- Email: [email protected]
Pathway of the primary afferent nerve fibres serving proprioception in monkey extraocular muscles
Ophthal. F’hysiol. Opt. Vol. 17, No. 3, pp. 225-231, 1997 0 1997 The College of Optometrists. Published by Elsevier Science Ltd Printed in Great Britain 0275.5408/97 $17.00 + 0.00
PII: SO275-5408(96)00070-l
Pathway of the primary afferent nerve fibres serving proprioception in monkey extraocular muscles Alex Applied London
Gentle* Vision EClV
and Gordon Laboratory, 7DD, UK
Ruskell
Department
of Optometry
and Visual
Science,
City
University,
Summary Following intracranial section of either the oculomotor or ophthalmic nerve, Wallerian degeneration studies revealed 1.38-3.7% of nerve fibres in the nerves to the inferior and superior rectus muscles were of ophthalmic nerve origin; more than half of them were unmyelinated. The results of the two experiments were complementary. The proportion of fibres identified as sensory is substantially smaller than the lo%, estimated in other studies, required to serve muscle receptors. These results indicate, contrary to some reports, that a substantial majority of proprioceptive fibres are conducted from extraocular muscles to the brainstem in the motor nerves and that their somata are not housed in the trigeminal ganglion. 01997 The College of Optometrists. Published by Elsevier Science Ltd.
clarification as inter-species variation occurs. It appears that while complete transfer of proprioceptive fibres to Vl occurs in species such as sheep, pig and rat (Manni et al., 1971; Manni & Bortolami, 1979; Daunicht, 1983; Daunicht et al., 1985; Bortolami et al., 1987b; Kubota et al., 1988), there is some segregation of this pathway in animals such as the cat (Batini et al., 1975; Sherif et al., 1981; Bortolami et al., 1987b). Interspecies differences in the complement of EOM proprioceptive organs (Harker, 1972; Maier et al., 1974; Ruskell, 1979, 1989, 1990; Porter & Donaldson, 1991; Lukas et al., 1994), further suggests that caution should be exercised when attempting to draw conclusions on the primate pathway from subprimate data. It is possible that methodological differences may account for the disagreement in primate studies. The labelling patterns observed by Porter and co-workers are partly consistent with the limited nerve fibre exchange noted by Ruskell, but a tracer study does not provide quantitative data as only a proportion of sensory axons will take up the label. In contrast, a degeneration technique potentially affects all neurons, provided the post-operative interval is adequate (TerGivBinen & Huikuri, 1969; Heym & Forssman, 1981; Ruskell, 1983). Intermuscular variation of the pathway is unlikely to explain the discrepancy as Porter and coworkers included the IOM in their survey. Previous studies suggest that fibre exchange involves
Introduction
The pathway of first order afferent nerve fibres subserving extraocular muscles (EOM) has been well documented over the years, and while there is agreement between workers on the passage of these fibres in subprimates, there is apparent discord over this pathway in primates. Porter et al. (1983) commented that labelling patterns following retrograde axonic transport of intramuscular horseradish peroxidase (HRP), identified the ophthalmic division of the trigeminal ganglion as the only location of proprioceptive somata in monkeys. The implicated pathway involved either intraorbital or intracranial exchange of axons between the motor nerves and the ophthalmic nerve (Vl), which is consistent with that observed in subprimates. Ruskell (1983), using a degeneration technique to study the axon populations in the nerve to the inferior oblique muscle (IOM), also found evidence for the exchange of axons, but their frequency (1.8 %) was arguably inadequate to represent the fin11proprioceptive complement. Studies of the subprimate pathway do not afford any *Present address: Department of Optometry, University of Wales College of Cardiff, Cardiff CF 3Y.J, UK Received: 11 March 1996 Revised form: 30 August 1996
225
226
Ophthal.
Physiol.
Opt.
1997
17: No 3
both myelinated and unmyelinated nerve fibres, but those subserving proprioception are apparently myelinated and 2.5-6 pm in diameter (Ruskell, 1978). The present study aimed to resolve the argument by employing a degeneration technique to allow quantitative assessment of fibre exchange between the Vl and branches of the oculomotor nerve (N III) that supply EOM other than the IOM. Surplus experimental animal material from another study became available and, fortunately, the surgical interventions were perfectly suited to the present objectives. Some recent experimental evidence regarding the role of proprioception in visual development and function is based on the understanding that fibre projections are conducted to the brainstem in the ophthalmic nerve (Trotter et al., 1991, 1993; Lewis et al., 1994). Should this prove not to be so, such evidence would require review. Materials
and methods
One adult rhesus (Macaca mulatta) and three young cynomolgus (Mucucu fusciculuris) monkeys were prepared using the following surgical procedure. Animals were sedated parenterally using 2-3 mg/kg ketamine, then anaesthetised with 15-25 mg/kg Sagatal (pentobarbital sodium), administered either via a saphenous vein or intraperitoneally . The scalp was incised sagitally and frontally, then the temporalis muscle was detached medially on each side and reflected downwards. Removal of the calvaria intact was effected by drilling five holes around the skull and sawing between the holes, protecting the brain with a spatula insert. Temporal bone was removed, using nibblers, to a position opposite the cranial floor. The dura mater was cut and reflected to reveal the brain surface and the left temporal lobe was elevated to expose the cranial nerves at the brainstem. Two of the animals underwent left ophthalmic neurectomy, the ophthalmic nerve sheath being opened parallel to the nerve axis close to the trigeminal ganglion, 3-4 mm proximal to the oculomotor entry into the roof of the cavernous sinus. The nerve was separated from attachments to the sheath and adjacent structures with a hook, then a small piece cut and removed. The oculomotor nerve was severed close to the brainstem in the left side of the other two animals. The distal stump of the nerve retracted, sufficient to deter or delay regeneration, and removal of a segment of the nerve was not necessary. Finally the dura mater was sutured, the calvaria replaced and secured by strapping the temporalis muscles across the crown with strong sutures. The wound was closed and each animal allowed to recover normally. Following recovery, the animals were monitored for signs of discomfort or behavioural abnormality and sedated if necessary. Two of the animals were killed three days, and the others seven and eight days respectively, after the operation; sufficient interval for induction of degeneration.
They were sedated, anaesthetised and given an injection of 1500 units of the anticoagulent heparin sodium. The external jugular vein and inferior vena cava were cut and the animals killed and fixed by cardiac perfusion of warm saline followed by approximately 3 litres of a cacodylate-buffered solution of 2 % glutaraldehyde/3 % paraformaldehyde @H 7.4). The heads were stored in the fixative at approximately 4°C and later dissected to free the left and right inferior rectus muscles (IRM) from two animals and the left and right superior rectus muscles (SRM) from the other two, while immersed in buffered sucrose. The proximal third of each of the muscles was dissected, trimmed of unwanted tissue, taking care to preserve the associated nerve stumps. The pieces were rinsed, immersed in an unbuffered solution of 1% osmium tetroxide for one hour. They were washed, dehydrated in graded ethanols and cleared in xylene. The specimens were placed in a solution of equal parts of xylene and Araldite, transferred to Araldite and polymerised in an oven at 60°C for 48 hours. Interrupted semithin sections were cut from the proximal end of each muscle portion and stained for light microscopy with 1% toluidine blue in an equal volume of 2.5 % sodium carbonate, until a complete transverse section of the nerve was obtained. Ultrathin sections were then cut using a diamond knife, and mounted on unfilmed copper grids for electron microscopy. These were stained with a saturated solution of uranyl acetate in 70 % ethanol for approximately 20 minutes, washed and placed in 0.4% lead citrate in 0.1 M sodium hydroxide. Counts of normal and degenerated nerve fibres were made using an electron microscope and nerves from the right sides used as controls. A minimum of 20 % of the nerve fibre complement was examined in each specimen after light microscopic evaluation to ensure that the counted area was representative of the nerve. Photographs of light microscopic sections were used as a guide during electron microscopy. Axons displaying some uniform microtubule/ neurofilament matrix, containing intact mitochondria, were considered to be normal. The integrity of the myelin sheath was also assessed. Results Inspection of the whole surface of each of the removed muscles before processing revealed no nerve entrance other than the single large branch of N III, which entered the muscle on its ocular aspect approximately one third of its length from the origin together with the principle artery and vein. Light microscopic examination of the nerves showed a range of myelinated fibre diameters l-20 pm. Both experimental and control sides of each animal were searched to identify any localised areas of axon diameter variation or myelin disruption. Apart from a few aggregations of smaller
Nerve fibres of extraocular
Figure rectus
1. Part muscle.
of a nerve at its entrance to the MR88, control. Bar: 20pm.
superior
fibres, which were included for subsequent electron microscopic (EM) inspection, no signs of fibre grouping were observed. Controls
The control sides displayed little evidence of disruption with light microscopy (Figure I). Occasional small cellular structures, unassociated with myelin, were noted between the myelinated fibres. These sometimes displayed a nucleus and were identified with EM examination as unmyelinated nerve fibres . EM survey revealed the myelinated fibres to have a uniformly distributed axoplasmic matrix consisting of neurofilaments and microtubules (Figure 2). Most axons contained frequent mitochondria, and Schmidt-Lantermann incisures and Schwann cell nuclei in the myelin sheaths were frequent. Unmyelinated nerve fibres were common and single bundles contained up to ten axons (Figure 2). A tiny fraction of nerve fibres failed to meet the strictest criteria for normality. Some 0.37-2.58% of fibres from all control preparations displayed a flocculent matrix with no distinguishable axoplasmic organelles. Of the myelinated fibres that showed these characteristics (0.18-0.35%), most had an irregular or fragmented myelin sheath, and there were signs of the Schwann cell membrane becoming separated from the myelin structure. These anomalies were probably mainly preparation artefacts and perhaps an expression of normal neuron aging, but they did not meet the criteria for normality adopted in this study and for counting purposes they were considered as degenerated. Despite efforts to sample nerves before they branched and dispersed within muscles, finger-like branches were observed. As they entered the muscle, axonal profiles tended. to be oblique in cross-section. Assessment of the
muscles:
A. Gentle and G. Ruskell
227
Figure 2. Electron micrographs of the nerve of Figure 1 to show the structure of nerve fibres in detail. Myelinated fibres show mild disruption of the myelin sheath attributed to preparation artefact; the split in the myelin of one of them is the commencement of a Schmidt-Lantermann incisure. The axoplasm is composed mainly of neurofilaments and small mitochondria. Two unmyelinated nerve fibre bundles are present (arrows). Bar: 2km.
condition of these nerve fibres was still found to be possible and examples were included in the survey. Discrete groups of small myelinated nerve fibres were encountered, separate from the main nerve and its branches, lying deeper in the muscle. These occurred both distal and proximal to the nerve entrance. They consisted of as many as 10 fibres and were usually sectioned transversely. These probably represent motor innervation of the multiply innervated Felderstruktur fibres, which are known to be present throughout the muscles, and they were therefore not included in the counts. The occasional unmyelinated nerve fibre bundle was found in the main nerve with vesicles in addition to mitochondria in their axoplasm. Oculomotor
neurotomy
Light microscopic (LM) examination of these preparations revealed widespread disruption of myelin sheaths and little or no staining of the axoplasm (Figure 3). Many large irregular cell profiles, with or without myelin fragments, were scattered throughout the nerve, and there was a substantial amount of extracellular space. EM examination of the main nerve revealed major changes in the axoplasm within a majority of myelinated fibres which met the criteria for degeneration (Table 1). The regular grainy matrix of microtubules, neurofilaments and mitochondria, characteristic of control side axoplasm, was replaced by clumps of a flocculent substance of variable density (Figure 4). Short arcuate or circular strands of darker staining material within the flocculent mass probably
228
Ophthal.
Physiol.
Opt.
1997
17: No 3
Figure 3. Part of a nerve at its entrance to the superior rectus muscle following oculomotor neurotomy. Many myelinated fibres appear to have lost their axoplasm, other myelin profiles are distorted with elimination of axoplasm. Wide spacing between nerve fibres contrasts with the tight packing seen in Figure 7. ML88. Bar: 1 Opm.
indicated degenerated mitochondria. There was widespread evidence of macrophage incursion and these cells and Schwann cells displayed phagocytosis. Both contained myelin fragments within cytoplasmic vesicles. Many, often large isolated Schwann cell processes were present, some of them containing myelin fragments. This abundant evidence of Schwann cell proliferation was associated with greater spacing between myelin sheaths. Degenerative signs in the animal that had an g-day post-operative period were advanced and caused so much disruption that accurate assessment of the numbers of degenerated nerve fibres, particularly unmyelinated fibres, became difficult. Consequently, the degenerated fibre counts for the experimental side of this animal were less reliable.
Figure 4. Electron micrograph of part of the nerve of Figure 3. The nerve fibres display various signs of degeneration, and none contains normal axoplasm. Several examples of Schwann cell proliferation and phagocytosis (arrows) are present. Bar: 2,um.
An occasional nerve fibre was found to have normal axoplasm (Figure 5). The example shown is of a medium size fibre but generally the myelinated fibres among them were small. Such nerve fibres represented 3.23-3.7% of
Table 1. Oculomotor nerve branch fibre counts* following ophthalmic (Vl) or oculomotor (III) nerve section Animal reference and nerve lesion
Myelinated degen./total
ML56 (F, Rh.8) III SRM
176211774
ML88 (M, Cyn.3) III SRM
1797/l
822
Unmyelinated degen./total 37185 100/148
Combined degen./total 179911859 1897/I
970
ML84 (M, Cyn.3) VI IRM
2212399
13/196
35t2595
ML86 (F, Cyn.7) VI IRM
2911981
27156
5612037
Bracketed *Estimated
items: from
gender, species, postoperative period. the proportion of the total examined.
Figure 5. Electron micrograph of a medium-size myelinated nerve fibre of normal appearance with a neighbouring degenerated nerve fibre following oculomotor neurotomy. ML56. Bar: 1 pm.
Nerve
fibres
Table 2. Proportion of fibres potentially originating in VI (fibres of normal appearance after Nlll section and degenerated after Vl section) Animal
reference
M56 M88 M84 M86
% m yelinated + unmyelinated 3.23 3.70 1.38 2.75
(2.58) (1.19) (0.59) (0.37)
% myelinated only 0.65 1.27 0.87 1.42
(0.26) (0.35) (0.29) (0.18)
Bracketed figures indicate % of control side nerve not meeting the criteria adopted for normality.
fibres
the total fibre population surveyed ( Table 2). Most of them were unmyelinated and only 0.65-l .27 % of myelinated fibres were undegenerated. Myelinated nerve fibre groups isolated within the muscle from the main nerve, and referred to earlier, showed gross disturbance of myelin detectable with LM. Ophthalmic
neurectomy
Gross examination of the texture and general appearance of the nerve at its entrance revealed no differences from the control side in both preparations. LM revealed sparse, isolated disruption of myelin within the nerve, often associated with a lack of axoplasmic staining. The affected nerve fibres appeared to be distributed randomly and their positions were noted on photographs preliminary to EM inspection. The normal appearance of the majority of nerve fibres was confirmed with EM but a few small myelinated fibres (l-3 pm) showed axoplasmic changes identical to those observed in the specimens following oculomotor neurotomy ( Table 1) Disrupted myelin was occasionally encountered in association with proliferation of Schwann cell cytoplasm, but there was little evidence of phagocytic activity. 1.38-2.75 % of the fibre population counted showed degenerative changes, myelinated fibres accounting for 0.87-1.42% (Table 2). Most unmyelinated nerve fibres had a normal appearance, but those showing changes were more frequent than in the myelinated population. Changes were not found in the small isolated groups of nerve fibres buried within muscle and probably innervating Felderstruktur muscle fibres. Discussion The presence of degenerated fibres in oculomotor nerve branches following intracranial section of the ophthalmic nerve is clear evidence that fibres from the oculomotor nerve joined the ophthalmic distal to the lesion. Similarly, the presence of nerve fibres of normal appearance in the oculomotor nerve branches following intracranial section of the oculomotor nerve is the expected corollary. These degeneration patterns are consistent with the findings of
of extraocular
muscles:
A. Gentle
and G. Ruskell
229
previous work on primates (Porter et al., 1983; Ruskell, 1983; Porter, 1986). Attempts to identify nerve fibre transfer by dissection of the primate ocular sensorimotor system have failed to reveal any significant transfer in the territory between the EOM and the orbital apex, whereas they are readily observed in some subprimate mammals such as the sheep (Ruskell, unpublished observations). This leaves the likely area of exchange within the cavernous sinus. Evidence for a functional proprioceptive system in monkey EOM is limited, but may be inferred from the results of previous studies. As monkey EOM contain no muscle spindles (Maier et al., 1974) and few Golgi tendon organs (Ruskell, 1979), it is reasonable to conclude the structure most likely to provide EOM stretch information to the brainstem is the myotendinous cylinder (Ruskell, 1978), in common with cats (Alvarado-Mallart & Pingon-Raymond, 1979). These structures were shown to be served by predominantly myelinated nerve fibres 2.5-6 pm in diameter. Electrophysiological evidence of a functional system supplying EOM stretch information to the brainstem in mammals has all been obtained from subprimates (Batini & HorcholleBossavit, 1974; Batini, 1979; Trotter et al., 1990, 1991, 1993). However, neuroanatomical studies leave little doubt that a similar system occurs in primates, with the putative proprioceptive fibres joining Vl and terminating in the central trigeminal complex (Porter, 1986), as HRP tracers injected into EOM produced labelling in the trigeminal ganglion, cuneate nucleus and ventrolateral areas of the spinal trigeminal tract. The N III fibres found in the present study leaving the nerves serving the vertical recti and crossing to Vl , are consistent with the results of an earlier study of the inferior oblique muscle (Ruskell, 1983), and qualitatively consistent with the results of the HRP tracer study. The present study addresses the question of whether or not Vl might conduct any or all proprioceptive information to the brainstem. Rather fewer than 350 innervated myotendinous cylinders were estimated to be present in the distal myotendinous region of each vertical rectus muscle of the rhesus monkey (Ruskell, 1978). Estimates of the fibre content of N III branches were 1859-2595, and it follows that proprioceptive fibres must account for at least 10% of these, assuming no convergence of the afferent nerve fibres occurs. The small numbers of fibres identified as transferring to Vl in this study (35-73), were they all proprioceptive, fall well short of this figure. The possibility that proprioceptive nerve fibres may or may not be myelinated cannot be ruled out, but the available evidence indicates that all myotendinous cylinders receive a myelinated fibre 2.5-6 pm in diameter (Ruskell, 1983). If unmyelinated fibres are excluded from the number of fibres transferred to Vl, the number of potential proprioceptive nerve fibres is substantially reduced (12-29). Moreover, most of these fail to meet the diameter criterion for myotendinous cylinders-the majority were l-3 pm in diameter.
230
Ophthal.
Physiol.
Opt. 1997 17: No 3
If, as appears likely, few or none of the fibres transferring to Vl are proprioceptive, the disparate conclusion of other workers needs to be addressed, particularly the observation that intramuscular HRP injections lead to labelling of trigeminal ganglion cells in monkeys (Porter et al., 1993; Porter, 1986). There is anatomical and indirect electrophysiological evidence for the existence of proprioceptive somata in the mesencephalic nucleus of the cat (Sherif et al., 198 1; Bortolami et al., 1987b), which are thought to receive an input from separate proprioceptive pathways via the motor and sensory nerves. Although this arrangement has been questioned (Porter & Donaldson, 1991) a similar organization could exist in primates with a large imbalance in favour of the motor pathway. A split pathway of this kind is conceivably consistent with the report of visual deficits following section of Vl (Lewis et al., 1994). Alternatively, the transferred fibres of this study and the HRP labeled ganglion cells found by Porter et al. (1983) could have a role other than proprioception. Bortolami and coworkers described EOM afferents, monitored in the trigeminal ganglion and brainstem in both primates and subprimates, responsive to pain and temperature (Bortolami et al., 1977, 1987a, 1987b, 1991; Manni et al., 1989). These fibres were small, both myelinated and unmyelinated, as were the transferred fibres found in this study. Another fraction of the intact unmyelinated fibres in N III following the lesions may be vascular afferents with somata in the trigeminal ganglion. Although this comment is speculative, as such fibres have not been identified within the extraocular muscles, their presence on the walls of arteries in the cavernous sinus in monkeys (Simons & Ruskell, 1988), and in the conjunctiva and superior tarsal muscle of rats (Luhtala et al., 1991), suggests that a sensory vascular distribution may be general. From 6.4-31% of unmyelinated nerve fibres within the nerve to the inferior oblique muscle in monkeys were estimated to be sympathetic, originating in the superior cervical ganglion (Ruskell, 1983). In the present study isolated nerve bundles, including preterminal axons containing vesicles, were identified as probably sympathetic and their most likely function vasomotor. Most sympathetic fibres serving the orbit issue from the retro-orbital plexus (Ruskell, 1970), remote from the position of the N III lesion and arguably unaffected by it, and consequently the surviving unmyelinated nerve fibres cannot all be identified as sensory. On balance, the evidence of this study indicates that the EOM nerve fibres transferring to V 1 are too few to meet the requirements of proprioception, and some or perhaps all of those that transfer serve functions other than proprioception.
References
Alvarado-Mallart, R. M. and Pincon-Raymond, M. (1979) The palisade endings of cat extraocular muscles, a light and electron microscopic study. Tissue Cell 11, 567-584. Batini, C. (1979) Properties of the receptors of the extraocular muscles. Progress in Bruin Res. 50, 301-314. Batini, C., Buisseret, P. and Buisseret-Delmas,C. (1975) Trigeminal pathway of the extrinsic eye muscle afferents in the cat. Bruin Res. 85, 74-78. Batini, C. and Horcholle-Bossavit, G. (1974) Extraocular muscle afferents and visual input interactions in the superior colliculus of the cat. Bruin Res. 50, 335-344. Bortolami, R., Veggetti, A., Callegari, E., Lucci, M. L. and Palmieri, G. (1977) Afferent fibers and sensory ganglion cells within the oculomotor nerve in some mammals and man. Arch. Ital. Biol. 115, 355-365. Bortolami, R., Lucchi, M. L., Callegari, E., Calza, L., Pettorossi, V. E. and Manni, E. (1987a) Analogies existing between the primary trigeminal afferent fibers running within the oculomotor nerve and the ventral root primary afferent fibers. Adv. Pain Res. Tnerap. 10, 59-63. Bortolami, R., Lucchi, M. L., Callegari, E., Pettorossi, V. E. and Manni, E. (1987b) Localization and somatotopyof sensory cells innervating the extraocular muscles of lamb, pig and cat. Histochemical and electrophysiological investigation. Arch. Ital. Biol. 125, 1-15. Bortolami, R., Galza, L., Lucchi, M. L., Giardino, L., Callegare, E., Manni, E., Pettorossi, V. E., Barazzon, A. M. and Costerbo, G. L. (1991) Peripheral territory and neuropeptidesof the trigeminal ganglion neurons centrally projecting through the oculomotor nerve demonstratedby fluorescent retrograde double-labelling combined with inununocytochemistry. Bruin Res. 547, 82-88. Daunicht, W. J. (1983) Proprioception in extraocular muscles of the rat. Brain Res. 278, 291-294. Daunicht, W. J., Jaworski, E. and E&miller, R. (1985) Afferent innervation of extraocular muscles in the rat studied by retrograde and anterogradehorseradish peroxidase transport. Neurosci. L&t. 56, 143-148. Harker, D. W. (1972) The structure and innervation of sheep superior rectus and levator palpebrae extraocular muscles. II Muscle spindles. Invest. Ophthalmol. 11, 970-979. Heym, Ch. and Forssman, W.-G. (1981) Techniques in Neuroanatomical Research. Springer, Berlin. Kubota, K., Matsuyarna, S., Kubota, M., Narita, N., Nagae, K., Hosaka, K., Lee, M.-S., Chang, C.-M., Yeh, Y.-C., Ohkubo, K. and Shibanai, S. (1988) Localization of proprioceptive neurons innervating the muscle spindle of pig extraocular muscles studied by horseradish peroxidase. Anat. Anz. 166, 117-132. Lewis, R. F., Zee, D. S., Gaymard, B. M. and Guthrie, B. L. (1994) Extraocular muscle proprioception functions in the control of ocular alignment and eye movement conjugacy. J. Neurophysiol. 72, 1028-1031. Luhtala, J., Palkama, A. and Uusitalo, H. (1991) Calcitonin gene related peptide immunoreactive nerve fibers in the rat conjunctiva. Invest. Ophthalmol. Vis. Sci. 32, 640-645. Lukas, J. R., Aigner, M., Blumer, R., Heinzl, H. and Mayr, R. (1994) Number and distribution of neuromuscular spindles in human extraocular muscles. Invest. Ophthalmol. Vis. Sci., 35, 4317-4327.
Acknowledgements
Sandip Doshi and Paul Dyer of City University provided technical assistance.
Maier, A., deSantis, M. and Eldred, E. (1974) The occurrence of muscle spindles in the extraocular muscles of various vertebrates. J. Morph. 143, 397-408. Manni, E. and Bortolami, R. (1979) Peripheral and central
Nerve
fibres
organization of the extraocular muscle proprioception in the ungulata. Prog. Bruin Rex 50, 291-299. Manni, E., Palmieri, G. and Marini, R. (1971) Peripheral pathway of the proprioceptive afferents from the lateral rectus muscle of the eye. Exp. Neurol. 30, 46-53. Manni, E., Draicchi, F., Pettorossi, V. E., Carobi, C., Grass, S., Bortolami, R. and Lucchi, M. L. (1989) On the nature of the afferent fibers of the oculomotor nerve. Arch. Ital. Biol. 127, 99-108. Porter, J. D. (1986) Brainstem terminations of extraocular muscle primary afferent neurons in the monkey. J. Comp. Neurol. 247, 133-143. Porter, J. D. and Donaldson, I. M. L. (1991) The anatomical substrate for cat extraocular muscle proprioception. Neuroscience, 43, 473-48 1. Porter, J. D., Guthrie, B. L. and Sparks, D. L. (1983) Innervation of monkey extraocular muscles: Localization of sensory and motor neurons by retrograde transport of horseradish peroxidase. J. Comp. Neurol. 218, 208-219. Ruskell, G. L. (1970) The orbital branches of the pterygopalatine ganglion and their relationship with internal carotid nerve branches in primates. J. Anat. 106, 323-339. Ruskell, G. L. (1978) The fine structure of innervated myotendinous cylinders in extraocular muscles of rhesus monkeys. J. Neurocytol. 7, 693-708. Ruskell, G. L. (1979) The incidence and variety of Golgi tendon organs in extraocular muscles of the rhesus monkey. J. Neurocytol. 8, 639-653. Ruskell, G. L. (1983) Fibre analysis of the nerve to the inferior oblique muscle in monkeys. J. Anat. 137, 445-455.
of extraocular
muscles:
A. Gentle
and G. Ruskell
231
Ruskell, G. L. (1989) The fine structure of human extraocular muscle spindles and their potential proprioceptive capacity. J. Anat. 167, 199-214. Ruskell, G. L. (1990) Golgi tendon organs in the proximal tendon of sheep extraocular muscles. Amt. Rec. 227, 25-31. Sherif, M. F., Papadopoulos, N. J. and Albert, E. N. (1981) Fiber contribution from the mesencephalic nucleus of the trigeminal nerve to the trochlear nerve in the cat. A histological quantitative study. Anat. Rec. 201, 669678.
Simons, T. and Ruskell, G. L. (1988) Distribution and termination of trigeminal nerves to the cerebral arteries in monkeys. J. Anut. 159, 57-71. Ter&%inen, H. and Huikuri, K. (1969) Effect of oculomotor and trigeminal nerve section on the ultrastructure of different myoneural junctions in rat extraocular muscles. Z. Zellforsch. mikrosk. Anut. 102, 466-482. Trotter, Y., Beaux, J. C., Pouget, A. and Imbert, M. (1990) Neuronal stereoscopic processing following extraocular proprioception deafferentation. Neuroreport, 1, 187-190. Trotter, Y., Beaux, J. C., Pouget, A. and Imbert, M. (1991) Temporal limits of the susceptibility of depth perception to proprioceptive deafferentation of extraocular muscles. Dev. Bruin Res. 59, 23-29. Trotter, Y., Celebrins, S., Beaux, J. C., Grandjea, B. and Imbert, M. (1993) Long term disfunctions of neural stereoscopic mechanisms after unilateral extraocular muscle proprioceptive deafferentation. J. Neurophysiol. 69, 15131529.
PII: SO275-5408(96)00070-l
Pathway of the primary afferent nerve fibres serving proprioception in monkey extraocular muscles Alex Applied London
Gentle* Vision EClV
and Gordon Laboratory, 7DD, UK
Ruskell
Department
of Optometry
and Visual
Science,
City
University,
Summary Following intracranial section of either the oculomotor or ophthalmic nerve, Wallerian degeneration studies revealed 1.38-3.7% of nerve fibres in the nerves to the inferior and superior rectus muscles were of ophthalmic nerve origin; more than half of them were unmyelinated. The results of the two experiments were complementary. The proportion of fibres identified as sensory is substantially smaller than the lo%, estimated in other studies, required to serve muscle receptors. These results indicate, contrary to some reports, that a substantial majority of proprioceptive fibres are conducted from extraocular muscles to the brainstem in the motor nerves and that their somata are not housed in the trigeminal ganglion. 01997 The College of Optometrists. Published by Elsevier Science Ltd.
clarification as inter-species variation occurs. It appears that while complete transfer of proprioceptive fibres to Vl occurs in species such as sheep, pig and rat (Manni et al., 1971; Manni & Bortolami, 1979; Daunicht, 1983; Daunicht et al., 1985; Bortolami et al., 1987b; Kubota et al., 1988), there is some segregation of this pathway in animals such as the cat (Batini et al., 1975; Sherif et al., 1981; Bortolami et al., 1987b). Interspecies differences in the complement of EOM proprioceptive organs (Harker, 1972; Maier et al., 1974; Ruskell, 1979, 1989, 1990; Porter & Donaldson, 1991; Lukas et al., 1994), further suggests that caution should be exercised when attempting to draw conclusions on the primate pathway from subprimate data. It is possible that methodological differences may account for the disagreement in primate studies. The labelling patterns observed by Porter and co-workers are partly consistent with the limited nerve fibre exchange noted by Ruskell, but a tracer study does not provide quantitative data as only a proportion of sensory axons will take up the label. In contrast, a degeneration technique potentially affects all neurons, provided the post-operative interval is adequate (TerGivBinen & Huikuri, 1969; Heym & Forssman, 1981; Ruskell, 1983). Intermuscular variation of the pathway is unlikely to explain the discrepancy as Porter and coworkers included the IOM in their survey. Previous studies suggest that fibre exchange involves
Introduction
The pathway of first order afferent nerve fibres subserving extraocular muscles (EOM) has been well documented over the years, and while there is agreement between workers on the passage of these fibres in subprimates, there is apparent discord over this pathway in primates. Porter et al. (1983) commented that labelling patterns following retrograde axonic transport of intramuscular horseradish peroxidase (HRP), identified the ophthalmic division of the trigeminal ganglion as the only location of proprioceptive somata in monkeys. The implicated pathway involved either intraorbital or intracranial exchange of axons between the motor nerves and the ophthalmic nerve (Vl), which is consistent with that observed in subprimates. Ruskell (1983), using a degeneration technique to study the axon populations in the nerve to the inferior oblique muscle (IOM), also found evidence for the exchange of axons, but their frequency (1.8 %) was arguably inadequate to represent the fin11proprioceptive complement. Studies of the subprimate pathway do not afford any *Present address: Department of Optometry, University of Wales College of Cardiff, Cardiff CF 3Y.J, UK Received: 11 March 1996 Revised form: 30 August 1996
225
226
Ophthal.
Physiol.
Opt.
1997
17: No 3
both myelinated and unmyelinated nerve fibres, but those subserving proprioception are apparently myelinated and 2.5-6 pm in diameter (Ruskell, 1978). The present study aimed to resolve the argument by employing a degeneration technique to allow quantitative assessment of fibre exchange between the Vl and branches of the oculomotor nerve (N III) that supply EOM other than the IOM. Surplus experimental animal material from another study became available and, fortunately, the surgical interventions were perfectly suited to the present objectives. Some recent experimental evidence regarding the role of proprioception in visual development and function is based on the understanding that fibre projections are conducted to the brainstem in the ophthalmic nerve (Trotter et al., 1991, 1993; Lewis et al., 1994). Should this prove not to be so, such evidence would require review. Materials
and methods
One adult rhesus (Macaca mulatta) and three young cynomolgus (Mucucu fusciculuris) monkeys were prepared using the following surgical procedure. Animals were sedated parenterally using 2-3 mg/kg ketamine, then anaesthetised with 15-25 mg/kg Sagatal (pentobarbital sodium), administered either via a saphenous vein or intraperitoneally . The scalp was incised sagitally and frontally, then the temporalis muscle was detached medially on each side and reflected downwards. Removal of the calvaria intact was effected by drilling five holes around the skull and sawing between the holes, protecting the brain with a spatula insert. Temporal bone was removed, using nibblers, to a position opposite the cranial floor. The dura mater was cut and reflected to reveal the brain surface and the left temporal lobe was elevated to expose the cranial nerves at the brainstem. Two of the animals underwent left ophthalmic neurectomy, the ophthalmic nerve sheath being opened parallel to the nerve axis close to the trigeminal ganglion, 3-4 mm proximal to the oculomotor entry into the roof of the cavernous sinus. The nerve was separated from attachments to the sheath and adjacent structures with a hook, then a small piece cut and removed. The oculomotor nerve was severed close to the brainstem in the left side of the other two animals. The distal stump of the nerve retracted, sufficient to deter or delay regeneration, and removal of a segment of the nerve was not necessary. Finally the dura mater was sutured, the calvaria replaced and secured by strapping the temporalis muscles across the crown with strong sutures. The wound was closed and each animal allowed to recover normally. Following recovery, the animals were monitored for signs of discomfort or behavioural abnormality and sedated if necessary. Two of the animals were killed three days, and the others seven and eight days respectively, after the operation; sufficient interval for induction of degeneration.
They were sedated, anaesthetised and given an injection of 1500 units of the anticoagulent heparin sodium. The external jugular vein and inferior vena cava were cut and the animals killed and fixed by cardiac perfusion of warm saline followed by approximately 3 litres of a cacodylate-buffered solution of 2 % glutaraldehyde/3 % paraformaldehyde @H 7.4). The heads were stored in the fixative at approximately 4°C and later dissected to free the left and right inferior rectus muscles (IRM) from two animals and the left and right superior rectus muscles (SRM) from the other two, while immersed in buffered sucrose. The proximal third of each of the muscles was dissected, trimmed of unwanted tissue, taking care to preserve the associated nerve stumps. The pieces were rinsed, immersed in an unbuffered solution of 1% osmium tetroxide for one hour. They were washed, dehydrated in graded ethanols and cleared in xylene. The specimens were placed in a solution of equal parts of xylene and Araldite, transferred to Araldite and polymerised in an oven at 60°C for 48 hours. Interrupted semithin sections were cut from the proximal end of each muscle portion and stained for light microscopy with 1% toluidine blue in an equal volume of 2.5 % sodium carbonate, until a complete transverse section of the nerve was obtained. Ultrathin sections were then cut using a diamond knife, and mounted on unfilmed copper grids for electron microscopy. These were stained with a saturated solution of uranyl acetate in 70 % ethanol for approximately 20 minutes, washed and placed in 0.4% lead citrate in 0.1 M sodium hydroxide. Counts of normal and degenerated nerve fibres were made using an electron microscope and nerves from the right sides used as controls. A minimum of 20 % of the nerve fibre complement was examined in each specimen after light microscopic evaluation to ensure that the counted area was representative of the nerve. Photographs of light microscopic sections were used as a guide during electron microscopy. Axons displaying some uniform microtubule/ neurofilament matrix, containing intact mitochondria, were considered to be normal. The integrity of the myelin sheath was also assessed. Results Inspection of the whole surface of each of the removed muscles before processing revealed no nerve entrance other than the single large branch of N III, which entered the muscle on its ocular aspect approximately one third of its length from the origin together with the principle artery and vein. Light microscopic examination of the nerves showed a range of myelinated fibre diameters l-20 pm. Both experimental and control sides of each animal were searched to identify any localised areas of axon diameter variation or myelin disruption. Apart from a few aggregations of smaller
Nerve fibres of extraocular
Figure rectus
1. Part muscle.
of a nerve at its entrance to the MR88, control. Bar: 20pm.
superior
fibres, which were included for subsequent electron microscopic (EM) inspection, no signs of fibre grouping were observed. Controls
The control sides displayed little evidence of disruption with light microscopy (Figure I). Occasional small cellular structures, unassociated with myelin, were noted between the myelinated fibres. These sometimes displayed a nucleus and were identified with EM examination as unmyelinated nerve fibres . EM survey revealed the myelinated fibres to have a uniformly distributed axoplasmic matrix consisting of neurofilaments and microtubules (Figure 2). Most axons contained frequent mitochondria, and Schmidt-Lantermann incisures and Schwann cell nuclei in the myelin sheaths were frequent. Unmyelinated nerve fibres were common and single bundles contained up to ten axons (Figure 2). A tiny fraction of nerve fibres failed to meet the strictest criteria for normality. Some 0.37-2.58% of fibres from all control preparations displayed a flocculent matrix with no distinguishable axoplasmic organelles. Of the myelinated fibres that showed these characteristics (0.18-0.35%), most had an irregular or fragmented myelin sheath, and there were signs of the Schwann cell membrane becoming separated from the myelin structure. These anomalies were probably mainly preparation artefacts and perhaps an expression of normal neuron aging, but they did not meet the criteria for normality adopted in this study and for counting purposes they were considered as degenerated. Despite efforts to sample nerves before they branched and dispersed within muscles, finger-like branches were observed. As they entered the muscle, axonal profiles tended. to be oblique in cross-section. Assessment of the
muscles:
A. Gentle and G. Ruskell
227
Figure 2. Electron micrographs of the nerve of Figure 1 to show the structure of nerve fibres in detail. Myelinated fibres show mild disruption of the myelin sheath attributed to preparation artefact; the split in the myelin of one of them is the commencement of a Schmidt-Lantermann incisure. The axoplasm is composed mainly of neurofilaments and small mitochondria. Two unmyelinated nerve fibre bundles are present (arrows). Bar: 2km.
condition of these nerve fibres was still found to be possible and examples were included in the survey. Discrete groups of small myelinated nerve fibres were encountered, separate from the main nerve and its branches, lying deeper in the muscle. These occurred both distal and proximal to the nerve entrance. They consisted of as many as 10 fibres and were usually sectioned transversely. These probably represent motor innervation of the multiply innervated Felderstruktur fibres, which are known to be present throughout the muscles, and they were therefore not included in the counts. The occasional unmyelinated nerve fibre bundle was found in the main nerve with vesicles in addition to mitochondria in their axoplasm. Oculomotor
neurotomy
Light microscopic (LM) examination of these preparations revealed widespread disruption of myelin sheaths and little or no staining of the axoplasm (Figure 3). Many large irregular cell profiles, with or without myelin fragments, were scattered throughout the nerve, and there was a substantial amount of extracellular space. EM examination of the main nerve revealed major changes in the axoplasm within a majority of myelinated fibres which met the criteria for degeneration (Table 1). The regular grainy matrix of microtubules, neurofilaments and mitochondria, characteristic of control side axoplasm, was replaced by clumps of a flocculent substance of variable density (Figure 4). Short arcuate or circular strands of darker staining material within the flocculent mass probably
228
Ophthal.
Physiol.
Opt.
1997
17: No 3
Figure 3. Part of a nerve at its entrance to the superior rectus muscle following oculomotor neurotomy. Many myelinated fibres appear to have lost their axoplasm, other myelin profiles are distorted with elimination of axoplasm. Wide spacing between nerve fibres contrasts with the tight packing seen in Figure 7. ML88. Bar: 1 Opm.
indicated degenerated mitochondria. There was widespread evidence of macrophage incursion and these cells and Schwann cells displayed phagocytosis. Both contained myelin fragments within cytoplasmic vesicles. Many, often large isolated Schwann cell processes were present, some of them containing myelin fragments. This abundant evidence of Schwann cell proliferation was associated with greater spacing between myelin sheaths. Degenerative signs in the animal that had an g-day post-operative period were advanced and caused so much disruption that accurate assessment of the numbers of degenerated nerve fibres, particularly unmyelinated fibres, became difficult. Consequently, the degenerated fibre counts for the experimental side of this animal were less reliable.
Figure 4. Electron micrograph of part of the nerve of Figure 3. The nerve fibres display various signs of degeneration, and none contains normal axoplasm. Several examples of Schwann cell proliferation and phagocytosis (arrows) are present. Bar: 2,um.
An occasional nerve fibre was found to have normal axoplasm (Figure 5). The example shown is of a medium size fibre but generally the myelinated fibres among them were small. Such nerve fibres represented 3.23-3.7% of
Table 1. Oculomotor nerve branch fibre counts* following ophthalmic (Vl) or oculomotor (III) nerve section Animal reference and nerve lesion
Myelinated degen./total
ML56 (F, Rh.8) III SRM
176211774
ML88 (M, Cyn.3) III SRM
1797/l
822
Unmyelinated degen./total 37185 100/148
Combined degen./total 179911859 1897/I
970
ML84 (M, Cyn.3) VI IRM
2212399
13/196
35t2595
ML86 (F, Cyn.7) VI IRM
2911981
27156
5612037
Bracketed *Estimated
items: from
gender, species, postoperative period. the proportion of the total examined.
Figure 5. Electron micrograph of a medium-size myelinated nerve fibre of normal appearance with a neighbouring degenerated nerve fibre following oculomotor neurotomy. ML56. Bar: 1 pm.
Nerve
fibres
Table 2. Proportion of fibres potentially originating in VI (fibres of normal appearance after Nlll section and degenerated after Vl section) Animal
reference
M56 M88 M84 M86
% m yelinated + unmyelinated 3.23 3.70 1.38 2.75
(2.58) (1.19) (0.59) (0.37)
% myelinated only 0.65 1.27 0.87 1.42
(0.26) (0.35) (0.29) (0.18)
Bracketed figures indicate % of control side nerve not meeting the criteria adopted for normality.
fibres
the total fibre population surveyed ( Table 2). Most of them were unmyelinated and only 0.65-l .27 % of myelinated fibres were undegenerated. Myelinated nerve fibre groups isolated within the muscle from the main nerve, and referred to earlier, showed gross disturbance of myelin detectable with LM. Ophthalmic
neurectomy
Gross examination of the texture and general appearance of the nerve at its entrance revealed no differences from the control side in both preparations. LM revealed sparse, isolated disruption of myelin within the nerve, often associated with a lack of axoplasmic staining. The affected nerve fibres appeared to be distributed randomly and their positions were noted on photographs preliminary to EM inspection. The normal appearance of the majority of nerve fibres was confirmed with EM but a few small myelinated fibres (l-3 pm) showed axoplasmic changes identical to those observed in the specimens following oculomotor neurotomy ( Table 1) Disrupted myelin was occasionally encountered in association with proliferation of Schwann cell cytoplasm, but there was little evidence of phagocytic activity. 1.38-2.75 % of the fibre population counted showed degenerative changes, myelinated fibres accounting for 0.87-1.42% (Table 2). Most unmyelinated nerve fibres had a normal appearance, but those showing changes were more frequent than in the myelinated population. Changes were not found in the small isolated groups of nerve fibres buried within muscle and probably innervating Felderstruktur muscle fibres. Discussion The presence of degenerated fibres in oculomotor nerve branches following intracranial section of the ophthalmic nerve is clear evidence that fibres from the oculomotor nerve joined the ophthalmic distal to the lesion. Similarly, the presence of nerve fibres of normal appearance in the oculomotor nerve branches following intracranial section of the oculomotor nerve is the expected corollary. These degeneration patterns are consistent with the findings of
of extraocular
muscles:
A. Gentle
and G. Ruskell
229
previous work on primates (Porter et al., 1983; Ruskell, 1983; Porter, 1986). Attempts to identify nerve fibre transfer by dissection of the primate ocular sensorimotor system have failed to reveal any significant transfer in the territory between the EOM and the orbital apex, whereas they are readily observed in some subprimate mammals such as the sheep (Ruskell, unpublished observations). This leaves the likely area of exchange within the cavernous sinus. Evidence for a functional proprioceptive system in monkey EOM is limited, but may be inferred from the results of previous studies. As monkey EOM contain no muscle spindles (Maier et al., 1974) and few Golgi tendon organs (Ruskell, 1979), it is reasonable to conclude the structure most likely to provide EOM stretch information to the brainstem is the myotendinous cylinder (Ruskell, 1978), in common with cats (Alvarado-Mallart & Pingon-Raymond, 1979). These structures were shown to be served by predominantly myelinated nerve fibres 2.5-6 pm in diameter. Electrophysiological evidence of a functional system supplying EOM stretch information to the brainstem in mammals has all been obtained from subprimates (Batini & HorcholleBossavit, 1974; Batini, 1979; Trotter et al., 1990, 1991, 1993). However, neuroanatomical studies leave little doubt that a similar system occurs in primates, with the putative proprioceptive fibres joining Vl and terminating in the central trigeminal complex (Porter, 1986), as HRP tracers injected into EOM produced labelling in the trigeminal ganglion, cuneate nucleus and ventrolateral areas of the spinal trigeminal tract. The N III fibres found in the present study leaving the nerves serving the vertical recti and crossing to Vl , are consistent with the results of an earlier study of the inferior oblique muscle (Ruskell, 1983), and qualitatively consistent with the results of the HRP tracer study. The present study addresses the question of whether or not Vl might conduct any or all proprioceptive information to the brainstem. Rather fewer than 350 innervated myotendinous cylinders were estimated to be present in the distal myotendinous region of each vertical rectus muscle of the rhesus monkey (Ruskell, 1978). Estimates of the fibre content of N III branches were 1859-2595, and it follows that proprioceptive fibres must account for at least 10% of these, assuming no convergence of the afferent nerve fibres occurs. The small numbers of fibres identified as transferring to Vl in this study (35-73), were they all proprioceptive, fall well short of this figure. The possibility that proprioceptive nerve fibres may or may not be myelinated cannot be ruled out, but the available evidence indicates that all myotendinous cylinders receive a myelinated fibre 2.5-6 pm in diameter (Ruskell, 1983). If unmyelinated fibres are excluded from the number of fibres transferred to Vl, the number of potential proprioceptive nerve fibres is substantially reduced (12-29). Moreover, most of these fail to meet the diameter criterion for myotendinous cylinders-the majority were l-3 pm in diameter.
230
Ophthal.
Physiol.
Opt. 1997 17: No 3
If, as appears likely, few or none of the fibres transferring to Vl are proprioceptive, the disparate conclusion of other workers needs to be addressed, particularly the observation that intramuscular HRP injections lead to labelling of trigeminal ganglion cells in monkeys (Porter et al., 1993; Porter, 1986). There is anatomical and indirect electrophysiological evidence for the existence of proprioceptive somata in the mesencephalic nucleus of the cat (Sherif et al., 198 1; Bortolami et al., 1987b), which are thought to receive an input from separate proprioceptive pathways via the motor and sensory nerves. Although this arrangement has been questioned (Porter & Donaldson, 1991) a similar organization could exist in primates with a large imbalance in favour of the motor pathway. A split pathway of this kind is conceivably consistent with the report of visual deficits following section of Vl (Lewis et al., 1994). Alternatively, the transferred fibres of this study and the HRP labeled ganglion cells found by Porter et al. (1983) could have a role other than proprioception. Bortolami and coworkers described EOM afferents, monitored in the trigeminal ganglion and brainstem in both primates and subprimates, responsive to pain and temperature (Bortolami et al., 1977, 1987a, 1987b, 1991; Manni et al., 1989). These fibres were small, both myelinated and unmyelinated, as were the transferred fibres found in this study. Another fraction of the intact unmyelinated fibres in N III following the lesions may be vascular afferents with somata in the trigeminal ganglion. Although this comment is speculative, as such fibres have not been identified within the extraocular muscles, their presence on the walls of arteries in the cavernous sinus in monkeys (Simons & Ruskell, 1988), and in the conjunctiva and superior tarsal muscle of rats (Luhtala et al., 1991), suggests that a sensory vascular distribution may be general. From 6.4-31% of unmyelinated nerve fibres within the nerve to the inferior oblique muscle in monkeys were estimated to be sympathetic, originating in the superior cervical ganglion (Ruskell, 1983). In the present study isolated nerve bundles, including preterminal axons containing vesicles, were identified as probably sympathetic and their most likely function vasomotor. Most sympathetic fibres serving the orbit issue from the retro-orbital plexus (Ruskell, 1970), remote from the position of the N III lesion and arguably unaffected by it, and consequently the surviving unmyelinated nerve fibres cannot all be identified as sensory. On balance, the evidence of this study indicates that the EOM nerve fibres transferring to V 1 are too few to meet the requirements of proprioception, and some or perhaps all of those that transfer serve functions other than proprioception.
References
Alvarado-Mallart, R. M. and Pincon-Raymond, M. (1979) The palisade endings of cat extraocular muscles, a light and electron microscopic study. Tissue Cell 11, 567-584. Batini, C. (1979) Properties of the receptors of the extraocular muscles. Progress in Bruin Res. 50, 301-314. Batini, C., Buisseret, P. and Buisseret-Delmas,C. (1975) Trigeminal pathway of the extrinsic eye muscle afferents in the cat. Bruin Res. 85, 74-78. Batini, C. and Horcholle-Bossavit, G. (1974) Extraocular muscle afferents and visual input interactions in the superior colliculus of the cat. Bruin Res. 50, 335-344. Bortolami, R., Veggetti, A., Callegari, E., Lucci, M. L. and Palmieri, G. (1977) Afferent fibers and sensory ganglion cells within the oculomotor nerve in some mammals and man. Arch. Ital. Biol. 115, 355-365. Bortolami, R., Lucchi, M. L., Callegari, E., Calza, L., Pettorossi, V. E. and Manni, E. (1987a) Analogies existing between the primary trigeminal afferent fibers running within the oculomotor nerve and the ventral root primary afferent fibers. Adv. Pain Res. Tnerap. 10, 59-63. Bortolami, R., Lucchi, M. L., Callegari, E., Pettorossi, V. E. and Manni, E. (1987b) Localization and somatotopyof sensory cells innervating the extraocular muscles of lamb, pig and cat. Histochemical and electrophysiological investigation. Arch. Ital. Biol. 125, 1-15. Bortolami, R., Galza, L., Lucchi, M. L., Giardino, L., Callegare, E., Manni, E., Pettorossi, V. E., Barazzon, A. M. and Costerbo, G. L. (1991) Peripheral territory and neuropeptidesof the trigeminal ganglion neurons centrally projecting through the oculomotor nerve demonstratedby fluorescent retrograde double-labelling combined with inununocytochemistry. Bruin Res. 547, 82-88. Daunicht, W. J. (1983) Proprioception in extraocular muscles of the rat. Brain Res. 278, 291-294. Daunicht, W. J., Jaworski, E. and E&miller, R. (1985) Afferent innervation of extraocular muscles in the rat studied by retrograde and anterogradehorseradish peroxidase transport. Neurosci. L&t. 56, 143-148. Harker, D. W. (1972) The structure and innervation of sheep superior rectus and levator palpebrae extraocular muscles. II Muscle spindles. Invest. Ophthalmol. 11, 970-979. Heym, Ch. and Forssman, W.-G. (1981) Techniques in Neuroanatomical Research. Springer, Berlin. Kubota, K., Matsuyarna, S., Kubota, M., Narita, N., Nagae, K., Hosaka, K., Lee, M.-S., Chang, C.-M., Yeh, Y.-C., Ohkubo, K. and Shibanai, S. (1988) Localization of proprioceptive neurons innervating the muscle spindle of pig extraocular muscles studied by horseradish peroxidase. Anat. Anz. 166, 117-132. Lewis, R. F., Zee, D. S., Gaymard, B. M. and Guthrie, B. L. (1994) Extraocular muscle proprioception functions in the control of ocular alignment and eye movement conjugacy. J. Neurophysiol. 72, 1028-1031. Luhtala, J., Palkama, A. and Uusitalo, H. (1991) Calcitonin gene related peptide immunoreactive nerve fibers in the rat conjunctiva. Invest. Ophthalmol. Vis. Sci. 32, 640-645. Lukas, J. R., Aigner, M., Blumer, R., Heinzl, H. and Mayr, R. (1994) Number and distribution of neuromuscular spindles in human extraocular muscles. Invest. Ophthalmol. Vis. Sci., 35, 4317-4327.
Acknowledgements
Sandip Doshi and Paul Dyer of City University provided technical assistance.
Maier, A., deSantis, M. and Eldred, E. (1974) The occurrence of muscle spindles in the extraocular muscles of various vertebrates. J. Morph. 143, 397-408. Manni, E. and Bortolami, R. (1979) Peripheral and central
Nerve
fibres
organization of the extraocular muscle proprioception in the ungulata. Prog. Bruin Rex 50, 291-299. Manni, E., Palmieri, G. and Marini, R. (1971) Peripheral pathway of the proprioceptive afferents from the lateral rectus muscle of the eye. Exp. Neurol. 30, 46-53. Manni, E., Draicchi, F., Pettorossi, V. E., Carobi, C., Grass, S., Bortolami, R. and Lucchi, M. L. (1989) On the nature of the afferent fibers of the oculomotor nerve. Arch. Ital. Biol. 127, 99-108. Porter, J. D. (1986) Brainstem terminations of extraocular muscle primary afferent neurons in the monkey. J. Comp. Neurol. 247, 133-143. Porter, J. D. and Donaldson, I. M. L. (1991) The anatomical substrate for cat extraocular muscle proprioception. Neuroscience, 43, 473-48 1. Porter, J. D., Guthrie, B. L. and Sparks, D. L. (1983) Innervation of monkey extraocular muscles: Localization of sensory and motor neurons by retrograde transport of horseradish peroxidase. J. Comp. Neurol. 218, 208-219. Ruskell, G. L. (1970) The orbital branches of the pterygopalatine ganglion and their relationship with internal carotid nerve branches in primates. J. Anat. 106, 323-339. Ruskell, G. L. (1978) The fine structure of innervated myotendinous cylinders in extraocular muscles of rhesus monkeys. J. Neurocytol. 7, 693-708. Ruskell, G. L. (1979) The incidence and variety of Golgi tendon organs in extraocular muscles of the rhesus monkey. J. Neurocytol. 8, 639-653. Ruskell, G. L. (1983) Fibre analysis of the nerve to the inferior oblique muscle in monkeys. J. Anat. 137, 445-455.
of extraocular
muscles:
A. Gentle
and G. Ruskell
231
Ruskell, G. L. (1989) The fine structure of human extraocular muscle spindles and their potential proprioceptive capacity. J. Anat. 167, 199-214. Ruskell, G. L. (1990) Golgi tendon organs in the proximal tendon of sheep extraocular muscles. Amt. Rec. 227, 25-31. Sherif, M. F., Papadopoulos, N. J. and Albert, E. N. (1981) Fiber contribution from the mesencephalic nucleus of the trigeminal nerve to the trochlear nerve in the cat. A histological quantitative study. Anat. Rec. 201, 669678.
Simons, T. and Ruskell, G. L. (1988) Distribution and termination of trigeminal nerves to the cerebral arteries in monkeys. J. Anut. 159, 57-71. Ter&%inen, H. and Huikuri, K. (1969) Effect of oculomotor and trigeminal nerve section on the ultrastructure of different myoneural junctions in rat extraocular muscles. Z. Zellforsch. mikrosk. Anut. 102, 466-482. Trotter, Y., Beaux, J. C., Pouget, A. and Imbert, M. (1990) Neuronal stereoscopic processing following extraocular proprioception deafferentation. Neuroreport, 1, 187-190. Trotter, Y., Beaux, J. C., Pouget, A. and Imbert, M. (1991) Temporal limits of the susceptibility of depth perception to proprioceptive deafferentation of extraocular muscles. Dev. Bruin Res. 59, 23-29. Trotter, Y., Celebrins, S., Beaux, J. C., Grandjea, B. and Imbert, M. (1993) Long term disfunctions of neural stereoscopic mechanisms after unilateral extraocular muscle proprioceptive deafferentation. J. Neurophysiol. 69, 15131529.