·Basic Research·
Is
there a primitive reflex residue underlying Marcus Gunn Syndrome? Rat
electrophysiology
Ying
Qiao1, Hou-Cheng Liang1,2,3, Jing-Dong Zhang1,2,
Pi-Fu Luo4, An-Le Su1, Ting Zhang1, Hong-Na
Zhu1
1Xi’an First Hospital, Key Clinic
Ophthalmology Lab, Shaanxi Province Eye Research Institute, Xi’an 710002,
Shaanxi Province, China
2Xi’an BRIGHT Eye Hospital, 234 West
Youyi Road, Xi’an 710068, Shaanxi Province, China
3Department of Ophthalmology,
Affiliated Hospital, Xi’an Jiaotong University, Xi’an 710002, Shaanxi Province,
China
4Department of Pathology, University
of Iowa College of Medicine, Iowa City, IA 52242, USA
Correspondence to: Jing-Dong Zhang. Department
Anesthesiology, University of Cincinnati Medical Center, Cincinnati, OH 45267,
USA. zhang2jd@ucmail.uc.edu; Hou-Cheng Liang. Xi’an BRIGHT Eye Hospital, 234
West Youyi Road, Xi’an 710068, Shaanxi Province, China. lianghc1@163.com
Received:
Abstract
AIM: To make an
electrophysiological demonstration of a possible jaw muscle
afferents-oculomotor neural pathway that was proposed by our previous works on
rats, which substantiates an early “release hypothesis” on pathogenesis of
human Marcus Gunn Syndrome (MGS).
METHODS: Extracellular unit
discharge recording was applied and both orthodromic and spontaneous unitary
firing were recorded in the oculomotor nucleus (III), and the complex of
pre-oculomotor interstitial nucleus of Cajal and Darkschewitsch nucleus (INC/DN),
following electric stimulation of the ipsilateral masseter nerve (MN) in rats.
RESULTS: Extracellular
orthodromic unit discharges, with latencies of 3.7±1.3 and 4.7±2.9ms, were
recorded unilaterally in the III, and the INC/DN neurons, respectively.
Spontaneous unit discharges were also recorded mostly in the INC/DN and less
frequently in the III. Train stimulation could prompt either facilitation or
inhibition on those spontaneous unit discharges. The inhibition pattern of
train stimulation on the spontaneous discharging was rather different in the
III and INC/DN. A slow inhibitory pattern in which spontaneous firing rate
decreased further and further following repeated train stimulation was observed
in the III. While, some high spontaneous firing rate units, responding promptly
to the train stimuli with a short-term inhibition and recovered quickly when
stimuli are off, were recorded in the INC/DN. However, orthodromic unit
discharge was not recorded in the III and INC/DN in a considerable number of
experiment animals.
CONCLUSION: A residual neuronal
circuit might exist in mammals for the primitive jaw-eyelid reflex observed in
amphibians, which might not be well-developed in all experimental mammals in
current study. Nonetheless, this pathway can be still considered as a
neuroanatomic substrate for development of MGS in some cases among all MGS with
different kind of etiology.
KEYWORDS: masseter
nerve; single unit discharge; oculomotor nucleus; pre-oculomotor neurons;
interstitial nucleus of Cajal/Darkschewitsch nucleus; Marcus Gunn Syndrome
DOI:10.18240/ijo.2020.01.05
Citation: Qiao
Y, Liang HC, Zhang JD, Luo PF, Su AL, Zhang T, Zhu HN. Is there a
primitive reflex residue underlying Marcus Gunn Syndrome? Rat
electrophysiology. Int J Ophthalmol 2019;13(1):29-35
INTRODUCTION
An oculomotor dysinnervation
disorder known as Marcus Gunn Syndrome (MGS) is characterized by abnormal eye
and eyelid movements and, most notably, eyelid retraction that is elicited by
jaw movements[1-2]. This
disorder is also termed trigemino-oculomotor synkinesis or Jaw-Winking. Early
clinical electromyography (EMG) studies on MGS cases showed distinct co-firing
of masticatory and extraocular muscles when EMG was recorded from both muscle
groups simultaneously[3-4].
Moreover, stimulation of the pterygoid muscle nerve elicited ipsilateral eyelid
retraction, and section of this nerve from the trigeminal motor root could
relieve the eyelid activity[3-4].
These findings suggested an intrinsic linkage between the masticatory and
extraocular muscle systems[3-4].
Based on these findings, a “release hypothesis” proposed[3-4] that a primitive masticatory oculomotor reflexive
circuit is preserved, but is normally suppressed in humans, and it can be
released and manifested whenever congenital or traumatized disorders occur in
the system. However, the existence of a primitive masticatory oculomotor
reflexive pathway remains to be shown. Coincidently, a neural tract tracing
study unveiled that central processes of the temporal muscle afferent
mesencephalic trigeminal nucleus (Vme) neurons project directly to oculomotor
(III) and trochlear nuclei (IV) in Xenopus toad[5].
This neuronal circuit could be useful in a hunting scenario in which amphibians
could simultaneously open their eyes and mouths widely to follow and catch a
prey. Interestingly, in our recent studies in rats[4-6], we also observed projections of the Vme neurons to the
III/IV, and even to their premotor neurons in the interstitial nucleus of Cajal
(INC) and Darkschewitsch nucleus (DN).
We considered INC and DN together as
a pre-oculomotor complex because early tract tracing study on monkey showed
projections from INC/DN to III/IV and termed this area as accessory oculomotor
nuclei[4,6-7].
Furthermore, in our own previous studies in rats, a large number of retrograde
labeled cells was constantly observed in this combined area and equally
distributed in both INC and DN following injection of retrograde tracer into
the III or IV[4,6-7].
In addition, more intriguing, c-Fos expression was consistently induced in this
combined area either by electric stimulation of masseter nerve (MN) or after
repeated and rapid down-stretching the lower jaw[6-7].
It is also noteworthy that a recent
clinic study in humans without any congenital ptosis, the authors observed the
same retraction of the upper eyelid by electric stimulation of pterygoid nerve
as reported previously in MGS cases[8]. This
report implied that this masticatory oculomotor pathway is not merely present
in MGS patients, but is also expressed in some people without congenital
ptosis. Nonetheless, the correlated reflex is strikingly manifested in MSG
patients according to “release hypothesis”. We have consistently witnessed
co-firing of eyelid and jaw closing muscles during steady occlusion (isometric
contraction of masseter without jaw movement) by recording of EMG from both the
eyelid and jaw muscles simultaneously in healthy human volunteers[9]. All evidences aforementioned substantiate the
existence of a residual of jaw muscle-oculomotor reflex pathway and supports
the “release hypothesis”. Therefore, in the present work, we attempt to further
our early[7] and recent exploration[4,6] of a masticatory oculomotor neural
pathway in the rat by electrophysiological extracellular unit recording from
neurons of the III and the INC complex (namely, the INC/DN) following electric
stimulation of the MN.
SUBJECTS AND METHODS
Ethical Approval All surgical procedures and animal care
were carried out in accordance with the Guidelines for the Care of Laboratory
Animals in Research issued by The Chinese Academy of Sciences. The experiment
protocol was approved by the Medical College and Affiliated Hospital.
Animals Totally 37 adult male Sprague-Dawley rats (300
Single Unit Recording Following Electric Stimuli of
Unilateral Masseter Nerve Animals were anesthetized with urethane (1.25
Figure 1 The location of bipolar
silver-wire stimulating electrode on the trunk of MN (A) and the position of
recording electrode on the III and INC/DN in the midbrain (B).
Extracellular recording electrodes
with a tip diameter of 1.0-2.0 μm were made from borosilicate glass pipettes (
RESULTS
Responses of the III Neurons to the MN Stimulation A total 13 extracellular units were recorded at the III
(Figure
Figure 2 Unitary discharges and
recording sites in III A, D, G: Recording sites
marked by sky blue (arrows); B, C: Orthodromic unitary responses recorded from
the III shown in the A, and downward arrowheads indicate the starting of the
stimulation; E: A spontaneous discharging unit encountered in the III in D with
a pattern of been excited, relaxed and further excited after each train stimuli
of the MN. Upward arrowheads indicate the start of the train stimuli. F: A real
time recording extracted from the framed area in E (indicating by a thin arrow;
from 46 000 to 56 000ms). The arrowhead points to the start of the train
stimuli. H: A real time recording from a spontaneous firing unit at III shown in
G. Upper track displays gradually reduced spontaneous firing rate following
train stimuli, bottom track shows the stimulation artifact. Aq: Aqueduct; cp:
Cerebral peduncle; mlf: Medial longitudinal fasciculus; PAG: Periaqueductal
gray; RN: Red nucleus; SC: Superior colliculus; SN: Substantia nigra.
In the 1st group, the
intensity of 1.7 to 2.5 T evoked a stable repeated single discharge. The mean
latency of the discharge was 3.9±1.8ms (average±standard deviation). In the 2nd
group, two subtypes of units could be identified: subtype 1 (n=1), the
unit showed spontaneous firing rate increases immediately after a train
stimulation (2.5 T, 52 Hz, 200ms duration), then the firing is relaxed shortly
and enhanced again (Figure 2E and
Responses of the INC/ND Neurons to the MN
Stimulation A total of 12 extracellular single units were recorded in
the INC/DN area from the aforementioned 11 rats (Figure
Figure 3 Unitary discharges and recording
sites in the INC/DN A, D, G:
Recording sites (arrows) stained by sky-blue. B, C: Orthodromic unitary
discharges recorded from the INC shown in A. The downward arrowheads point the
starting of the stimulation. E: Train stimuli immediately increased firing rate
(pointing by head-down arrows) of a spontaneous discharging unit in the INC and
the recording site is shown in D. Upward arrowheads on the bottom indicate
starting of each train stimuli. F: Real time recording clipped from the
selected area in E (indicating by a thin arrow; from 10 000 to 12 000ms), an
upward arrowhead points to starting of the train stimuli. H: A spontaneous
discharging unit recorded in the DN was immediately or rapidly inhibited after
each train stimuli (pointing by upward arrowheads). Reduced firing rates are
marked by asterisks. The recording site is shown in G. I: A real time recording
extracted from the framed area in H (pointing by a thin arrow; from 8000 to 10
000ms), and artifact of the train stimuli (bottom line) is on the bottom. 3V:
Third ventricle; fr: Fasciculus retroflexus; pc: Posterior commissure.
In the group of spontaneous firing
units, there were also 2 subtypes, but the responding manner were different
from those recorded from the III group. Subtype 1 (n=5), units with low firing
rate that showed immediate increment of discharges following each train
stimulation, and declined quickly back to the initial level until the next
train (Figure 3E,
DISCUSSION
The present
work showed neurophysiological profile of a masticatory oculomotor neuronal
pathway that we found in recent decade[4,6-7], which is probably the residue of a primitive jaw-eye
cooperative reflex. Electric stimulation of the MN, the largest branch of the
trigeminal motor root in rats and cats, has been widely used to study the
central pathways of jaw muscle spindle Vme afferents in the rats and cats[10-11]. There are three principal
types of somatic nerve fibers in the MN. The majority of them are efferent
fibers of motoneurons in the trigeminal motor nucleus (Vmo)[12].
The second most common fibers are from primary jaw muscle spindle Vme afferents[10-11], those conduct muscle
proprioception. Trigeminal ganglion afferent fibers that transmit nociception
from deep jaw muscle and/or connective tissues are the smallest component[13]. Stimulation of the axons of trigeminal motoneurons
innervating masticatory muscles evokes antidromic impulses that are only
conducted to the Vmo[12-14].
And ganglionic afferents carrying jaw muscle nociception project restrictedly
to the dorsolateral rim of the spinal trigeminal nucleus and lamina I/II of
dorsal horn at upper cervical levels[13]. In
contrast, the Vme afferents project broadly, but unilaterally to the brainstem
through their central processes. Their terminations are found in a wide variety
of areas in the brainstem[10-11],
including ascending projections[10] to midbrain.
Thus, the MN stimuli evoked orthodromic unitary responses that were recorded in
the III and/or INC/DN in the present work are most likely elicited by inputs of
jaw muscle Vme afferents. Moreover, an interesting finding in our recent study
was that repeated and rapid down-stretch of the lower jaw induced bilateral
c-Fos expression in pre-oculomotor neurons located in INC/DN[6].
It is known that specific sensory stimulation can induce c-Fos expression in
relay neurons along a pertinent functional pathway[15].
Therefore, this observation substantiates the participation of jaw muscle
spindle Vme afferents in masticatory oculomotor pathway. It further suggests
that this pathway is somehow functional even under normal conditions.
Our recent
neuronal tract tracing study suggested that projections from the Vme neurons to
the III/IV and INC/DN are direct projections[4],
namely the monosynaptic innervations. In general, people use responsive latency
to judge whether a recorded unitary discharge is conducted via a
monosynaptic or through a multiple synaptic pathway[7,12,14,16]. However,
the latencies of unitary discharges after initial of stimulation is prominently
variated in the present work in units recorded from both III and INC/DN
(2.1-5.7ms and 1.8-7.6ms, respectively). Nonetheless, this kind of broad
variation of unitary discharge latencies following the MN stimulation was also
observed in our previous studies[16-17].
On one hand, it is possible that a masseter muscle afferent impulse prompted an
interneuron excitation firstly and then ensued an III motoneuron or INC/DN
pre-oculomotor neuron discharge. Dual synaptic pathway from jaw muscle
afferents to ipsilateral III through ipsilateral INC/DN premotor neurons can be
excluded because in the rat the INC/DN premotor neurons project only to the
contralateral III/IV[7]. On the other hand, if
monosynaptic innervation of jaw muscle afferents to the III and/or INC/DN
neurons are truly existent, some action potential transmitting may have been
delayed or vanished in the bifurcations of neuronal processes. It is because
all of intracellular recording followed by tracer injection studies on jaw
muscle spindle Vme afferents of cats and rats have revealed that Vme neuronal
processes are highly branched and the central axons commonly have 3-order
branches before finally reaching the targets[10-11,14]. Considering a linear
correlation between axon diameter and conduction speed[18-19], it is possible that action potential transmission is
decelerated when it prompts from thicker axons to thinner offspring branches.
Besides, the previous studies had displayed action potential transmission
failure at axon bifurcation when an investigated nerve was repeatedly
stimulated[20]. A single III motoneuron or INC
neuron may be innervated by more than one Vme neuronal axons based on previous
intracellular tract tracing studies[10-21].
If action potentials carried by these axons are partially vanished at their
bifurcation, fewer impulses would finally arrive to a targeted III or INC/DN
neuron. In this situation, the duration from postsynaptic membrane
depolarization to action potential initiation may be prolonged due to fewer
spatial or weaker temporal summation[22]. That
might be also a reason for fewer orthodromic unit discharge was recorded than
expected.
In the present work, we observed
inhibitory effects on both III and INC/DN in response to train stimulation of
the MN. However, based on established chemical neuroanatomy and neurophysiology
data[6,23], projections from
the Vme to the III and INC/DN should be primarily excitatory. INC/DN neurons
use both glutamate and GABA as neurotransmitters[24-25], and they can exert both excitatory and inhibitory
effects on the III neurons[26]. In addition,
bilateral INC neurons communicate each other through their projecting fibers
traveling in the posterior commissure[4,6-7]. These neuroanatomical frameworks established a
potential inhibitory circuits from the INC/DN to the III/IV and/or between the
INC of both sides. But in the rat, the INC/DN only innervate the contralateral
III[7]. Hence, in the present work, if any
inhibition onto the III is from the INC/DN, the circuit of this inhibitory
impulses may be from ipsilateral INC/DN that is activated by stimulation of the
MN to the contralateral one, and then back to the ipsilateral III. Whereas, in
front eyed mammals and humans, the pre-oculomotor INC neurons project to both
contra- and ipsilateral III[25]. The existence of
these neuronal circuits in humans had been suggested by a clinic phenomenon:
bilateral MGS[1-2,27].
Simultaneous bilateral levator palpebrae superioris (LPS) EMG recordings from a
patient suffering bilateral MGS revealed an explicit and complete inhibition of
LPS activity, whenever the contralateral LPS fired. This concomitant excitation
and inhibition persistent whenever the patient’s lower jaw kept moving
right-left horizontally[1-2,27].
Although the current
electrophysiological study adds new data for the masticatory oculomotor pathway
that we have been exploring, we still can’t exclude involvement of other
possible neuronal pathways. In recent years, a set of reports suggest that some
Vme neurons innervate mechanoreceptors in Müeller’s muscle in the superior
tarsal plate of human and rat[28-29].
In the rat, the central axons of these tarsal plate afferent Vme neurons
project to ipsilateral III, and terminate on motoneurons those innervate
slow-twitch skeletal muscles in the LPS[28-29]. Hence, it is possible that these tarsal plate
afferent Vme neurons may play a role in excitement of the LPS motoneuron in the
III. In addition, a neural tract tracing study in Macaque monkey has also
demonstrated a limited number of labeled Vme neurons following injection of
retrograde tracers into the LPS and superior rectus muscles[30].
Notably, the authors, who showed tarsal plate Vme afferents projecting to the
III, also discovered that tarsal plate Vme afferents connect with jaw muscle
spindle and/or periodontal Vme neurons through gap junctions, since they
observed spreading of gap-junction permeable dye from tarsal plate Vme
afferents into abutting Vme neurons[28].
Electrotonic coupling through gap-junction between the Vme neurons was reported
decades ago[31]. More importantly, somatofugal
action potentials have been recorded from the Vme neurons those coupling to the
other discharging Vme neurons, when their resting potential reaches the
threshold to initiate action potentials[31]. It
is likely that tarsal plate afferent Vme neurons may be co-fired through this
electrotonic coupling mechanism when jaw muscle afferent Vme neurons are
excited by either electric stimuli, or mechanic stimulation such as vigorously
stretching of the lower jaw[6].
Finally, it is notable that in a
considerable number of experimental rats in our current work, the single neuron
unit or the recorded unitary spontaneous discharges did not explicitly respond
to stimulation of the MN. Although it has been well known that the
physiological condition of each animal may be largely varied, the possibility
that this pathway is not well developed in some rats could not be excluded.
This situation is probably parallel to the finding in genetically healthy
humans that LPS retraction was evoked by stimulation of the ipsilateral
trigeminal motor root in about 1/3 of subjects[8].
Also, some reported cases manifest MGS only temporally in life[1-3]. These findings suggest the neural
circuit we explored may be better developed in some creatures, but not in all
of them.
ACKNOWLEDGEMENTS
We warmly thank Dr. James Keblesh
for his critical reading and English correction on this manuscript.
Foundation: Supported by Natural Sciences
Research Funding from Shaanxi Province (No.2009K01-74).
Conflicts of Interest: Qiao Y, None; Liang HC,
None; Zhang JD, None; Luo PF, None; Su AL, None; Zhang
T, None; Zhu HN, None.
REFERENCES