited in normal circumstance. This implication is supported by experimental results showing all of physiological responses recorded were evoked by enforced electric stimulations, but a nature stimulation such as lowering down the jaw failed to elicit responses. Based on the results from animal studies, it is reasonable to deduce that a similar “distinguished” masticatory - oculomotor neural pathway may exist in human. Indeed, clinical EMG studies on the MGS cases in which cofiring of masticatory and extraocular muscle was observed when EMG was recorded simultaneously on both muscle groups [4]. Stimulation of the pterygoid muscle nerve elicited ipsilateral eyelid retraction, and ablation of the nerve from trigeminal motor root relieved the eyelid activity [7]. Moreover, some patients manifested only temporal MGS or temporal trigemino-oculomotor synkinesis in life, and others exhibited a pattern of alternative healing and relapse [4, 8]. Although these data were from the MGS patients [4, 7], not from healthy people, it is quite possible, in light of these evidences, that the neural pathway uncovered in the present study, or the “primitive” trigemino-oculomotor synkinetic reflex pathway also exists in healthy human. This pathway is most likely shut off in normal environment, but “released” under certain disease conditions. Therefore, our results support the “release hypothesis” on the pathogenesis of MGS [3, 4, 7, 8]. In addition, the SR/ LP EMG can be recorded usually after stimulation for 10-15s, implying involvement of an arousing process or a process of disinhibition. This phenomenon stimulated us to study Fos expression and the Fos-like immunoreactivity was observed in the INC/DN.

The Fos expression induced by special stimulation has been widely used in exploring functional neuronal pathway for many years [21, 22]. The present study revealed the projections of the Vme neurons to pre-oculomotor neurons in the INC/DN and these projecting terminals do contact with retrogradely labeled premotor neurons of the III. Orthodromic unit discharges were recorded in the INC/DN following electric stimulation of the TR and distinct SR/LP EMG was also evoked by similar stimulation. Further, Fos expression was induced by the similar stimulation. Interestingly, Fos expression was not found in the central mesencephalic and paramedian pontine reticular formation following TR stimulation, and nor projections from the Vme neurons to these areas.

It has been well documented that the INC is a pre-oculomotor center that controls verticaltorsional eye and head movements and the DN was a member of accessory oculomotor nuclei reported in early studies [11-13]. The III/IV motoneurons themselves predominately innervate the vertical and torsinal eye movement muscles, as well as the LP - the eyelid elevation muscles [12, 13, 23]. However, the premotor neurons those manipulate horizontal eye movement are demonstrated to locate in the central mesencephalic and pontine reticular formation [10, 13]. Hence, innervations of those Vme neurons to the pre-oculomotor neurons in the INC/ DN may not be simply a non-functional residual circuit. Whether this pathway is functional and what kind function this pathway may have are still unknown. It is probably related to vertical and torsional eye movements including eyelid activity with implications for MGS pathogenesis, based on evidences from both our earlier [15] and current works and previous studies carried out by others [7, 11-13, 23].