Peripheral projections of a subpopulation of dorsal root ganglion neurons defined by ovalbumin immunoreactivity

Summary Previous studies from this laboratory have used antisera to aldehyde-conjugated ovalbumin to localize ovalbumin-like immunoreactivity within a subpopulation of sensory neurons. We have now combined retrograde tracing and immunohistochemical procedures to identify the tissues innervated by sensory neurons which are either immunoreactive or non-immunoreactive for ovalbumin. The fluorescent tracer Di-I was administered to feather follicles, flexor ulnar muscle, subdermis, expansor secundariorum, heart and liver and identified seven days later within corresponding dorsal root ganglia. Most neurons innervating the follicles had large cell somata, and fewer than 3% were immunoreactive for ovalbumin. In contrast, most sensory neurons projecting to subdermis, muscle and expansor secundariorum muscle were of a medium diameter. Approximately 25% of those neurons projecting to the expansor secundariorum, and 60% projecting to the subdermis and muscle, were immunoreactive for ovalbumin. Sensory neurons innervating heart and liver were the smallest, and only 8% were immunoreactive for ovalbumin. The study indicates that sensory neurons innervating different organs have somata with significantly different sizes, suggesting a functional specificity. Moreover, neurons demonstrating either the ovalbumin-IR positive or negative phenotypes show distinct peripheral projections, suggesting that this phenotype may be at least partially controlled by retrograde signals derived from the cells they innervate.     

Characterization of substances which promote or repel sympathetic fiber growth in vitro.

To determine whether a recognition mechanism is involved in determination of sympathetic innervation patterns of various tissues, tissue-derived substances were applied to a restricted test surface region of dishes and the responses of cultured sympathetic neurites were examined. Sympathetic fibers exhibited a turning or ramifying response, resulting in a dense fiber growth on test regions coated with particulate (adheron) fractions of a conditioned-medium (CM) from expansor secundariorum, heart, peripheral blood vessel or abdominal aorta, whereas on test regions coated with those from lung, skeletal muscle or dorsal aorta the neurite growth was repelled and sparse fiber growth was observed. Particulate fractions of brain- or gizzard-CM had no effect. These patterns in vitro were in parallel with the dense sympathetic innervation in expansor secundariorum, heart, peripheral blood vessel and abdominal aorta, but little or no sympathetic innervation in lung, skeletal muscle and dorsal aorta in vivo. These results suggest that adheron particles may participate in determination of sympathetic innervation patterns. Activity which repels or promotes the sympathetic fiber growth was inactivated by pronase E or trypsin but not by DNase or neuraminidase. Repelling activity was lost after treatment with heparinase or heparitinase but not with chondroitinase ABC or hyaluronidase. Promoting activity was retained after treatment with these glycosidases. These results suggest that the factor(s) possessing a repellent effect is a heparan sulfate proteoglycan and one(s) possessing a promoting effect is a protein.     

Smooth muscle sensitization and neuromuscular depression induced by chronic administration of antimalarial drugs.

The functional effects on chick smooth and skeletal muscle of chronic administration of 60 mg kg-1 chloroquine or quinacrine given as daily intraperitoneal injections for 70 days have been investigated. Noradrenaline and potassium chloride (KCl) contracted the normal expansor secundariorum muscle, a smooth muscle from the wing of chicks wholly innervated by noradrenergic nerves. The muscle was unresponsive to acetylcholine and histamine. Chronic administration of chloroquine or quinacrine induced supersensitivity of expansor muscles to KCl and the muscles were contracted by acetylcholine and histamine. These actions were more pronounced in quinacrine-treated chicks and could be due to direct smooth muscle sensitization that may result in postjunctional changes. The oesophagus is a smooth muscle that is predominantly under parasympathetic control. The oesophagus from chronically-treated chicks was more sensitive to acetylcholine and KCl than the control muscles. This sensitization was more marked for chloroquine than quinacrine. Chronic administration of chloroquine and quinacrine depressed skeletal muscle contractions evoked by acetylcholine and potassium chloride. These findings indicate that chronic chloroquine and quinacrine administration sensitise smooth muscle to agonist drugs but depress neuromuscular transmission.     

Changes in smooth muscle sensitivity to agonist drugs during development.

The receptors in the expansor secundariorum muscle of chicks were characterized pharmacologically and the changes in their response to nerve stimulation and agonist drugs determined during development. The muscle responded to noradrenergic nerve stimulation, noradrenaline and 5-hydroxytryptamine without any change in sensitivity during development. Expansor muscles from 15-day-old chicks were more sensitive to isoprenaline than muscles from older animals. The muscle from 15-day-old chicks responded to acetylcholine and histamine; the sensitivity to both drugs decreased progressively with increasing age of the chicks and disappeared by day 40 posthatching. The normal developmental decrease in response to acetylcholine and histamine were prevented by surgical denervation of the muscle; an intervention that also induced supersensitivity to noradrenaline greater than isoprenaline greater than 5-hydroxytryptamine. The muscle responded to potassium chloride without any change in sensitivity during development or following surgical denervation. These findings indicate that sympathetic nerves influence the responsiveness of the expansor secundariorum muscle to drugs, especially the development decrease in response to acetylcholine and histamine.     

Immunohistochemical localization of parvalbumin in rat and monkey autonomic ganglia

Summary The cellular distribution of parvalbumin-like immunoreactivity in autonomic ganglia such as superior cervical sympathetic ganglia, paravertebral sympathetic chain ganglia (T₆), ciliary ganglia and enteric ganglia was investigated by immunohistochemical peroxidase—antiperoxidase methods using an antiserum against rat skeletal muscle parvalbumin. We detected parvalbumin-like immunoreactivity in almost all neurons of rat superior cervical sympathetic ganglia and other paravertebral sympathetic chain ganglia, where the antigen was located in the cytoplasm but the nuclei were not labelled. No neurons positive for parvalbumin-like immunoreactivity were observed in rat ciliary ganglia or enteric ganglia. In monkey, almost all neurons of the superior cervical sympathetic ganglia contained parvalbumin-like immunoreactivity, but none of the neurons of the ciliary ganglia were labelled with the antiserum to parvalbumin. These results suggest that parvalbumin-like immunoreactivity exists in a specific subpopulation of the neurons of the autonomic nervous system.     

Reconsideration of the Autonomic Cranial Ganglia: An Immunohistochemical Study of Mid‐Term Human Fetuses

The cranial parasympathetic ganglia have been reported to paradoxically contain the sympathetic nerve marker, tyrosine hydroxylase (TH), in addition to neurons expressing parasympathetic markers such as vasoactive intestinal peptide (VIP) and neuronal nitric oxide synthase (nNOS). However, the distribution of these molecules in the cranial ganglia of human fetuses has not yet been examined. Using paraffin sections from 10 mid‐term human fetuses (12–15 weeks), we performed immunohistochemistry for TH, VIP, and nNOS in the parasympathetic ciliary, pterygopalatine, otic, and submandibular ganglia, and for comparison, the sensory inferior vagal ganglion. The ciliary and submandibular ganglia contained abundant TH‐positive neurons. In the former, TH‐positive neurons were much more numerous than nNOS‐positive neurons, whereas in the latter, nNOS immunoreactivity was extremely strong. No or a few cells in the pterygopalatine, otic, and inferior vagal ganglia expressed TH. Ciliary TH neurons appeared to compensate for classically described sympathetic fibers arising from the superior cervical ganglion, whereas in the submandibular ganglion, nNOS‐positive neurons as well as TH neurons might innervate the lingual artery in addition to the salivary glands. Significant individual variations in the density of all these markers suggested differences in sensitivity to medicine affecting autonomic nerve function. Consequently, in the human cranial autonomic ganglia, it appears that there is no simple dichotomy between sympathetic and parasympathetic function. Anat Rec, 2012. © 2011 Wiley Periodicals, Inc.     

Properties of postganglionic sympathetic neurons with axons in the right thoracic vagus.

The resting and reflex-evoked activities of single postganglionic sympathetic neurons with axons in the right thoracic vagus were tested in chloralose-anaesthetized cats. The properties of a majority of neurons were found to be similar. Cardiac- and inspiration-related rhythmicities were present in the resting activity of sympathetic neurons. Their resting activity was not affected by hyperventilation which abolished phrenic nerve discharges. Systemic hypoxia (2 min; 8% O2 in N2) increased the activity of the neurons more effectively in the deafferented state than when both sinus nerves remained intact. Injection of 0.1 ml 1 M sodium bicarbonate saturated with CO2, which activates peripheral chemoreceptors in the right or left carotid sinus, usually evoked a decrease in sympathetic activity in animals with both sinus nerves intact. We concluded that activation of peripheral chemoreceptors may inhibit the activity of the sympathetic neurons with axons in the right thoracic vagus. We suggest that the described sympathetic neurons may be a functionally homogeneous population which may innervate the conducting system of the heart. The close localization of sympathetic and parasympathetic axons in the vagus nerve may facilitate sympathetic-parasympathetic interaction at the level of their endings in the heart.     

Novel concept for the epaxial/hypaxial boundary based on neuronal development

Trunk muscles in vertebrates are classified as either dorsal epaxial or ventral hypaxial muscles. Epaxial and hypaxial muscles are defined as muscles innervated by the dorsal and ventral rami of spinal nerves, respectively. Each cluster of spinal motor neurons passing through dorsal rami innervates epaxial muscles, whereas clusters traveling on the ventral rami innervate hypaxial muscles. Herein, we show that some motor neurons exhibiting molecular profiles for epaxial muscles follow a path in the ventral rami. Dorsal deep-shoulder muscles and some body wall muscles are defined as hypaxial due to innervation via the ventral rami, but a part of these ventral rami has the molecular profile of motor neurons that innervate epaxial muscles. Thus, the epaxial and hypaxial boundary cannot be determined simply by the ramification pattern of spinal nerves. We propose that, although muscle innervation occurs via the ventral rami, dorsal deep-shoulder muscles and some body wall muscles represent an intermediate group that lies between epaxial and hypaxial muscles.     

Morphology of the accessory ciliary ganglion of the cat

Summary A microdissection of the orbital nerves of the cat was made paying particular attention to the accessory ciliary ganglion. The structures were studied with the light microscope by histological and histochemical techniques. The accessory ciliary ganglion consisted of cells similar to those in the main ganglion. The neurons of the accessory ciliary and those of the main ciliary ganglion emitted no specific fluorescence for catecholamines. After injection of horseradish peroxidase-wheat germ agglutinin conjugates into the chambers of the eye, most ganglion cells of the accessory and main ciliary ganglia were labeled. A few labeled ectopic cells were also found in proximity of the accessory ganglion in the lateral short ciliary nerve or in the communicating branch from the trigeminal nerve. The present results indicate that the accessory ciliary and ectopic ganglion cells are parasympathetic in nature and may innervate the intrinsic eye musculatures.     

Principal mechanisms controlling penile retraction and protrusion in rabbits

The effects on penile volume of nerve stimulations and drugs injected into the systemic circulation were studied plethysmographically. Dilator responses at selective perfusion of the penile artery were studied by measuring the perfusion pressure. The main results and conclusions are: The penis has an adrenergic vasoconstrictor supply coming from the sacrococcygeal parts of the sympathetic chains. A very low (0.2 Hz) vasomotor tone keeps the penis relaxed. If this tone is interrupted the penis will protrude but autoregulation will soon take over and eventually produce hyperinvolution of the penis. Two vasodilator paths, both with pelvic ganglionic relays, were found. 1) The pelvic parasympathetic nerves, probably having mainly non-cholinergic postganglionic neurons and operating quite effectively at low frequencies. 2) The sympathetic hypogastric nerves, presumably having at least partly cholinergic postganglionic neurons which, apart from muscarinic dilation of minute inflow resistance vessels to the erectile tissue, may also work by suppression of excitatory adrenergic neurotransmission. The pelvic and hypogastric vasodilator outflows work synergistically. The vasoconstrictor nerves are very strong and efficient antagonists of the vasodilator nerves.     

Development of cholinergic traits in the quail ciliary ganglion: expression of choline acetyltransferase-like immunoreactivity

The avian ciliary ganglion is a parasympathetic ganglion derived from the neural crest whose neurons provide cholinergic innervation to the eye. Here, we describe the time course of appearance and the morphology of cholinergic cells in the ciliary ganglion, as assessed by antibodies against choline acetyltransferase. Choline acetyltransferase immunoreactivity was first observed in 5.5-day-old quail embryos, 1 day after condensation of the ciliary ganglion. Both the intensity of choline acetyltransferase immunoreactivity and size of the choline acetyltransferase-immunoreactive cells increased with ganglionic age. By 12 days, a second population of choline acetyltransferase-immunoreactive cells, possibly corresponding to choroid neurons, was observed whose cells were smaller and less intensely stained than earlier differentiating choline acetyltransferase-immunoreactive cells. The percentage of choline acetyltransferase-immunoreactive cells was initially high, constituting approximately 50% of the total cell population. As a function of time, the proportion of cholinergic cells decreased, probably due to proliferation of non-neuronal cells and naturally-occurring cell death. Our results confirm the existence of two morphologically distinct populations of cholinergic neurons in the avian ciliary ganglion and demonstrate that these neuronal subpopulations express choline acetyltransferase immunoreactivity at different times in development. Because choroid neurons innervate their targets later than ciliary neurons, this finding is consistent with the hypothesis that target interactions regulate expression of choline acetyltransferase.     

Nervous regulation of the tone of the retractor penis muscle in the goat

The nervous control of the retractor penis muscle (rp) was investigated in the anaesthetized goat. Also, isolated field stimulated strips of the muscle were studied. The noradrenaline (NA) and acetylcholine (ACh) content of the rp was determined, and histochemistry for adrenergic and acetylcholinesterase (AChE) positive nerves was performed. The muscle exhibited spontaneous activity that persisted after section of all nerves. There was, however, also a tendency of the activity to follow the general vasomotor tone, which disappeared after section of the sympathetic chains. The excitatory adrenergic nerves which innervate the muscle come from the sympathetic chains and run along the pudendal, the hypogastric and the pelvic nerves. The rp has a dense network of adrenergic fibres and is very sensitive to excitatory adrenergic stimulation. It has a fairly large NA content, which is higher in old goats (5.95 ± 0.42 μg g^-1) than in young goats (2.87 ± 0.78 μg g-¹). Inhibitory non-adrenergic non-cholinergic (NANC) innervation reaches it via the pelvic and the hypogastric nerves. The maximum inhibitory response is reached at low frequencies (2–4 Hz). Cholinergic prejunctional inhibition of the excitatory response to sympathetic chain stimulation was effected by simultaneous stimulation of the hypogastric nerves. In vitro experiments confirmed the presence of endogenous cholinergic muscarinic suppression of the excitatory adrenergic neurotransmission. Significant amounts of ACh (0.81 7 plusmn; 0.18 μg g^-1) are present in the muscle, and it contains strongly AChE positive nerve fibres and nerve cell bodies. It is concluded that the goat rp is innervated by sympathetic adrenergic excitatory nerves and parasympathetic NANC inhibitory nerves. It further has a direct sympathetic inhibitory NANC innervation, and an indirect inhibitory cholinergic innervation which at least in part is sympathetic.     

Sensory and autonomic innervation of the rat eyelid: Neuronal origins and peptide phenotypes

Neuronal origins, peptide phenotypes and target distributions were determined for sensory and autonomic nerves projecting to the eyelid. The retrograde tracer, Fluoro-Ruby, was injected into the superior tarsal muscle and meibomian gland of Sprague-Dawley rats. Labelled neurons were observed within the pterygopalatine (31 ± 6 of a total of 8238 ± 1610 ganglion neurons), trigeminal (173 ± 43 of 62 082 ± 5869) and superior cervical ganglia (184 ± 35 of 21 900 ± 1741). Immunostaining revealed vasoactive intestinal polypeptide immunoreactivity (VIP-ir) in nearly all Fluoro-Ruby-labelled pterygopalatine ganglion neurons (86 ± 5%) but only rarely in trigeminal (0.3 ± 0.3%) or superior cervical (1.4 ± 1.4%) ganglion neurons. Calcitonin gene-related peptide (CGRP)-ir was not observed in pterygopalatine or superior cervical ganglion somata, but was present in 24 ± 4% of trigeminal neurons. Bright dopamine β-hydroxylase (DBH) immunofluorescence was observed in the majority of eyelid-projecting neurons within the superior cervical ganglia (65 ± 5%) and lighter staining was detected in pterygopalatine neurons (63 ± 3%), but no DBH-ir was observed in trigeminal neurons. Examination of eyelid sections revealed dense VIP-ir innervation of meibomian gland acini and vasculature and modest distribution within tarsal muscle. CGRP-ir fibers surrounded ductal and vascular elements of the meibomian gland and the perimeter of tarsal muscle. DBH-ir fibers were associated with meibomian gland blood vessels and acini, and were more densely distributed within tarsal muscle. This study provides evidence for prominent meibomian gland innervation by parasympathetic pterygopalatine ganglion VIP-ir neurons, with more restricted innervation by sensory trigeminal CGRP-ir and sympathetic neurons. Tarsal muscle receives abundant sympathetic innervation, as well as moderate parasympathetic and sensory CGRP-ir projections. The eyelid contains substantial non-CGRP-ir sensory innervation, the targets of which remain undetermined. The distribution of identified autonomic and sensory fibers is consistent with the idea that meibomian gland function, as well as that of the tarsal muscle, is regulated by peripheral innervation.     

Developmental regulation of parasympathetic nerve density by sympathetic innervation in the tarsal smooth muscle of the rat.

The developmental influence of sympathetic innervation on parasympathetic nerve density was investigated in the tarsal smooth muscle of the rat. Specificity of acetylcholinesterase staining as a marker for parasympathetic innervation was first determined by acute selective denervations. Excision of the ipsilateral superior cervical ganglion caused a 39% reduction in the density of acetylcholinesterase-positive nerves seven days later, indicating that sympathetic nerves contribute to cholinesterase-positive tarsal muscle innervation. Excision of the pterygopalatine ganglion concurrent with superior cervical ganglionectomy caused a virtually complete disappearance of acetylcholinesterase-positive innervation within seven days, indicating that non-sympathetic cholinesterase-positive fibers derive from the pterygopalatine ganglion and are presumed to be parasympathetic. Analysis of the control population indicated that parasympathetic nerve density did not vary significantly between males and females, between the superior and inferior muscles, or in rats studied at four and 12 months of age. The influence of sympathetic innervation on parasympathetic nerve density during postnatal development was examined by conducting surgical sympathectomies on postnatal day 5 and quantifying acetylcholinesterase-positive nerve density at four months of age. Neonatal sympathectomy caused a 46% reduction in cholinesterase-positive nerve density beyond that which occurred in acutely sympathectomized adult controls. It is concluded that sympathetic innervation is required for developing parasympathetic nerves to attain their normal density within the rat tarsal muscle. This finding is consistent with the idea that sympathetic nerves can exert positive effects on parasympathetic nerve outgrowth during development.     

Neurotransmitters in parasympathetic ganglionic neurons and nerves in mouse lower airway smooth muscle

In most species, including humans, lower airway smooth muscle (ASM) contains nerve terminals from two distinct populations of parasympathetic ganglionic neurons based on neurotransmitter phenotype: cholinergic and non-adrenergic non-cholinergic (NANC), causing contraction and relaxation, respectively, of ASM. Using immunohistological staining, the density and distribution of NANC-associated neurotransmitters, vasoactive intestinal peptide (VIP) and nitric oxide synthase were 6% of total nerve profiles compared to 19% cholinergic nerves in ASM in mouse (C57BL/6) central airways. The location of the NANC parasympathetic neurons innervating the tracheal ASM, as determined by retrograde neuronal tracer from the trachealis muscle, was the myenteric plexus of the esophagus, closely associated with the outer striated longitudinal muscle layers; the majority of the retrograde-labeled neurons were VIP- and NOS-IR. The results of these experiments provide the first direct evidence that VIP-IR and NOS-IR neurons intrinsic to the mouse esophagus project axons to the adjacent trachealis muscle.     

Regulation of A-currents by cell-cell interactions and neurotrophic factors in developing chick parasympathetic neurones.

1. The developmental regulation of ion channel expression was studied in parasympathetic neurones isolated from the chick ciliary ganglion. Whole-cell patch clamp recordings were made from ciliary ganglion neurones that were removed from the embryo on the ninth embryonic day (E9) and maintained in dissociated cell culture for an additional 4 days. Previous studies have shown that the expression of a transient voltage-activated K+ current (IA) is regulated by unidentified environmental stimuli during these developmental stages. 2. The effect of interactions between neurones and target tissue on the expression of IA was tested by co-culturing ciliary ganglion neurones with chick striated muscle cells. Neurones from the nerve-muscle co-cultures expressed normal amplitudes of IA, but the neurones did not express normal levels of IA when they were plated onto lysed muscle fibres. 3. The effect of interactions between ganglionic neurones and non-neuronal ganglionic cells was tested by culturing ganglia as explants rather than as dissociated cells. Neurones isolated from the explant cultures did not express normal levels of IA. Similarly, when dissociated ganglionic neurones were co-cultured with fibroblasts isolated from embryonic chick skin, they did not express normal amplitudes of IA. 4. Chronic depolarization caused by growing ciliary ganglion neurones in the presence of elevated K+ concentrations did not allow for the normal expression of IA, although it did promote the survival of these neurones in vitro. 5. Addition of 40 ng ml-1 of recombinant human ciliary neurotrophic factor (CNTF) or basic fibroblast growth factor (bFGF) to the cell culture medium had no effect on IA expression in developing chick ciliary ganglion neurones. However, 40 ng ml-1 of acidic fibroblast growth factor (aFGF) stimulated the expression of IA. All trophic factors promoted the growth and survival of ciliary ganglion neurones in vitro. 6. Dissociated ciliary ganglion neurones were maintained in a culture medium containing an aqueous extract of the whole brain. Neurones developing under these conditions expressed normal levels of IA. The stimulatory activity of the brain extract was destroyed by heating. 7. The expression of IA in chick ciliary ganglion neurones developing in vitro can be regulated by soluble growth factors and by interactions with certain other cell types. Similar interactions may regulate the expression of IA in ciliary ganglion neurones developing in situ.     

A light and electron microscopic study of the innervation of pulmonary arteries in the cat

Nerve terminal-smooth muscle relationships were studied in pulmonary arteries of the cat using 5-hydroxydopamine to help differentiate adrenergic and nonadrenergic terminals. There was a periarterial plexus of nerves in the walls of pulmonary arteries that extended into the lung to innervate even small arteries having a single layer of smooth muscle cells. Adrenergic nerves surrounded all arteries and extended into the tunica media of the large arteries. There were also apparent cholinergic nerves around the pulmonary arteries, although this was confirmed by electron microscopy for medium- and small-sized arteries only. The relationships of nerve terminals to smooth muscle cells in pulmonary arteries suggest that release of norepinephrine by adrenergic terminals can produce both decreased compliance and increased resistance in the pulmonary vascular bed, and that acetylcholine released by cholinergic terminals may act directly on vascular smooth muscle or on adrenergic terminals to modulate release of norepinephrine. These results suggest that both sympathetic and parasympathetic nerves may have a regulatory role in the pulmonary circulation.     

Formation of Neuromuscular Junction‐Like Structure Between Primary Sensory Terminals and Skeletal Muscle Cells in Vitro

Sensory nerve cross‐anastomosis provides a modified trophic environment by modulating neurotrophic factor synthesis in muscle. Target tissues contribute to the phenotype and function of sensory neurons. Whether formation of neuromuscular junction (NMJ)‐like structure between sensory neurons and skeletal muscle (SKM) cells in vitro remains unknown. In this study, a neuromuscular coculture model of dissociated dorsal root ganglion (DRG) neurons and SKM cells was established. The relationship between DRG neurons and SKM cells was observed by light microscopy and scanning electron microscopy (SEM). The results showed that: (1) DRG neuronal axons frequently terminated on or adhered to the SKM cells; (2) the crossing axons adhered to each other, hence displacement of the terminal axons on the contracting SKM cells would also oscillate the proximally crossing axonal network; (3) the configurations of the axon terminal observed by SEM were variable in different culture conditions; (4) the enlarged nerve endings terminated on the surface of SKM cells which formed NMJ‐like structure. These results offered new clues for a better understanding of the relationship between sensory neurons and SKM cells. Anat Rec, 2011. © 2010 Wiley‐Liss, Inc.     

Nerve pathways between the pterygopalatine ganglion and eye in cats

By dissection of thiocholine-stained orbital preparations, it has been determined that three different nerve pathways link the pterygopalatine ganglion and the eye in cats. 1) Nerves from the proximal half of the ganglion join a plexus of nerves and ganglion cells in the rete mirabile of the maxillary artery. Branches of the internal carotid nerve also supply this plexus. Fine nerves from the plexus travel to the optic nerve and then to the eye, accompanying both the nasociliary nerve that passes through the rete and the ciliary arteries that arise from the rete. 2) One or more nerves from the nerve of the pterygoid canal and from a prominent accessory ganglion near the orbital apex course to the inferior optic nerve surface at the optic foramen; these then run distally along the optic nerve to fuse with ciliary nerves or to accompany ciliary arteries entering the eye. 3) Other nerves from the pterygopalatine ganglion travel medially around the extraocular muscle cone to join the ethmoidal and infratrochlear branches of the nasociliary nerve; some nerves from the ganglion then take a retrograde course to the optic nerve, where they join ciliary nerves or arteries to the eye. All three pathways may transmit sympathetic, parasympathetic and somatic sensory nerve fibers.