Palisade Endings in Extraocular Muscles of the Monkey are Immunoreactive for Choline Acetyltransferase and Vesicular Acetylcholine Transporter

Abstract

purpose. To analyze palisade endings in extraocular muscles (EOMs) of a primate species and to examine our previous findings in cat that palisade endings are putative effector organs.

methods. Eleven monkeys (Macaca fascicularis) of both sexes, between 4 and 6 years of age were analyzed. Whole EOM myotendons were immunostained with four combinations of triple-fluorescent labeling and examined by confocal laser scanning microscopy. Labeling included antibodies against choline acetyltransferase (ChAT), vesicular acetylcholine transporter (VAChT), neurofilament, and synaptophysin. Muscle fibers were counterstained with phalloidin.

results. Palisade endings were observed in all monkey EOMs. Nerve fibers extended from the muscle into the tendon and looped back to divide into a terminal arborization (palisade ending) around a single muscle fiber tip. In approximately 30% of the cases, nerve fibers supplying palisade endings often established motor terminals outside the palisade complex. Nerve fibers forming palisade endings were ChAT-neurofilament positive. Axonal branches of palisade endings were ChAT-neurofilament positive as well. All palisade nerve terminals exhibited ChAT-synaptophysin immunoreactivity. Within the palisade complex, palisade nerve terminals exhibited VAChT immunoreactivity. All palisade nerve terminals were VAChT-synaptophysin immunoreactive.

conclusions. The results confirm that in the monkey, palisade endings contain acetylcholine and are therefore most likely effector organs. Palisade endings are also present in human EOMs and because of their location at the myotendinous junction, these organs are of crucial interest for strabismus surgery.

The role played by proprioceptive innervation of extraocular muscles (EOMs) is one of the least understood in the neuronal control of eye movements. There has been a long-lasting debate about whether afferent signals from proprioceptors (inflow theory) or a copy of the central motor command (outflow theory) provide information about the eye position that is necessary for the interpretation of visual information. ^{1 2} In typical mammalian skeletal muscles, proprioceptive information derives from muscle spindles and Golgi tendon organs. EOMs in mammals represent an exception to this organization plan, as muscle spindles and Golgi tendon organs are absent in most species. ³  

Dogiel ⁴ was one of the first scientists who described palisade endings (innervated myotendinous cylinders) in the EOMs of several mammals and humans. The observations of Dogiel were confirmed later in monkeys, ⁵ cats, ^{6 7 8} and humans. ^{9 10} Palisade endings were also detected and confirmed in rabbit EOMs ^{11 12} and have been described recently in EOMs of sheep, ¹³ and rats. ¹⁴  

Palisade endings are encapsulated nerve end organs unique to EOMs of mammals. They are located at the myotendinous junction and consist of a dense ramification of preterminal axons and their vesicle-laden nerve terminals around a single muscle fiber tip. Most palisade nerve terminals have established contacts to the tendon and only a few of them have contacts to the muscle fiber surface. Palisade endings in rabbits and rats are an exception, since exclusively neuromuscular contacts have been observed. ^{11 14} Palisade endings arise from nerve fibers that, coming from the muscle, extend into the tendon and then turn back 180° to terminate around a single muscle fiber tip. Each palisade ending is associated with a multiply innervated muscle fiber that has several motor contacts along its length. 

Although physiological evidence is missing, there is a consensus that palisade endings are sensory organs, ^{1 2 5 6 7 14 15} and Spencer and Porter ¹⁶ have concluded that palisade endings potentially represent the principal proprioceptor type for mammalian EOMs. Nevertheless, in a few publications, a motor role ^{11 17} or a sensory–motor role ⁹ of this EOM specific organ is favored. 

In our latest study, ⁸ we analyzed palisade endings in cat EOMs by using antibodies against choline acetyltransferase (ChAT). The antibodies against ChAT label cholinergic nerve fibers in the central and peripheral nervous system. ^{18 19 20} We showed that the nerve fibers supplying palisade endings and the palisade complexes themselves were ChAT positive. In a few palisade endings, we observed neuromuscular contacts that were α-bungarotoxin positive as well. Morphologically, we detected several times that the same nerve fiber supplying palisade endings established motor contacts outside the palisade complex. In summary, the results of our previous study provided evidence that palisade endings exhibit molecular and morphologic characteristics of effector organs. ⁸  

It was the purpose of the present study to examine our surprising finding that palisade endings are putative effector organs in a primate species that is evolutionary closely related to humans. In addition to antibody against ChAT we used antibody against vesicular acetylcholine transporter (VAChT) to analyze palisade endings. VAChT is a vesicle membrane protein and is responsible for the uptake of acetylcholine into synaptic vesicles. VAChT is used to identify cholinergic nerve terminals in the central and peripheral nervous system. ^{18 19}  

Materials and Methods

All animals used in this study were treated in accordance with the ARVO statement for the Use of Animals in Ophthalmic and Vision Research. 

Eleven monkeys (Macaca fascicularis), eight females and three males aged between 4 and 6 years and weighing between 2.4 and 3.8 kg, were analyzed in the study. The eyeballs, including the distal parts of the EOMs were obtained from Novartis Pharma AG (Basel, Switzerland). The distal parts of the four rectus EOMs and two oblique EOMs, including the tendons, were removed. Tissue was immersion fixed with 4% paraformaldehyde in 0.1 M phosphate buffer (PB; pH 7.4) for 24 hours and then rinsed in 0.1 M phosphate-buffered saline (PBS; pH 7.4). The EOMs were further processed for confocal laser scanning microscopy. 

Confocal Laser Scanning Microscopy

A total of 132 EOMs were analyzed. Distal EOM myotendons were divided into four groups, each group containing 33 EOMs, both rectus and oblique. Different combinations of triple fluorescent labeling were applied to the whole specimens. These wholemounts were labeled with: (1) phalloidin and antibodies against ChAT and neurofilament; (2) Phalloidin and antibodies against ChAT and synaptophysin; (3) phalloidin and antibodies against VAChT and neurofilament; and (4) phalloidin and antibodies against VAChT and synaptophysin. 

Procedure for Triple-Fluorescent Labeling

A detailed description of the staining protocol of EOM wholemounts is given by Konakci et al. ⁸ Herein, we provide a condensed version. The marker for muscle fibers, the primary antibodies, and secondary antibodies, and the working dilutions and the sources, are listed in Table 1 . 

After fixation, specimens were immersed for 1 hour in prechilled acetone at 4°C, rinsed in PBS with 0.05% Tween, and subsequently transferred into a blocking solution for 1 hour at room temperature. The blocking solution always corresponded to the host(s) of the second antibodies. The preparations were incubated in a mixture of primary antibodies for 48 hours at 15°C and then were incubated in secondary antibodies that were applied successively, each for 4 hours at room temperature. Finally, specimens were labeled for 30 minutes with phalloidin. Between the incubation steps, specimens were extensively rinsed in PBS containing 0.05% Tween. After the staining procedure, specimens were mounted in agar (Citifluor, Stansted, UK). 

Wholemounts were examined under a confocal laser scanning microscope (CLSM; model LSM 410; Carl Zeiss Meditec, Oberkochen, Germany). When palisade endings were identified, series of longitudinal virtual CLSM sections of 0.4- to 1-μm thickness were cut through the specimens. Each section was photodocumented, and three-dimensional projections were formulated on computer (LSM Image Examiner software; Carl Zeiss Meditec). 

Control Experiments

In negative control experiments, the primary antibodies were omitted and the secondary antibodies were used alone. In all cases, the omission of the primary antibodies resulted in a complete lack of immunostaining. 

Positive control experiments were performed on cryostat sections of monkey EOMs. Sections were double labeled with anti-ChAT and α-bungarotoxin or alternatively with anti-VAChT and α-bungarotoxin. Muscle fibers were counterstained with phalloidin. In both experiments α-bungarotoxin-positive motor endplates were also positive for ChAT and VAChT. 

Results

Motor Nerve Endings

Motor endplates in monkey EOMs were positive for α-bungarotoxin and immunoreactive for ChAT and VAChT as well (Fig. 1) . The nerve fibers supplying motor terminals were ChAT immunoreactive (Fig. 1A) . 

Morphology of Palisade Endings

In accordance with Ruskell ⁵ we confirmed the presence of palisade endings in monkey EOMs. We observed palisade endings at the distal myotendinous junctions in both the rectus and the oblique EOMs. 

Palisade endings were supplied by thin nerve fibers (2–3 μm in diameter) which came from the muscle and extended into the tendon. Within the tendon, the nerve fibers curved back to divide into a terminal arborization around single muscle fiber tips (Figs. 2 3 4) . In several cases, the nerve fibers reached the palisade complex at the muscle fiber tip more directly without previously turning back from the tendon (Fig. 3C) . 

We traced nerve fibers supplying palisade endings toward the muscle belly. In many cases, the nerve fiber supplying a palisade ending intermingled with others at the myotendinous junction and could not be traced any farther. In approximately 30% of the cases, we observed that the nerve fiber supplying a palisade ending established motor contacts outside the palisade complex. In two images (Figs. 2A 3C) , nerve fibers, before forming a palisade ending, established at least one motor terminal on the muscle fiber associated with the palisade complex. 

Palisade endings consisted of axonal branches that formed nerve terminals that lay within the tendon and around the muscle fiber surface (Figs. 2 3 4 5) . Other palisade nerve terminals lay between the muscle fiber processes that attached the muscle fiber to the tendon (Fig. 3A , inset). In some palisade endings, we observed nerve terminals exclusively in the tendon compartment (Figs. 2B 3B 5B) . 

Molecular Characteristics of Palisade Endings

We used the following antibodies to characterize palisade endings in EOMs: anti-ChAT, anti-VAChT, anti-neurofilament, and anti-synaptophysin. Anti-ChAT and anti-VAChT are markers for cholinergic nerve fibers and cholinergic nerve terminals, respectively. Anti-neurofilament is a general marker for nerve fibers, and anti-synaptophysin is a general marker for nerve terminals. In addition, we used phalloidin to counterstain muscle fibers. Four different combinations of triple labeling were performed in EOM wholemounts (Table 1) . 

Labeling with Phalloidin and Antibodies against ChAT and Neurofilament

All nerve fibers forming palisade endings were positive for neurofilament as well as for ChAT (Fig. 2) . We never found a palisade complex that originated from a neurofilament-positive but ChAT-negative nerve fiber. We also did not detect palisade endings that had dual innervation—namely, by a nerve fiber’s being positive for ChAT-neurofilament and by another nerve fiber’s being only neurofilament positive. 

The axonal branches of the palisade endings were also neurofilament-ChAT immunoreactive. Axonal branches of palisade endings, being only positive for neurofilament were never observed (Fig. 2) . 

Labeling with Phalloidin and Antibodies against ChAT and Synaptophysin

We observed ChAT-positive nerve fibers that split into palisade endings at the myotendinous junction. The axonal branches of the palisade endings were also ChAT immunoreactive. All palisade nerve terminals were immunoreactive for synaptophysin and for ChAT (Fig. 3) . Nerve terminals that stained solely synaptophysin positive were never observed. 

Labeling with Phalloidin and Antibodies against VAChT and Neurofilament

Neurofilament-positive nerve fibers coming from the muscle penetrated the tendon and then looped back to supply palisade endings on muscle fiber tips. The axonal branches of the palisade endings established nerve terminals that exhibited VAChT immunoreactivity (Fig. 4) . 

Labeling with Phalloidin and Antibodies against VAChT and Synaptophysin

By using this combination of triple labeling, we stained the muscle fibers and the palisade nerve terminals. The nerve fibers supplying the palisade endings and the axonal branches of the palisade endings were not labeled. The region of thepalisade endings, the EOM tendon junction, was easy to identify because there the phalloidin-labeled muscle fibers were attached to the unstained tendon. All palisade nerve terminals exhibited immunoreactivity for synaptophysin and VAChT (Fig. 5) . We detected no nerve terminals that were only synaptophysin positive. 

Discussion

Although there are conflicting reports on the functional properties of palisade endings, it is generally assumed that these EOM-specific organs are sensory and monitor EOM function. ^{1 2 5 6 7 14 15 16} Recently, we have shown in cats that palisade endings exhibit molecular characteristics that suggest an effector function. ⁸ It was the purpose of the present study to examine this surprising finding in palisade endings of a primate species, the monkey, by using antibodies against ChAT and as a novel antibody against VAChT, two markers for cholinergic nerve fibers and cholinergic nerve terminals. 

Consistent with our morphologic findings in cat, ⁸ we observed in monkey that nerve fibers supplying palisade endings often establish motor terminals outside the palisade complex. In monkey, approximately 30% of the palisade endings exhibit such an additional motor innervation from the same nerve fiber. As in the cat, ⁸ we showed in the monkey that palisade endings arise from nerve fibers that contain ChAT. Axonal branches of palisade endings and palisade nerve terminals contain ChAT as well. Moreover, in this study we demonstrated for the first time that palisade nerve terminals contain the VAChT protein. The VAChT immunoreactivity of palisade nerve terminals correlates well with morphologic findings in palisade endings. In particular, fine structural investigations of palisade endings in monkey, ⁵ cat, ^{6 8} sheep, ¹³ and humans ⁹ demonstrate that palisade nerve terminals, both at the tendon and at the muscle fiber, exhibit dense aggregations of clear vesicles. VAChT is a vesicle membrane protein and anti-VAChT is a marker for synaptic vesicles. ^{18 19} The VAChT immunoreactivity of palisade nerve terminals strongly suggests that the VAChT protein is a membrane constituent of the vesicles in palisade nerve terminals. 

This is the first report that demonstrated a ChAT-VAChT immunoreactivity of palisade endings. ChAT is the synthesizing enzyme for the neurotransmitter acetylcholine and VAChT is a necessary requisite to transport acetylcholine into synaptic vesicles of axonal terminals. Thus, the present study in monkey definitively proves that palisade endings synthesize acetylcholine, which is stored in the vesicles of palisade nerve terminals. Because acetylcholine is the classic neurotransmitter of nerve terminals that in other locations are universally motor, we conclude that palisade endings may act as effectors organs. We confirmed the putative effector role of palisade endings in monkey, a primate species that is evolutionary close to humans. It is therefore tempting to speculate that palisade endings in human EOMs may also fulfill an effector function. 

Recent studies have shown that multiple neurotransmitters (glutamate ²¹ ) and neuromodulatory agents (CGRP ²² and substance P ²³ ) are found in nerve terminals. It would be a challenge to conduct future studies to examine whether palisade endings contain other neurotransmitters–neuromodulatory agents beside acetylcholine. 

Until now, physiological studies on palisade endings are lacking. The major arguments to classify palisade endings as proprioceptors came from fine structural investigations ^{5 6 13} and a neuronal tracing experiment. ⁷ In fine structural studies of rhesus monkey, ⁵ cat, ⁶ and sheep, ¹³ palisade nerve terminals contacting the collagen fibrils as target structures were arguably considered to be sensory. Palisade nerve terminals contacting the muscle fiber lack a basal lamina in the synaptic cleft that is analogous to sensory neuromuscular contacts in muscle spindles. After injecting neuronal tracers into the trigeminal ganglion that is supposed to harbor the perikarya of the nerve fibers innervating EOM proprioceptors, Billig et al. ⁷ described three kinds of labeled nerve endings in cat EOMs—one type resembling palisade endings. 

Morphologically, we observed in the monkey that nerve fibers supplying palisade endings establish motor terminals outside the palisade complex. Immunohistochemically, we confirmed that palisade endings are cholinergic structures. In particular, labeling monkey EOMs with a cholinergic marker (anti-ChAT) and a general marker for nerve fibers (anti-neurofilament) demonstrated that palisade endings are exclusively supplied with cholinergic nerve fibers. All axonal branches of palisade endings are cholinergic as well. Noncholinergic nerve fibers were never found to form palisade endings. Further, a dual innervation of the palisade complex with cholinergic and noncholinergic nerve fibers was also never observed. In the same line, labeling EOMs with antibody against ChAT and a general marker for nerve terminals (anti-synaptophysin) showed that all nerve terminals of the palisade complexes are cholinergic. Labeling monkey EOMs with another cholinergic marker (anti-VAChT) confirmed the cholinergic phenotype of palisade nerve terminals. Immunostaining with antibodies against VAChT-synaptophysin revealed that all palisade nerve terminals are cholinergic. In summary, morphologic findings and immunoreactivity for cholinergic markers strongly favor an effector role of palisade endings. 

Recently, Tata et al. ²⁴ detected that small and medium-sized sensory neurons in the dorsal root ganglion of rats are VAChT immunoreactive. Based on these findings, one might speculate that palisade endings are cholinergic sensory structures. Terminals on the muscle fiber from the same nerve fiber supplying a palisade ending would have to be interpreted as either sensory as well, or the nerve fiber would have to serve a dual (sensory and motor) cholinergic function. In the present study, we provided evidence that palisade endings exhibit ChAT–VAChT immunoreactivity which is identical with classic motor terminals. Further, we showed that palisade endings arise from nerve fibers establishing motor terminals outside the palisade complex. We conclude that these findings are strong arguments in support of our assumption that palisade endings are effectors. 

Supporting evidence that palisade endings are putative effector organs comes from degeneration experiments ¹⁷ and α-bungarotoxin labeling. ^{8 11} After stereotactic lesions of the abducens, trochlear, and occulomotor nuclei in cats, Sas and Scháb ¹⁷ found that besides motor nerve terminals, all palisade endings were degenerated in the EOMs. In cat, palisade nerve terminals contacting the muscle fiber were observed in few palisade endings. ⁸ Such neuromuscular contacts stained positively for α-bungarotoxin. ⁸ In palisade endings of rabbit EOMs, neuromuscular contacts had a basal lamina in the synaptic cleft which is typical of motor terminals. ¹¹ Binding of α-bungarotoxin to neuromuscular contacts confirmed the motor role. ¹¹ It is important, however, to note that palisade endings in rabbit EOMs are different, since neurotendinous contacts that are a regular feature in palisade endings of all other species, including humans, are absent. 

In humans, the hypothesis was put forward that palisade endings may combine motor and sensory functions. Lukas et al. ⁹ found sensorylike neurotendinous contacts and motor neuromuscular contacts in palisade endings of human EOMs. Motor neuromuscular contacts were endowed with a basal lamina in the synaptic cleft and bound α-bungarotoxin. In the cat ⁸ and now confirmed in the monkey, we showed that exclusively cholinergic nerve fibers form palisade endings that are endowed exclusively with cholinergic nerve terminals. These findings suggest a motor innervation of this EOM-specific organ. 

Functional Considerations

The functional significance of palisade endings with a putative effector role is difficult to predict. In palisade endings, most nerve terminals are at the tendon and are in intimate contact with the surrounding collagen bundles. It is unclear what effect a release of neurotransmitter (acetylcholine) might have on the neighboring collagen. Otherwise, neurotendinous contacts are at a great distance from the muscle fiber, and after acetylcholine release there is a long distance for the neurotransmitter to diffuse to a cholinergic receptor site on the muscle fiber surface. Nerve terminals being far away from their target would mean that palisade endings belong to a motor system that is different from the classic motor system in skeletal muscle, because there, motor terminals are always in close contact with the muscle fiber, and the neurotransmitter release site and receptor site reside close together. 

The importance of afferent signals from EOM proprioceptors is demonstrated in several studies. In cats, it was shown that EOM proprioception plays a role in orienting behavior and in depth perception. ^{25 26} In humans, several studies support the assumption that EOM proprioceptors provide information about the eye position in the orbit. ^{27 28 29 30} In particular, in behavioral experiments, it has been shown that subjects in whom the afferent input from the EOMs has been either disrupted or altered, fail to localize targets in the surrounding space exactly. Such deficits were interpreted as a sign that subjects receive insufficient afferent signals about the eye position, which in turn led to misinterpretation in the spatial localization of visual targets. ^{27 28 29 30}  

The endowment with classic proprioceptors (Golgi tendon organs and muscle spindles) in mammalian EOMs shows striking interspecies variations. Muscle spindles and Golgi tendon organs are numerous in even toed ungulates, ^{31 32 33 34 35} but only a few muscle spindles and one to two Golgi tendon organs were observed in the EOMs of rhesus monkey. ^{1 36} Human EOMs contain muscle spindles, whereas Golgi tendon organs are reported to be absent. ^{1 37 38 39} In EOMs of the cat, rabbit, guinea pig, and rat, neither muscle spindles nor Golgi tendon organs are found. ³ Particularly in those EOMs in which muscle spindles and/or Golgi tendon organs are few or absent, palisade endings are considered to be an alternative sensory end organ. ¹⁵ In this study, we confirmed in a primate species that palisade endings are putative effector organs and therefore these organs would not provide information about the eye position. Our findings that palisade endings may be effectors are not only of scientific interest but also have clinical relevance. Palisade endings are also present in human EOMs and are located at the myotendinous junction, the very target area of strabismus surgery. 

 Supported by Grant P15478 from the Fonds zur Foerderung der Wissenschaftlichen Forschung.

 Submitted for publication June 9, 2005; revised August 2, 2005; accepted September 28, 2005.

 Disclosure: K.Z. Konakci, None; J. Streicher, None; W. Hoetzenecker, None; I. Haberl, None; M.J.F. Blumer, None; G. Wieczorek, Novartis Pharma AG, Basel (E, F); J.G. Meingassner, Novartis Pharma AG, Vienna (E, F); S.L. Paal, None; D. Holzinger, None; J.-R. Lukas, None; R. Blumer, None

 The publication costs of this article were defrayed in part by page charge payment. This article must therefore be marked “advertisement” in accordance with 18 U.S.C. §1734 solely to indicate this fact.

 

Table 1.

 

List of the Marker for Muscle Fibers, Primary Antibodies, and Secondary Antibodies Used in the Study in Different Combinations of Triple Staining

Triple Staining	Marker for Muscle Fibers, Working Dilution, Source	Primary Antibodies, Working Dilutions, Sources	Secondary Antibodies, Working Dilutions, Sources
Phalloidin, anti-choline acetyltransferase (ChAT), anti-neurofilament	Alexa Fluor 633 conjugated phalloidin, 1:80; Molecular Probes, Eugene, OR	Goat anti ChAT, 1:100; Chemicon, Temecula, CA	Donkey anti-goat Alexa Fluor 488, 1:500; Molecular Probes
				Mouse anti-neurofilament, 1:100; Chemicon	Donkey anti-mouse rhodamine, 1:100; Chemicon
Phalloidin, anti ChAT, anti-synaptphysin	Alexa fluor 633 conjugated phalloidin	Goat anti ChAT, 1:100; Chemicon	Rabbit anti-goat Alexa Fluor 488, 1:500; Molecular Probes
		Mouse anti-synaptophysin, 1:200; Chemicon	Goat anti-mouse rhodamine, 1:200; Chemicon
Phalloidin, anti-vesicular acetylcholine transporter (VAChT), anti-neurofilament	Alexa fluor 633 conjugated phalloidin	Goat anti-VAChT, 1:100; Santa Cruz Biotechnology, Santa Cruz, CA	Donkey anti-goat Alexa Fluor 488, 1:500; Molecular Probes
				Donkey anti-mouse rhodamine, 1:100; Chemicon
			Mouse anti-neurofilament, 1:100; Chemicon
Phalloidin, anti-VAChT, anti-synapthophysin	Alexa fluor 633 conjugated phalloidin	Goat anti-VAChT, 1:100; Santa Cruz	Rabbit anti-goat Alexa Fluor 488, 1:500; Molecular Probes
		Mouse anti-synaptophysin, 1:200; Chemicon	Goat anti-mouse rhodamine, 1:200; Chemicon

Figure 1.

 

CLSM image of motor terminals labeled with α-bungarotoxin (red) and alternatively with antibodies against ChAT (green) and VAChT (green). Muscle fibers were counterstained with phalloidin (white). (A) A motor terminal double positive for α-bungarotoxin and ChAT. The nerve fiber (N) supplying the motor terminals was also ChAT positive. (B) Motor terminal double positive for α-bungarotoxin and VAChT. Scale bar, 100 μm.

Figure 2.

 

CLSM image of palisade endings. Nerve fibers were immunostained with anti-neurofilament (red) and with anti-ChAT (green). Muscle fibers were labeled with phalloidin (white). The tendon attached to the muscle fiber was not stained. (A) A nerve fiber running beside the muscle extended into the tendon and turned back to form a palisade ending around the muscle fiber tip. Before, forming the palisade complex, the nerve fiber established a motor terminal (arrowhead) on the muscle fiber. The nerve fiber and axonal branches of the palisade ending stained positively for neurofilament and ChAT. (B) Two neurofilament- and ChAT-positive nerve fibers coming from the tendon formed a palisade ending around the muscle fiber tip. Likewise, axonal branches of the palisade ending were neurofilament-ChAT immunoreactive. Scale bar, 100 μm.

Figure 3.

 

CLSM image of palisade endings. Nerve fibers were labeled with anti-ChAT (green), nerve terminals with anti-synaptophysin (red), and muscle fibers with phalloidin (white). The tendon was not stained. (A–C) ChAT-positive nerve fibers forming palisade endings around muscle fiber tips. In the palisade complexes, axonal branches were positive for ChAT, whereas nerve terminals stained positively for ChAT and synaptophysin. (A, inset) Detail of a palisade ending with ChAT-positive axonal branches and ChAT- and synaptophysin-positive nerve terminals between the muscle fiber processes that attach the muscle fiber to the tendon. (A) The nerve fiber supplying a palisade ending runs alongside the muscle fiber and loops back in the tendon. (B) Nerve fibers coming from the tendon and supplying a palisade ending. (C) A nerve fiber running alongside the muscle fiber branching immediately at the muscle fiber tip to form a palisade ending. Outside the palisade ending, the nerve fiber established a motor contact (arrowhead) on the muscle fiber. Scale bar, 100 μm.

Figure 4.

 

CLSM image of palisade endings. Nerve fibers are immunostained with anti-neurofilament (red), nerve terminals with anti-VAChT (green), and muscle fibers with phalloidin (white). The tendon is not stained. (A, B) Neurofilament-positive nerve fibers coming from the muscle extended into the tendon and turned back to supply palisade endings at the muscle fiber tips. The axonal branches of the palisade ending were neurofilament positive, palisade nerve terminals exhibited VAChT immunoreactivity. Scale bar, 100 μm.

Figure 5.

 

CLSM image of palisade nerve terminals. Palisade nerve terminals were immunostained with anti-synaptophysin (red) and anti-VAChT (green). The nerve fibers supplying the palisade endings and axonal branches of the palisade endings were not labeled. Muscle fibers were labeled with phalloidin (white). The tendon was not stained. (A, B) At the muscle tendon junction, the phalloidin-labeled muscle fibers were attached to the unstained tendon. (A) Palisade nerve terminals were in the tendon and around the muscle fiber tip. All palisade nerve terminals exhibited both synaptophysin and VAChT immunoreactivity. (B) Showing synaptophysin- and VAChT-positive palisade nerve terminals in the tendon. Scale bars, 100 μm.

The authors thank Christiane Krivanek and Marietta Lipowec for valuable technical assistance.