Pupillary light reflex circuits in the macaque monkey: the preganglionic Edinger–Westphal nucleus
- 98 Downloads
The motor outflow for the pupillary light reflex originates in the preganglionic motoneuron subdivision of the Edinger–Westphal nucleus (EWpg), which also mediates lens accommodation. Despite their importance for vision, the morphology, ultrastructure and luminance-related inputs of these motoneurons have not been fully described in primates. In macaque monkeys, we labeled EWpg motoneurons from ciliary ganglion and orbital injections. Both approaches indicated preganglionic motoneurons occupy an EWpg organized as a unitary, ipsilateral cell column. When tracers were placed in the pretectal complex, labeled terminals targeted the ipsilateral EWpg and reached contralateral EWpg by crossing both above and below the cerebral aqueduct. They also terminated in the lateral visceral column, a ventrolateral periaqueductal gray region containing neurons projecting to the contralateral pretectum. Combining olivary pretectal and ciliary ganglion injections to determine whether a direct pupillary light reflex projection is present revealed a labeled motoneuron subpopulation that displayed close associations with labeled pretectal terminal boutons. Ultrastructurally, this subpopulation received synaptic contacts from labeled pretectal terminals that contained numerous clear spherical vesicles, suggesting excitation, and scattered dense-core vesicles, suggesting peptidergic co-transmitters. A variety of axon terminal classes, some of which may serve the near response, synapsed on preganglionic motoneurons. Quantitative analysis indicated that pupillary motoneurons receive more inhibitory inputs than lens motoneurons. To summarize, the pupillary light reflex circuit utilizes a monosynaptic, excitatory, bilateral pretectal projection to a distinct subpopulation of EWpg motoneurons. Furthermore, the interconnections between the lateral visceral column and olivary pretectal nucleus may provide pretectal cells with bilateral retinal fields.
KeywordsPupil Autonomic Luminance Midbrain Near response
Biotinylated dextran amine
Caudal central subdivision
Preganglionic Edinger–Westphal nucleus
Interstitial nucleus of Cajal
Left ciliary ganglion
Lateral visceral column
Medial dorsal nucleus
Medial geniculate nucleus
Medial longitudinal fasciculus
Medial pretectal nucleus
Nucleus of Darkschewitsch
Nucleus of the optic tract
Nucleus of the posterior commissure
Olivary pretectal nucleus
Posterior pretectal nucleus
Rough endoplasmic reticulum
Right ciliary ganglion
Wheat germ agglutinin-conjugated horseradish peroxidase
The preganglionic motoneurons whose axons travel with the third cranial nerve are located in the preganglionic subdivision of the Edinger–Westphal nucleus (EWpg) (Kozicz et al. 2011; May et al. 2008a), a cell group associated with the oculomotor nucleus (III) that was first described by Edinger (1885) and Westphal (1887) over a century ago. These parasympathetic motoneurons synapse on postganglionic motoneurons in the ciliary ganglion that, in turn, supply the intraocular muscles of the eye. There are two populations within EWpg: one controls lens accommodation by activating the ciliary muscle and the other controls pupillary constriction by activating the pupillary sphincter muscle (Gamlin et al. 1984, 1994; Hultborn et al. 1973; McDougal and Gamlin 2015; May et al. 2019b).
Two behaviors use this parasympathetic outflow: the near response and the pupillary light reflex. The near response is initiated when animals direct their eyes to a nearby object. To do this, they execute three interrelated actions. (1) The lines of sight are rotated nasally (converged), pointing the foveae of both eyes toward the object. (2) The curvature of the lens is increased through the actions of the ciliary muscle, to focus the closer object. (3) The pupils constrict to produce greater depth of focus and less spherical aberration. The simultaneous occurrence of accommodation, convergence and pupillary constriction form the near triad. The pupillary light reflex acts to regulate the amount of light falling on the retina to optimize luminance levels for the photoreceptors. The short latency response of the pupil is primarily regulated by the parasympathetic input to the iris (Loewenfeld 1993). The pathway for the pupillary light reflex originates with retinal ganglion cells that can be characterized as broad-field luminance detectors. These cells contain melanopsin and have been characterized as intrinsically photoreceptive retinal ganglion cells (ipRGCs), although they also receive photoreceptor input (Güler et al. 2008; Hannibal et al. 2014). They send their axons to the olivary pretectal nucleus (OPt), with the temporal retina projecting to the ipsilateral OPt and the nasal retina providing input via the chiasm to the contralateral OPt. Like their retinal inputs, the cells in OPt are also classified as broad-field luminance detectors (Gamlin et al. 1995). This nucleus is believed to subsequently project to the EWpg. The action of this pathway is inhibited when there is increased activation of sympathetic pathways to the pupillary dilator muscle under low illumination conditions and due to changes in the state of the animal. In the latter, the pupils dilate in response to increased states of attention, arousal and/or cognitive load (Kahneman and Beatty 1966; Beatty 1982; Gabay et al. 2011; Szabadi 2013; Joshi et al. 2016), as well as with saccades (Wang and Munoz 2015).
Despite the critical role of EWpg motoneurons in these two important visual functions, there is relatively little information available about their inputs and ultrastructure in the primate, compared to other mammalian species (Ichinohe et al. 1996; Klooster et al. 1995; Sun and May 2014a, b). Even the organization of the primate EWpg is a matter of dispute. Some authorities describe this nucleus as a unitary column stretching rostrocaudally, dorsal to III (Akert et al. 1980; May et al. 2008a), whereas others have divided the nucleus into a number of subdivisions (Burde 1983; Burde and Williams 1989). In fact, it has been suggested that the pupillary preganglionic motoneurons inhabit one of these subdivisions, the lateral visceral column (Burde 1983; Büttner-Ennever et al. 1996; Kourouyan and Horton 1997). The precise pattern of connections between the pretectum and EWpg is also a matter of dispute. Some believe that OPt projects bilaterally, with decussating fibers traveling in the posterior commissure (Benevento et al. 1977; Klooster et al. 1995; Kourouyan and Horton 1997). However, other reports indicate a strictly contralateral projection (Steiger and Büttner-Ennever 1979; Clarke et al. 1985a). In fact, a monosynaptic projection of the pretectum onto preganglionic motoneurons has not been proven in primates.
To fill these gaps in our knowledge, we have undertaken a comprehensive characterization of these preganglionic parasympathetic motoneurons and their synaptic contacts in the macaque monkey. Furthermore, we have utilized neuronal tracers to label synaptic input from the pretectum to discriminate motoneurons involved in pupillary constriction. Brief reports of some of these data have appeared previously (Sun and May 1995; Erichsen et al. 1998; May et al. 2008b).
The surgical procedures described below are in accordance with NIH guidelines and were approved by the Institutional Animal Care and Use Committee of the University of Mississippi Medical Center. Material from Macaca fascicularis (n = 13) and Macaca mulatta (n = 5) monkeys (> 3.0 kg) of both sexes was used in this study. Some of these cases were also used in other, non-conflicting studies. The animals were initially sedated with ketamine HCl (10 mg/kg, IM). For intraocular injections, animals were anesthetized with ketamine (22 mg/kg, IM) and xylazine (1 mg/kg, IM), and then proparacaine drops were applied to the cornea. For central injections and injections of the ganglia, an i.v. line was put in place to maintain hydration and a tracheal tube was introduced to allow induction of inhalation anesthesia with 3% isoflurane. Vital signs including temperature and heart rate were monitored and maintained within normal limits. The animals were given dexamethasone (1 mg/kg, IV) and atropine sulfate (0.05 mg/kg, IV) to control brain edema and respiratory secretions, respectively. All surgical procedures took place in a surgical suite and utilized sterile technique. The animals received Butorphanol (0.01 mg/kg, IM) or Buprenex (0.001 mg/kg, IM) as a postsurgical analgesic. Before cardiac perfusion, they were sedated with ketamine HCl (10 mg/kg, IM) and then deeply anesthetized with sodium pentobarbital (50 mg/kg, IP).
Case injection details
We used a lateral approach to inject the ciliary ganglion. An incision was made in the skin of the left temple. The anterior edge of the temporalis muscle was disinserted and retracted caudally. The skull over the lateral caudal aspect of the orbit was removed to allow visualization of the lateral rectus muscle, which was disinserted from the globe and retracted. A syringe with a 27 G needle was used to remove aqueous from the anterior chamber to partially deflate the globe. Blunt dissection revealed the optic nerve, and then the short ciliary nerves, which accompany it, were followed back to the ciliary ganglion. Connective tissue within this ganglion makes it difficult to inject using a needle. Therefore, we utilized small insect pins mounted on orange sticks to place the tracer in the ganglion. A paste of WGA–HRP and HRP was dried onto the pin tips. These were inserted into the ganglion and held until the tracer dissolved. The muscles were reattached and the overlying skin edges sutured together. One of these animals also received a pretectal biocytin injection, as described below.
A number of different tracers were used to label pretectal inputs to the EWpg: WGA–HRP, biocytin, biotinylated dextran amine (10,000 MW) (BDA) and Phaseolus vulgaris leukoagglutinin (PhaL) (see Table 1 for details). Due to the small size of the olivary pretectal nucleus (OPt), this nucleus was not injected in isolation. The dorsal surface of the midbrain was visualized by aspirating the overlying cortex. PhaL was injected iontophoretically using a glass micropipette with a 25-µm tip (7 µA, for 10 min, 50% duty cycle positive current). The other tracers were injected using a 1-µl Hamilton syringe. In each case, the needle or pipette was angled between 23o and 30o tip up from vertical in the parasagittal plane. The defect produced by the aspiration was filled with Gelfoam and the incision closed. After the appropriate survival times, the animals were deeply anesthetized and then perfused with buffered saline followed by 1.0% paraformaldehyde and 1.25% glutaraldehyde in 0.1 M, pH 7.2 PB (WGA–HRP and BDA) or 2.0% paraformaldehyde, 1.0% glutaraldehyde in 0.1 M, pH 7.2 PB (Biocytin and PhaL).
After perfusion, the brainstem was blocked in the stereotaxic frontal plane, removed and postfixed in the fixative solution for 2 h. It was then stored in phosphate buffer at 4 °C until it could be cut and processed. The brainstem was cut into 50- or 100-μm sections in the frontal plane by use of a vibratome (Leica). Alternatively, the samples were cryoprotected in 30% sucrose and frozen sectioned on a sliding microtome (AO) at 40 or 80 μm. Ordered 1 in 3 series were reacted to reveal the tracers. Those tagged with HRP were reacted using the tetramethylbenzidine (TMB) procedure of Olucha et al. (1985) (see Perkins et al. 2009 for details). In other cases, we used the nitroprusside TMB method (Mesulam 1978). To reveal tracers tagged with biotin (Biocytin and BDA), the tissue was reacted with avidin-conjugated horseradish peroxidase (avidin–HRP) (Vector Labs) and then the HRP was revealed with the chromogen diaminobenzidine (DAB) using the procedure of Adams (1977) (see Perkins et al. 2009 for details). To visualize the PhaL, we used the method of Gerfen and Sawchenko (1984) (see Wang et al. 2013 for details) that uses biotinylated goat anti-PhaL (Vector Labs) and a goat ABC kit (Vector Labs). The HRP was visualized by use of DAB, as described above. For the dual-tracer experiment, the TMB protocol of Olucha was followed, and the blue reaction product was protected by DAB to reveal retrogradely transported WGA–HRP, followed by the biocytin protocol that used DAB intensified with nickel and cobalt (see Perkins et al. 2009 for details).
For light microscopy, the sections were mounted, counterstained with cresyl violet or neutral red, dehydrated, cleared and coverslipped. For electron microscopy (EM), areas containing labeled preganglionic motoneurons were excised under visual control using a Wild M8 stereomicroscope and prepared for EM. Care was taken to exclude the adjacent oculomotor nucleus, which contained scattered labeled motoneurons due to the spread of tracer from the ganglion. The sections were then prepared for light microscopy to allow sample area verification. The EM samples were processed and cut for EM using standard procedures (Barnerssoi and May 2016). Semithin sections were taken from the block face, stained with Toluidine, and used to direct further trimming to the area of interest. Ultrathin sections were photographed with a Zeiss 10C or a Leo transmission electron microscope. Synaptic contacts were photographed at a standard magnification of 21,560X on the EM, but the magnification for somata and dendrites varied.
Quantitative analysis of electron micrographs
The original electron micrographs were scanned to JPEG images and the terminals classified. Terminal classification, based on vesicle shape and synaptic density, and terminal measurements were made by separate individuals who were blinded to each other’s findings. The JPEG files were converted into tagged image format (TIF) to enable the use of NIH Image for Macintosh to process the images [https:/imagej.nih.gov/nih-image/download.html]. The images were first scaled to provide the correct number of pixels representing each micrometer. Then a ‘markup’ macro provided with NIH Image was loaded and a sufficient number of distinct colors were reserved in the look-up table (LUT) to allow the required number of image features to be traced with different colors. Boundaries of axon terminals, dendrites and cell bodies were marked onto the image along with lengths of membrane contacted by profiles and by synapses. Once colored graphics were extracted for measurement, NIH Image was used to ‘density slice’ through the LUT colors and separately measure the cross-sectional area, perimeter, major and minor axes, as well as the length of profile contact and synaptic contact. Where spines were visible, their values were not included in the ‘true’ profile and synaptic totals. The numerical data produced were saved to a spreadsheet and subsequently analyzed. Data were compared using a Student’s T test, and significance was set at p values less than 0.05.
Ultrastructure of preganglionic motoneurons and inputs
Quantitative ultrastructural findings
Somata n = 13
102.23 ± 8.86
30.25 ± 1.97
19.85 ± 1.26
15.12 ± 2.41
3.30 ± 0.42
All dendrites n = 47
37.43 ± 5.20
11.76 ± 1.51
7.00 ± 1.03
7.12 ± 0.93
1.52 ± 0.19
Proximal dendrites n = 33
47.27 ± 6.26
14.90 ± 1.80
8.73 ± 1.21
7.95 ± 1.20
1.69 ± 0.23
Distal dendrites n = 14
9.16 ± 1.29
3.16 ± 0.48
1.66 ± 0.16
1.94 ± 0.30
0.45 ± 0.08
We then turned our attention to the types of terminal in synaptic contact with the retrogradely labeled preganglionic motoneurons. The axon terminals (n = 226) innervating these motoneurons showed a majority (n = 103) of Type 1 profiles, with fewer Type 2 (n = 60) and Type 1D (n = 45) terminals. There were relatively few Type 2D terminals (n = 18), and Type 3 terminals (n = 3) were very rare. In this case, we did see a clear difference between the pupillary and other preganglionic motoneuron samples (Fig. 15b). The axon terminals (n = 55) contacting pupillary motoneurons showed a majority (n = 26) of Type 2D profile type, in which the vesicles were mostly clear pleomorphic with some scattered dense-core vesicles. In fact, this type was 39.2% more common on pupillary motoneurons than on other motoneurons. Type 1 profiles were relatively less common (n = 15), and equal quantities of clear Type 1D (n = 7) and Type 2 (n = 7) were present. No Type 3 profiles were observed on pupillary motoneurons. When we broke this analysis down with respect to neuronal geography, the preponderance of At2D profiles on pupillary neurons was still evident on the membranes of somata (Fig. 15c), proximal dendrites (Fig. 15 d) and distal dendrites (Fig. 15e). It is noteworthy that the terminal type associated with the actual pretectal input in the pupillary population of motoneurons, Type 1D, was only sampled on pupillary proximal dendrites (Fig. 15c–e). However, we found examples in which proximal dendrites that were contacted by labeled pretectal terminals displayed continuity with distal dendrites or somata (Fig. 14e, f).We analyzed these examples to characterize more of the pupillary population. Also of note is the fact that Type A1 terminals appear to be much less common on the distal dendrites of both motoneuron types, compared to the proximal dendrites and somata (Fig. 15c–e).
Several characteristics of the four main terminal types were measured. The mean terminal area ranged between 1 and 2 µm2, and varied slightly between terminal classes, but no significant differences were found between the areas of the terminal classes that contact the two motoneuron categories. Similarly, the mean length of the synaptic contacts ranged between 0.35 and 0.55 µm for these four types of profile, but the profile lengths of these different classifications revealed no significant differences for the two types of motoneuron.
Since we were struck by the unusual prevalence of the relatively rare Type 2D terminal on the pupillary preganglionic motoneurons, we undertook a separate comparison using just synaptic contact characteristics. We confined this analysis to samples where we were confident of the symmetrical or asymmetrical classification. We found no significant difference between the two motoneuron categories with respect to the length of profile apposition, the actual synaptic density length or the synaptic coverage. However, we did see differences with respect to the area of the terminals (Fig. 15f). The mean area of the terminals contacting the pupillary motoneurons was larger than that of terminals contacting motoneurons in the other category. Only the symmetrical terminal measures reached significance (p < 0.05). Symmetric contacts on pupillary preganglionic motoneurons had a mean area of 1.90 µm2 (s.e. 0.23 µm2), and those contacting the other motoneurons had a mean area of 1.29 µm2 (s.e. 0.12 µm2).
The results of this study indicate that the motoneurons in the macaque EWpg are organized into a single column that runs longitudinally, dorsal to III. Hopefully, this will lay to rest arguments over whether the monkey EWpg contains cytoarchitectonically distinct subdivisions. The results strongly suggest that the OPt provides a monosynaptic input to a subpopulation of EWpg motoneurons. This is presumably the substrate for the pupillary light reflex, and our data indicate it is a bilateral projection, supporting both the direct and consensual responses. The ultrastructural characteristics of this projection are congruent with the pretectal projection being excitatory, and the presence of dense-core vesicles suggests peptide neuromodulators may be present. Beyond receiving pretectal input, pupillary motoneurons differ from lens-related motoneurons at the ultrastructural level in that they receive a higher percentage of inhibitory inputs. Finally, this study indicates the presence of a feedback projection to the pretectum from a novel source, the lvc. This feedback may play a role in constructing the bilateral retinal fields characteristic of many OPt neurons.
Several of the techniques we used to label preganglionic motoneurons also produced labeling of somatic motoneurons. This made it difficult to differentiate extraocular motoneurons found in the C-group from preganglionic motoneurons belonging to EWpg. However, trans-synaptic transport of WGA–HRP, while it only produced light labeling of the preganglionic motoneurons, produced no labeling of somatic motoneurons. The consistency in the labeling pattern in EWpg between these tracer types provides further support for our conclusions.
The small size of the OPt and the fact that it is embedded within other pretectal nuclei made it difficult to inject this nucleus without involving adjacent structures. All of our injections spread to other nuclei, making interpretation of any one case difficult. However, the retrograde tracer studies presented in the companion paper (May and Warren 2019) suggest that there are only a few neurons projecting to the region containing EWpg in the nuclei surrounding OPt. The fact that different tracers produced similar patterns of anterograde labeling in the EWpg and SOA when the OPt was involved strongly supports our claims.
Only one dual-tracer experiment was undertaken. This lessens the impact of the argument in support of the monosynaptic projection of the OPt to the EWpg motoneurons. However, the patterns of label seen in this case were identical to all those observed with single injections of retrograde or anterograde tracer. This consideration also applies to the ultrastructural evidence for this projection. Furthermore, our quantitative analysis comparing pupillary preganglionic motoneurons to other preganglionic motoneurons rests on a relatively small sample where we could observe labeled pretectal terminals contacting labeled motoneurons in EWpg.
Organization of the Edinger–Westphal nucleus
Although the preganglionic motoneurons of the EW form a discrete nucleus in the birds that have been investigated (Reiner et al. 1983; Gamlin et al. 1984), in most mammals, the cholinergic cells of EWpg do not form a discrete nucleus, and are instead scattered in the vicinity of III (Sugimoto et al. 1978; Toyoshima et al. 1980; Kozicz et al. 2011; Sun and May 2014b). In contrast, the division of the EW that contains peptidergic, centrally projecting neurons (EWcp) often forms a fairly discrete nucleus in most mammals (May et al. 2008a; Kozicz et al. 2011). In primates, on the other hand, the motoneurons of EWpg generally form a more discrete nucleus located dorsal to III and extending into AM (present results; Warwick 1954; Akert et al. 1980; Burde and Loewy 1980; Clarke et al. 1985a; Sun and May 1993; May et al. 2008b), whereas the peptidergic cells of EWcp are somewhat more dispersed, lying on the midline between the oculomotor nuclei, within the SOA and lateral to III (Horn et al. 2008; May et al. 2008a; Kozicz et al. 2011).
It has been suggested that the EWpg of monkeys is divided into a number of cytoarchitectonic subdivisions, including dorsal, lateral and medial visceral columns (Burde 1983; Burde and Williams 1989). However, we believe that there are no distinct cytoarchitectonic subdivisions of EWpg for the following reasons: (1) no subdivisions were observed when an antibody to choline acetyltransferase was used to identify motoneurons dorsal to III (Horn et al. 2008; May et al. 2008a). (2) Subdivisions were not seen using retrograde trans-synaptic transport of rabies virus from the ciliary body (May et al. 2018). (3) In the present study, we did not find subdivisions of EWpg following injections of the ciliary ganglion or through the use of trans-synaptic transport of either WGA or WGA–HRP from the globe.
Preganglionic motoneuron ultrastructure
The ultrastructural examination of EWpg motoneurons indicated that these cells receive input from five different ultrastructural types of terminal: two presumably excitatory classes containing clear spherical vesicles—At1 and At1D, two presumably inhibitory classes containing clear pleomorphic vesicles—At2 and At2D, and a few terminals dominated by large dense-core vesicles—At3. It is always possible that some of the terminals classified as At1 and At2 might simply represent cuts where the sparse small dense-core vesicles were not present. While this sampling effect may have biased the counts towards the At1 and At2 categories, it seems unlikely that sampling can account for all the terminals present in these two categories, as these were generally more common than the corresponding At1D and At2D classes. Similar terminal types were observed contacting EWpg motoneurons in a cat study although a more complex classification system that also considered vesicle packing was employed (Sun and May 2014b).
The presence of multiple terminal types may reflect the presence of multiple inputs to these cells. For lens-related preganglionic motoneurons, inputs from both the near-response neurons in SOA and disjunctive saccade neurons in the central mesencephalic reticular formation should be present (May et al. 2016, 2018, 2019a). The near triad, which controls the eye with respect to target distance, involves the modulation of vergence angle, lens accommodation and pupillary diameter (Mays 1984; Zhang et al. 1992; McDougal and Gamlin 2015; May et al. 2019b). Consequently, many neurons in both these structures that fire for convergence would be expected to provide both lens-related and pupil-related motoneurons in EWpg with excitatory input (Mays, 1984; Mays et al. 1986; Judge and Cumming 1986; Zhang et al. 1992; Waitzman et al. 2008; Das 2011, 2012). These structures also contain populations that fire for divergent eye movements, so they may also provide EWpg motoneurons with inhibitory inputs. In addition, the pupillary preganglionic motoneurons receive an excitatory input from the OPt for the pupillary light reflex (Gamlin et al. 1995). Here, we have demonstrated that the OPt terminals fall into the At1D category of presumed excitatory inputs. This was also found to be true in the cat (Sun and May 2014b). The modulatory neuropeptide contained in their dense-core vesicles remains to be determined. Inhibitory (At2 and At2D) terminals to pupil-related motoneurons may arise from the hypothalamic regions that activate pupillary dilation via hypothalamospinal pathways (Sillito and Zbrožyna 1970a, b). The locus coeruleus is also believed to provide an inhibitory input to pupillary neurons in EWpg (Breen et al. 1983; Samuels and Szabadi 2008). This may be part of a system that dilates the pupil with respect to affect, interest and more general activation of the nervous system (Szabadi 2013; Joshi et al. 2016).
We compared the ultrastructure of preganglionic motoneurons that received OPt input, presumably pupillary motoneurons, and other preganglionic motoneurons that did not receive OPt input, a population highly enriched in, but not exclusively, lens-related motoneurons. These two populations share many general features of organization. The pupil-related motoneurons did, however, differ significantly from the mainly lens-related motoneurons in the following respects. Larger numbers of type At2D terminals were observed in contact with the pupillary motoneurons and presumably inhibitory symmetric contacts on pupillary motoneurons were larger. This suggests that pupillary motoneurons receive an additional inhibitory input that is not present on lens-related motoneurons. Perhaps these are tied to decreasing pupillary sphincter tone during dilation. However, while the activity patterns of the lens-related motoneurons have been described in detail (Gamlin et al. 1994), those of the pupil-related motoneurons have not (but see McDougal and Gamlin 2015).
Pupillary light reflex pathways
The present data indicate that the OPt projection to EWpg is a bilateral one (Fig. 16). In fact, it suggests that, in addition to a crossed projection by way of the posterior commissure, the contralateral EWpg also receives input via axons that decussate beneath the cerebral aqueduct. The terminal pattern we observed following OPt injections matches that reported previously in monkeys (Benevento et al. 1977; Büttner-Ennever et al. 1996) and in cats (Distler and Hoffmann 1989; Sun and May 2014b). Some retrograde studies have only reported a contralateral (Steiger and Büttner-Ennever 1979; Clarke et al. 1985a) or predominantly contralateral projection (Gamlin et al. 1995). It is possible that the ipsilateral anterograde labeling we observed here was due entirely to fiber-of-passage uptake. However, we believe this is unlikely, given that the same result was observed with several different tracers (present results), including tritiated amino acids (Büttner-Ennever et al. 1996). Thus, it would appear that morphological substrates for balanced direct and consensual pupillary response are present at the level of the retinal input to OPt and the projection of the OPt to the EWpg.
The area of the EWpg containing motoneurons that had close associations with OPt terminals lay roughly in the middle third of the rostrocaudal extent of the EWpg column. This location differs from that observed in the cat (Erichsen and May 2002; Sun and May 2014b). A simple explanation for this difference might be that the region of EWpg that receives the most contacts in both species lies closest to OPt. Only a portion of the close associations may represent actual synaptic contact. In fact, cells with numerous close associations, like those shown in Fig. 10b, represented quite a small proportion of the motoneuron population. These were found ventrally in EWpg, consistent with a ventral location for pupillary preganglionics reported in marmosets (Clarke et al 1985b, 2003a).
Lateral visceral column
One of the novel findings described here and in a previous short report (May et al. 2008b) was the presence of a cluster of neurons located dorsolateral to EWpg that projected to the contralateral pretectum. This area received both ipsilateral and contralateral input from the pretectum (Figs. 6 and 7). The location of these cells and terminal fields appears to be the same as that designated as the lvc in three previous studies. In one, trans-synaptic anterograde terminal labeling was seen from retinal injections in macaque monkeys (Kourouyan and Horton 1997). In the others, terminal labeling with a contralateral predominance was observed following pretectal injections (Baleydier et al. 1990; Büttner-Ennever et al. 1996). The designation lvc was used by these authors in the expectation that this nucleus represented one of the EW subdivisions proposed by Burde (1983). As noted above, we have not seen evidence of such subdivisions, and the region in question does not contain cholinergic cells, as a portion of EWpg would (Horn et al. 2008; May et al. 2008a). To maintain some consistency in terminology, we have provisionally maintained the designation lvc, with the proviso that it is not part of EW.
In the companion paper (May and Warren 2019), we described terminals in OPt after injections aimed at EWpg that included lvc. Based on a cross case analysis of our anterograde and retrograde cases, it seems likely that the lvc receives bilateral input from OPt, and that it projects to the contralateral OPt. Moreover, close associations between the pretectal terminals and the projection neurons suggest that this nucleus provides a monosynaptic feedback to OPt. Approximately 40% of the neurons in the primate OPt display bilateral receptive fields that are driven from both eyes (Clarke et al. 2003b). Direct retinal input from the ipsilateral temporal and contralateral nasal retina provides the drive for the contralateral visual field, but central pathways are needed to contribute to the ipsilateral visual field representation. As noted in the companion paper (May and Warren 2019), there is little evidence of commissural OPt connections that could provide this information. Thus, the pattern of connections between the lvc and OPt makes it a good candidate to provide ipsilateral visual field input to the OPt (Fig. 16).
We would like to thank Ms. Malinda Danielson, Jinrong Wei and Olga Golanov for their technical assistance with respect to surgeries and processing of the brains, as well as preparation of the figures. We are also indebted to Mr. Glen Hoskins for processing and cutting tissue for electron microscopy.
PJM helped to design the experiments, carry out the experiments, analyze the data, write the manuscript and edit the manuscript. WS helped to carry out the experiments, analyze the data, and edit the manuscript. NFW helped to analyze the data, write the manuscript and edit the manuscript. JTE helped to design the experiments, analyze the data, and edit the manuscript.
Portions of the material presented here were supported by funds from National Institute of Health Grants: EY07166 to Paul J. May, EY014263 to Paul J. May, Paul D.R. Gamlin and Susan Warren, and National Science Foundation Grant IBN-0130954 to Martha Bickford and Paul J. May. Nick Wright was supported by a grant from the National Lottery Charities Board (now the Community Fund).
Compliance with ethical standards
Conflict of interest
None of the authors has any perceived or real conflicts of interest with respect to this submission.
Ethical use of animals
All applicable international, national, and/or institutional guidelines for the care and use of animals were followed. All procedures performed in studies involving animals were in accordance with the ethical standards of the institution at which the studies were conducted. Specifically, they were undertaken under protocols approved by the Institutional Animal Care and Use Committee of the University of Mississippi Medical Center (USDA Animal Welfare Assurance #D16-00174).
- Edinger L (1885) Ueber den verlauf der centralen hirnnervehbahen mit demonstration von präparten. Neurol Zbl 4:309Google Scholar
- Erichsen JT, Wright NF, May PJ (1998) Light and electron microscopic observations on motoneurons subserving the near triad in the macaque. Invest Ophthalmol Vis Sci 39:S1049Google Scholar
- Gerfen CR, Sawchenko PE (1984) An anterograde neuroanatomical tracing method that shows the detailed morphology of neurons, their axons and terminals: immunohistochemical localization of an axonally transported plant lectin, Phaseolus vulgaris leucoagglutinin (PHA-L). Brain Res 290:219–238PubMedPubMedCentralCrossRefGoogle Scholar
- Güler AD, Ecker JL, Lall GS, Haq S, Altimus CM, Liao HW, Barnard AR, Cahill H, Badea TC, Zhao H, Hankins MW, Berson DM, Lucas RJ, Yau KW, Hattar S (2008) Melanopsin cells are the principal conduits for rod-cone input to non-image-forming vision. Nature 453(7191):102–105PubMedPubMedCentralCrossRefGoogle Scholar
- Ichinohe N, Shoumura K, Takahashi H (1996) Quantitative electromicroscope study of the oculomotor parasympathetic neurons projecting to the ciliary ganglion in cats: comparison of the synaptic (axon-somatic and axo-proximal dendritic) organization of anterior-dorsal and ventral cell groups. Anat Embryol (Berl) 193:229–238CrossRefGoogle Scholar
- Loewenfeld IE (1993) The Pupil: Anatomy Physiology and Clinical Applications. Iowa State University Press, AmesGoogle Scholar
- May PJ, Reiner A, Gamlin PD (2019b) Autonomic regulation of the eye. Oxford research encyclopedia, neuroscience; neuroendocrine and autonomic systems. Oxford University Press, Oxford, pp 1–27Google Scholar
- McDougal DH, Gamlin PD (2015) Autonomic control of the eye. Comp Physiol 5:439–473Google Scholar
- Samuels ER, Szabadi E (2008) Functional neuroanatomy of the noradrenergic locus coeruleus: its roles in the regulation of arousal and autonomic function part II: physiological and pharmacological manipulations and pathological alterations of locus coeruleus activity in humans. Curr Neuropharmacol 6:254–285PubMedPubMedCentralCrossRefGoogle Scholar
- Sun W, May PJ (1995) Morphology and connections of the pupillary light reflex pathway in cat and monkey. Invest Ophthamal Vis Sci 36:S12Google Scholar
Open AccessThis article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article's Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article's Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/licenses/by/4.0/.