The caudal part of the macaque ventrolateral prefrontal (VLPF) cortex hosts several distinct areas or fields—45B, 45A, 8r, caudal 46vc, and caudal 12r—connected to the frontal eye field (area 8/FEF). To assess whether these areas/fields also display subcortical projections possibly mediating a role in controlling oculomotor behavior, we examined their descending projections, based on anterograde tracer injections in each area/field, and compared them with those of area 8/FEF. All the studied areas/fields displayed projections to brainstem preoculomotor structures, precerebellar centers, and striatal sectors that are also targets of projections originating from area 8/FEF. Specifically, these projections involved: 1) the intermediate and superficial layers of the superior colliculus; 2) the mesencephalic and pontine reticular formation; 3) the dorsomedial and lateral pontine nuclei and the reticularis tegmenti pontis; and 4) the body of the caudate nucleus. Furthermore, area 45B projected also to the regions around the trochlear nucleus and to the raphe interpositus. The present data provide evidence for a role of the caudal VLPF areas/fields in controlling oculomotor behavior not only through their connections to area 8/FEF, but also in parallel through a direct access to preoculomotor brainstem structures and to the cerebellar and basal ganglia oculomotor loops.
The caudal part of the ventrolateral prefrontal cortex (VLPF) is connected to the frontal eye field (area 8/FEF; e.g., Huerta et al. 1987; Stanton et al. 1993, 1995; Schall et al. 1995). Recent studies (Gerbella et al. 2010; Gerbella, Borra et al. 2013a) have shown that this prefrontal region hosts several connectionally distinct areas or fields (Fig. 1): area 8r, located rostral to area 8/FEF; the caudalmost part of ventrocaudal area 46 (caudal 46vc); area 45B, located in the prearcuate bank ventral to area 8/FEF; area 45A, located on the rostrally adjacent convexity cortex; and the caudal part of area 12r. These studies showed that all of these areas are connected not only to area 8/FEF but also to the supplementary eye field (SEF), and some of them (8r, caudal 46vc, and 45B) are also connected to the lateral intraparietal (LIP) area. As area 8/FEF, the SEF and area LIP are well known to be important nodes in the cortical network for initiating and controlling voluntary eye movement (see, e.g., Lynch and Tian 2006), these data suggest that the caudal VLPF areas take part in this network, thus playing a role in controlling oculomotor behavior. Indeed, the role of a cortical sector including areas 8r and caudal 46vc in controlling visually and memory-guided saccades has already been well assessed (e.g., Funahashi et al. 1993; Takeda and Funahashi 2002; Watanabe et al. 2006; Kuwajima and Sawaguchi 2007). Conversely, a possible role in oculomotor control of the more ventral areas 45B, 45A, and caudal 12r has not yet been systematically investigated.
It is possible that caudal VLPF areas could exert influence on oculomotor control not only through area 8/FEF, but also through direct projections to oculomotor brainstem structures. Indeed, indirect data, based on tracer injections in the superior colliculus (SC; Leichnetz et al. 1981; Fries 1984; Lock et al. 2003), showed that the caudal VLPF is a source of corticotectal projections. Furthermore, recent studies (Koval et al. 2011; Johnston et al. 2013) have shown that inactivation of the caudal part of area 46, including area 46v, affects saccade-related neural activity in the SC.
The subcortical projections of the caudal VLPF areas are still virtually unknown. In contrast, much is known about the descending pathways possibly mediating the role of area 8/FEF in initiating and controlling eye movements (see, e.g., Lynch and Tian 2006). Specifically, area 8/FEF is a source of 1) projections to the SC and to other brainstem preoculomotor centers (e.g., Huerta et al. 1986; Stanton et al. 1988b); 2) cortico-striate projections, which are at the root of the so-called oculomotor basal ganglia circuit (e.g., Alexander and DeLong 1985; Hikosaka et al. 1989); and 3) cortico-pontine projections, which ultimately target oculomotor-related zones of the cerebellum (e.g., Thier and Möck 2006).
The aim of the present study was to assess whether caudal VLPF areas display subcortical projections possibly mediating a role in controlling oculomotor behavior in parallel with area 8/FEF. To this purpose, we examined the descending projections of these areas based on injections of anterograde tracers confined to individual areas/fields and compared them with those of area 8/FEF. Parts of this article have been published previously in an abstract form (Luppino et al. 2012).
Materials and Methods
Subjects, Surgical Procedures, and Selection of Injection Sites
The present study is based on results from injections of anterograde neural tracers placed in the architectonic areas 8/FEF, 8r, 45B, and 45A, and in the caudal part of areas 46vc and 12r (Fig. 2 and Table 1) in 7 macaque monkeys (5 Macaca mulatta and 2 Macaca fascicularis). All these cases, except for the tracer injection in area 8r in Case 56l, have already been used in previous studies focused on the cortical connectivity of these areas (Gerbella et al. 2010; Gerbella, Borra et al. 2013a; Borra et al. 2011). The animal handling as well as surgical and experimental procedures complied with the European guidelines (86/609/ EEC and 2003/65/EC Directives) and Italian laws in force on the care and use of laboratory animals, and were approved by the Veterinarian Animal Care and Use Committee of the University of Parma and authorized by the Italian Health Ministry.
|35||M. mulatta||R||8/FEFa||FR 10%||1 × 1|
|36||M. fascicularis||R||8/FEFa||FR 10%||1 × 1|
|R||45Ba||BDA 10%||1 × 1|
|37||M. mulatta||R||45Ba||FR 10%||1 × 1|
|R||45Aa||BDA 10%||1 × 1|
|39||M. fascicularis||L||45Aa||FR 10%||2 × 1|
|43||M. mulatta||R||46vcc||FR 10%||1 × 1|
|48||M. mulatta||L||12rb||LYD 10%||1 × 1.3|
|56||M. mulatta||L||8r||FR 10%||2 × 1|
|35||M. mulatta||R||8/FEFa||FR 10%||1 × 1|
|36||M. fascicularis||R||8/FEFa||FR 10%||1 × 1|
|R||45Ba||BDA 10%||1 × 1|
|37||M. mulatta||R||45Ba||FR 10%||1 × 1|
|R||45Aa||BDA 10%||1 × 1|
|39||M. fascicularis||L||45Aa||FR 10%||2 × 1|
|43||M. mulatta||R||46vcc||FR 10%||1 × 1|
|48||M. mulatta||L||12rb||LYD 10%||1 × 1.3|
|56||M. mulatta||L||8r||FR 10%||2 × 1|
Under general anesthesia and aseptic conditions, each animal was placed in a stereotaxic apparatus, and an incision was made in the scalp. The skull was trephined to remove the bone overlying the target region, and the dura was opened to expose the VLPF. The criteria for the selection of the injection sites have been described in detail in previous studies (see Table 1). After the tracer injections were placed, the dural flap was sutured, the bone was replaced, and the superficial tissues were sutured in layers. During surgery, hydration was maintained with saline, and temperature was maintained using a heating pad. Heart rate, blood pressure, respiratory depth, and body temperature were continuously monitored. Upon recovery from anesthesia, the animals were returned to their home cages and closely monitored. Dexamethasone and prophylactic broad-spectrum antibiotics were administered pre- and postoperatively. Analgesics were also administered intra- and postoperatively.
Tracer Injections and Histological Procedures
Once the appropriate site was chosen, the mostly anterograde tracer biotinylated dextran amine [BDA; dextran molecular weight (MW) 10 000, 10% 0.1 M phosphate buffer, pH 7.4; Invitrogen] and the retro-anterograde tracer dextran (MW 10 000) conjugated with tetramethylrhodamine [Fluoro-Ruby (FR), 10% 0.1 M phosphate buffer, pH 7.4; Invitrogen] or with lucifer yellow [Lucifer Yellow Dextrane (LYD), 10% in 0.1 M phosphate buffer, pH 7.4; Invitrogen-Molecular Probes] were slowly pressure-injected through a glass micropipette (tip diameter: 50–100 µm) attached to a 5- or 10-µL Hamilton microsyringe (Reno, NV), positioned with a stereotaxic holder. Table 1 summarizes the locations of the injections, the injected tracers, and their amounts.
Twenty-eight days following the injections, each animal was deeply anesthetized with an overdose of sodium thiopental and perfused consecutively with saline, 3.5–4% paraformaldehyde, and 5% glycerol, prepared in 0.1 M phosphate buffer and pH 7.4, through the left cardiac ventricle. Each brain was then blocked coronally on a stereotaxic apparatus, removed from the skull, photographed, and placed in 10% buffered glycerol for 3 days and 20% buffered glycerol for 4 days. Finally, each brain was cut, frozen, into coronal sections of 60-μm thickness.
As in all the subjects the controlateral hemisphere was used for injecting retrograde fluorescent tracers (Gerbella et al. 2010; Borra et al. 2011; Gerbella, Borra et al. 2013a), for technical reasons, after cutting, the 2 hemispheres were separated and only the sections through the ipsilateral hemisphere, together with those through the entire brainstem, were processed for the visualization of the BDA, FR, and LYD labeling. In Case 35r FR, some sections through the mesencephalon were damaged and the superficial layers of the caudal part of the SC were missing.
In Cases 36r and 37r, one series of each fifth section was processed for the visualization of BDA (incubation 60 h), using a Vectastain ABC kit (Vector Laboratories, Burlingame, CA) and 3,3′-diaminobenzidine (DAB) as a chromogen. The reaction product was intensified with cobalt chloride and nickel ammonium sulfate. In these 2 cases, as well as in Cases 35r, 39l, 43r, 48l, and 56l, in which BDA was injected in other areas that were not the object of the present study, one series of each fifth section was processed to visualize FR and BDA, or LYD and BDA, using the double-labeling protocol described in detail in Gerbella et al. (2010). Briefly, the sections were first processed to visualize BDA, with a shorter (i.e., overnight) incubation period in the ABC solution, and BDA was then stained brown using DAB. The sections were then incubated overnight in avidin-biotin blocking reagent (Vector SP-2001), for 72 h at 4 °C in a primary antibody solution of rabbit anti-FR or rabbit anti-LY (1:3000; Invitrogen) in 0.3% Triton, 5% normal goat serum in PBS, and for 1 h in biotinylated secondary antibody (1:200, Vector) in 0.3% Triton, 5% normal goat serum in PBS. Finally, FR or LYD labeling was visualized using the Vectastain ABC kit (Vector Laboratories) and the Vector SG peroxidase substrate kit (SK-4700, Vector) as a chromogen. With this procedure, BDA labeling was stained brown, and the FR or the LYD labeling was stained blue in the same tissue sections. In all cases, one series of each fifth section was stained using the Nissl method (0.1% thionin in 0.1 M acetate buffer, pH 3.7).
The distribution of the anterograde labeling in the subcortical structures was analyzed and plotted in sections every 300 µm, together with the outlines of the section, the bundles of fibers, the aqueduct and ventriculi, and the blood vessels. The attribution of the labeling to the different structures was made by superimposing adjacent Nissl-stained sections on the plots of the labeling, with the aid of a microprojector. Differences in shrinkage between adjacent, differentially processed sections were corrected by slightly changing the magnification of the microprojector. The various subcortical structures were designated mostly according to the nomenclature of the atlas of Paxinos et al. (2000), and SC layers were defined according to May (2006).
As has been the case in previous studies (Borra et al. 2010, 2012), in order to visualize the distribution of the BDA, FR, and LYD anterogradely labeled terminals in the midbrain and the brainstem, a grid pattern (100 × 100 µm) was overlaid on high magnification digitalized images of the sections acquired with a 20× objective through a digital camera incorporated into the microscope with an automatic acquisition system (NisElement; Nikon Co., Tokyo, Japan). A marker was then placed in a square of the grid when at least 3 labeled terminals were located within the square. After preliminary analysis, we defined 2 types of marked squares, based on the number of terminals located within: squares displaying 3–8 labeled terminals, and squares displaying >8 labeled terminals. These two types reflected relatively sparser (e.g., en-passant varicosities) and relatively richer (e.g., clusters of synaptic endings) labeling observed in 100 × 100 × 60 µm tissue voxels (Supplementary Fig. 1).
In the striatum and in the subthalamic nucleus, the labeling was organized in patches, always visible at low magnification, where labeled terminals were very dense. Thus, in order to visualize the distribution of the observed cortico-striatal projections, as in other studies (e.g., Selemon and Goldman-Rakic 1985; Parthasarathy et al. 1992; Calzavara et al. 2007), we extracted the labeling from digitalized photographs taken with a ×10 objective. Specifically, using Adobe Photoshop (Adobe Systems, Inc., San Jose, CA), the images were converted into a black-and-white image applying a threshold appropriate to separate the labeling, stained in black or blue, from the lighter background. The threshold was the same for all the images from a single case. Indeed, to avoid possible sources of variability among sections from the same case, all the sections selected for one type of immunostaining were processed in the same solutions.
The distribution of the labeling in the SC was also visualized in 2D reconstructions as in a previous study (Borra et al. 2012). Specifically, in each plotted section every 600 µm, the SC was flattened at the level of a virtual line running in the mediolateral direction along the middle of SC thickness. The flattened sections were then aligned to correspond with the midline, and the labeling was distributed along the space between the 2 consecutive sections.
In order to collect information on the possible participation in oculomotor control of the caudal VLPF areas—connected to area 8/FEF—through subcortical projections, the distribution of the anterograde labeling observed after tracer injections in each caudal VLPF area or field was compared with that observed after tracer injections in area 8/FEF. The results showed that all the caudal VLPF areas/fields under study displayed projections to brainstem preoculomotor structures, precerebellar centers, and striatal sectors, which are major targets of the descending projections of area 8/FEF.
Projections to Brainstem Preoculomotor and Precerebellar Structures
After tracer injections in area 8/FEF (Cases 36r FR and 35r FR), we observed projections to several brainstem preoculomotor structures (Fig. 3, Supplementary Fig. 2, and Table 2). In both cases, the main projection focus was the SC. The labeling involved more extensively the intermediate layers (stratum griseum intermediale, SGI), but also the superficial layers, considering both the stratum opticum (SO) and the stratum griseum superficiale (SGS), and was very limited in the deep layers (Supplementary Fig. 3A). After the more ventral injection (Case 36r FR), the labeling extended mostly in the middle of the SC for almost the entire rostrocaudal extent, except for the caudalmost part (Fig. 3C,D,E, and Supplementary Fig. 3B). After the more dorsal injection (Case 35r FR), the labeling was more concentrated in the caudal SC (Supplementary Fig. 2E). Labeled terminals were also observed in the contralateral SC, mainly laterally in the SGI (not shown), at the rostrocaudal levels where the labeling in the ipsilateral SC was richer. Furthermore, in the mesencephalon, labeled terminals were also located in the nucleus of Darkschewitsch (Fig. 3A and Supplementary Fig. 2A), in the ventral periaqueductal gray above the oculomotor (III) and trochlear (IV) nuclei, in the mesencephalic reticular formation (MRF) near the interstitial nucleus of Cajal, and extending ventrolaterally at more caudal levels (Fig. 3A–D and Supplementary Fig. 2A–D), and also in the ventrally adjacent pontine reticular formation (PnO; Fig. 3D,E and Supplementary Fig. 2C–E). After the more dorsal injection, labeled terminals were also observed in the pretectal region (Supplementary Fig. 2B). Finally, in the pons, the anterograde labeling involved the pontine raphe, the region around the abducens (VI) nucleus, and the nucleus prepositus hypoglossi (PH; Fig. 3F–H and Supplementary Fig. 2F,G).
p-IV, peri-trochlear zone; p-VI, peri-abducens zone; rMRF, rostral MRF; STN, subthalamic nucleus.
Labeling observed in one out of two cases.
Dense labeling was also observed in precerebellar structures. Specifically, dense spots of labeled terminals were located mainly in the rostral mediodorsal and, to a lesser extent, in the dorsolateral pontine nuclei (PN), and in the reticularis tegmenti pontis (RtTg), mainly ipsilaterally but also controlaterally, especially after the dorsal injection (Fig. 3A–E and Supplementary Fig. 2A–E). Labeled terminals were also observed in the medial part of the parvicellular red nucleus (RNp), especially after the more ventral injection (Fig. 3B).
After tracer injections in all the caudal VLPF areas/fields under study, the anterograde labeling was found in several brainstem structures that are also targets of projections from area 8/FEF (Figs 4–8, Supplementary Figs 4 and 5, and Table 2). In general, the labeling in these structures appeared to be weaker than that observed after the tracer injections in area 8/FEF. Note, however, that the density of the labeling observed after injections of different tracers and even of the same tracer can be affected by several factors intrinsic to the tract-tracing experimental approach (e.g., differences in amount, spread, and sensitivity of injected tracers). Accordingly, the present data do not allow us to make any reliable quantitative comparison between the projections originating from the various studied areas.
First, in all cases, relatively rich anterograde labeling was observed in the ipsilateral SC where some variability was observed across the different cases in the laminar distribution of the labeling (Supplementary Fig. 3A). In general, the highest proportion of labeled voxels was located in the SGI, especially after the injection in caudal area 46vc. However, in all the cases the labeling also involved the superficial layers. Specifically, a significant proportion of labeled voxels was observed in the SO, especially after injections in areas 45A, 8r, and 1 of the 2 injections in area 45B, and in the SGS, especially after the injection in area 8r. After injections in area 45B and caudal areas 46vc and 12r, the topographic distribution of the labeling in the SC tended to be similar to that observed after the ventral injection in area 8/FEF (Figs 4C–E, 7B,C, 8A–D, and Supplementary Figs 3B and 4C,D), whereas after the injection in area 8r the labeling tended to be richer in the caudal half (Fig. 6C,D). After the injections in area 45A, the labeling tended to be concentrated relatively more rostrally in Case 39l FR (Supplementary Fig. 5C,D) and relatively more caudally in Case 37r BDA (Fig. 5B–D and Supplementary Fig. 3B), suggesting a possible topographic organization of the corticotectal projections of this area. Much weaker labeling was observed in the controlateral SC at the rostrocaudal levels where the labeling in the ipsilateral SC was richer.
Second, in all cases, anterograde labeling was consistently located in the MRF and in the ventrally adjacent PnO. Specifically, after injections in areas 45A and 8r, the labeling in the MRF tended to be located in the lateral most part, also involving the parabigeminal nucleus (Figs 5A–C, 6B,C, and Supplementary Fig. 5A–C). Furthermore, after injections in area 45B, labeled terminals were consistently observed in the rostral MRF, in the region around the III and IV nuclei and in the pontine raphe (Fig. 4A–C,F and Supplementary Fig. 4A,B,E), while after injection in area 8r, some anterogradely labeled terminals were found around the VI nucleus and in the PH (Fig. 6E,F).
Finally, all the areas under study were a source of projections to the PN and RtTg. In the PN, dense spots of anterograde labeling were located in the rostral mediodorsal and the dorsolateral parts in all cases. These projections appeared relatively rich after injections in area 45B (Fig. 4A–E and Supplementary Fig. 4A–C), relatively weak after injections in areas 45A and 8r (Figs 5A–C and 6A,B and Supplementary Fig. 5A–C), and very weak after injections in caudal areas 46vc and 12r (Figs 7A,B and 8A,B,D).
Projections to the Basal Ganglia
After tracer injections were placed more ventrally and more dorsally in area 8/FEF (Cases 36r FR and 35r FR, respectively), dense patches of labeled terminals were found in the striatum showing a very similar distribution (Fig. 9 and Supplementary Fig. 6). In both cases, the most densely labeled territory was located in the middle part of the caudate body, extending for several millimeters in a rostrocaudal direction from ∼2 to ∼10 mm caudal to the anterior commissure (AC). Within this caudate sector, the dense patches of labeled terminals tended to be located laterally with some variability in their mediolateral extent across the 2 cases. Smaller, less-dense clusters of terminals were located more rostrally, up to 2 mm rostral to the AC. In the putamen, dense patches of labeled terminals were located in the caudalmost part, and a smaller and less-dense labeled zone was observed dorsomedially, at ∼4–8 mm caudal to the AC. Additional anterograde labeling was observed in both cases (Case 36r FR shown in Fig. 10) in the ventrolateral part of the subthalamic nucleus, mostly in the caudal half. Finally, in both cases clusters of labeled terminals were observed in the dorsolateral part of the substantia nigra pars compacta (SNpc; Fig. 3A,B and Supplementary Figs 2A,B and 7).
After tracer injections in all the various caudal VLPF areas/fields under study, dense patches of anterograde labeling were observed in the caudate, which mostly involved the same sector labeled after tracer injections in area 8/FEF (Figs 9, 11, and 12 and Supplementary Fig. 8). Specifically, the densest labeling extended from about 1 mm rostral to about 7 mm caudal to the AC for caudal areas 46vc and 12r (Fig. 12), and to about 10 mm caudal to the AC for areas 8r, 45B, and 45A (Figs 9 and 11 and Supplementary Fig. 8). Within this sector, some small variations in dorsoventral and mediolateral location of the dense patches of labeled terminals were observed, even between cases of tracer injections in the same area. In some cases, these patches appeared to very closely match the location of the patches of terminals labeled after tracer injections in area 8/FEF, whereas, in other cases, they appeared to be located slightly more medially and/or ventrally. In all the cases, less-dense anterograde labeling was located more rostrally up to about 2.5 mm rostral to the AC, where the projections from area 8/FEF appeared quite weaker. Some small, sparser clusters of labeled terminals were also found in the caudal putamen after injections in areas 8r, 45B, and 45A (Figs 9 and 11 and Supplementary Fig. 8). After both the tracer injections in area 45A, the tracer injection in caudal area 12r and 1 of the 2 tracer injections in area 45B anterograde labeling was also located in the caudate tail (Figs 11 and 12 and Supplementary Fig. 8). Additional labeling was found in the ventrolateral part of the subthalamic nucleus after both the tracer injections in area 45B and 1 of the 2 injections in area 45A (Fig. 10). Only very few, if any, sparse labeled terminals were observed in the other cases. Finally, after injections in areas 45B and 45A and, more sparsely, after injections in caudal area 12r, clusters of labeled terminals were observed in the SNpc (Supplementary Fig. 7).
The present study showed that the caudal VLPF areas/fields 8r, 45B, 45A, caudal 46vc, and caudal 12r display subcortical projections that, for many aspects, resemble those originating from area 8/FEF. Accordingly, the present data provide evidence for these areas'/fields' role in controlling oculomotor behavior not only through their connections to area 8/FEF, but also through direct access to preoculomotor brainstem structures and to the cerebellar and basal ganglia oculomotor loops.
Descending Projections of Caudal VLPF Areas
The present data are the first to describe the descending projections originating from areas 8r, 45B, 45A, caudal 46vc, and caudal 12r, as these areas or fields have been only recently defined as distinct cortical entities (Gerbella et al. 2007, 2010; Gerbella, Borra et al. 2013a; Borra et al. 2011). Evidence for caudal VLPF projections to the SC has thus far been available only from indirect data (Leichnetz et al. 1981; Fries 1984; Lock et al. 2003). Furthermore, projections to the caudate body have been observed by Yeterian and Pandya (1991) after a tracer injection in the ventral prearcuate cortex, and by Yeterian and Van Hoesen (1978) after tracer injections involving the location of areas 46v, 12r, and 45A.
Our data from tracer injections in area 8/FEF, used in the present study as a framework of reference, were in substantial agreement with those from other studies, in which tracer injections were placed either in the ventral or in the dorsal part of this area (Huerta et al. 1986; Stanton et al. 1988a, 1988b; Shook et al. 1990, 1991; Yeterian and Pandya 1991; Parthasarthy et al. 1992; Cui et al. 2003), where small and large saccades are represented, respectively (Bruce et al. 1985). Some differences from the above-mentioned studies' results may be attributed to the use in the present study of more sensitive anterograde tracers. Specifically, the labeling in the pretectal area observed in the present study, especially after the dorsal area 8/FEF injection, could not be attributed to true anterograde labeling or to fibers by Stanton et al. (1988b). Furthermore, we were able to observe labeled terminals in the SNpc after injections either in area 8/FEF, in which they were not observed in other studies, or in areas 45B, 45A, and caudal area 12r.
Recent connectional data (Gerbella et al. 2010; Gerbella, Borra et al. 2013a; Borra et al. 2011) have provided evidence for the affiliation of the caudal VLPF areas with the oculomotor cortical network. First, all these areas/fields display a differential pattern of connectivity to area 8/FEF: area 8r and caudal area 46vc are connected to the entire extent of area 8/FEF, area 45B and caudal area 12r are mostly connected to the ventral part of this area, and area 45A is mostly connected to the dorsal part. Furthermore, all these areas/fields are also connected to the SEF, and areas 8r, 45B, and caudal area 46vc are also connected to area LIP.
In agreement with these data, functional studies have shown that the cortical sector corresponding to area 8r and caudal area 46vc appears to play a role in the generation and control of visually guided and memory-guided saccades (e.g., Funahashi et al. 1993; Takeda and Funahashi 2002; Watanabe et al. 2006; Kuwajima and Sawaguchi 2007), and in controlling vergence and ocular accommodation (Gamlin and Yoon 2000). Indeed, this cortical sector has already been included in the oculomotor cortical network, and is part of the so-called prefrontal eye field (Lynch and Tian 2006). The role of the more ventral areas 45B, 45A, and caudal area 12r in oculomotor control has not yet been systematically investigated. Though the functional properties of area 45B are still poorly understood, we have suggested, based on connectional data, that this area is a “preoculomotor” area, in which rostral prefrontal, orbitofrontal, inferotemporal, and amygdalar inputs guide the exploration of visual scenes for the perception of objects, actions, and faces (Gerbella et al. 2010; Gerbella, Baccarini et al. 2013b). Area 45A is connected to higher order auditory and multisensory areas of the superior temporal gyrus (Gerbella et al. 2010). Functional data, showing that this sector is involved in multisensory processing of communication stimuli (see Romanski and Averbeck 2009) and activates during actions and faces observation (Nelissen et al. 2005; Tsao et al. 2008), suggested a role in communication behavior for this area. The connections of area 45A to the dorsal part of area 8/FEF and to the amygdala (Gerbella et al. 2010; Gerbella, Baccarini et al. 2013b) could represent the neural substrate of the role in communication behavior of gaze direction, an important communicative signal in social interactions (e.g., Emery 2000; Ghazanfar et al. 2006).
The present study shows that all of the various caudal VLPF areas/fields are sources of projections to brainstem pre-oculomotor structures, albeit to variable extents; this suggests for them a role in oculomotor control in parallel with area 8/FEF. Indeed, all these areas/fields are sources of projections to the intermediate layers of the SC, in which neurons are activated prior to saccadic eye movements and project to premotor regions in the brainstem oculomotor system (Sparks 2002). These SC layers are a well-known major target of projections from area 8/FEF, the SEF, and area LIP, exerting an excitatory drive considered to play a crucial role in controlling voluntary gaze shifts (see, e.g., Segraves and Goldberg 1987; Lynch and Tian 2006; Wurtz 2009). Furthermore, corticotectal projections from caudal VLPF areas/fields, similarly to area 8/FEF, also involve the SC superficial layers, suggesting a role in modulating visual information processing at this level. Our data showed that, according to the injected area, corticotectal projections tended to be richer either in the rostral or in the caudal part of the SC, to which the ventral and the dorsal part of area 8/FEF, respectively, mostly project (Stanton et al. 1988b; present data). Interestingly, in all cases, the labelling tended to involve almost the entire medio-lateral extent of the superficial and intermediate SC layers, but substantially avoiding the medialmost and the lateralmost parts. This pattern of corticotectal projections markedly distinguishes caudal VLPF areas from the rostrally adjacent hand-related prefrontal fields (intermediate area 12r and rostral area 46vc), which mostly project to the lateral part of the intermediate and deep SC layers (Borra et al. 2012), and from the rostral part of the VLPF (rostral area 12r and area 46vr), which does not appear to project to the SC (personal observations).
The corticotectal projections from caudal area 46vc described in the present study could explain the changes in neural activity observed in the SC after inactivation of this area (Johnston et al. 2013) and could represent a neural substrate for the proposed role of the caudal periprincipalis region in suppressing unwanted saccades (Pierrot-Deselligny et al. 1991; Sweeney et al. 1996; Condy et al. 2007; Koval et al., 2011; Johnston et al. 2013).
Furthermore, the projections of areas 45B and 8r to other brainstem pre-oculomotor structures that are also targets of projections from area 8/FEF, though apparently relatively weak, suggest a role in oculomotor control relatively close to the motor output. Specifically, area 45B projects to the rostral MRF, involved in controlling vertical saccades (Handel and Glimcher 1997; Waitzmann et al. 2000) and to the RI, likely where omnipause neurons are located (e.g., Büttner-Ennever et al. 1988) and area 8r very weakly projects to the peri-abducens region and the PH, possibly involving the territory of the inhibitory burst neurons (Strassman et al. 1986). Finally, areas 8r, 45A, and 45B (only one case) project to the PBG (and the adjacent MRF), which is a visuo-oculomotor structure (Sherk 1979; Cui and Malpeli 2003) receiving from, and projecting to, the SC (Graybiel 1978; May 2006) and projecting to the lateral geniculate nucleus (Wilson et al. 1995).
A further finding of the present study was that all the caudal VLPF areas/fields project to the PN and RtTg, suggesting their involvement in cerebro-cerebellar loops. Cortico-pontine projections from area 45B and the caudal part of area 46v have been already described by Schmahmann and Pandya (1997). Our data indicate that areas 45A and 8r and caudal area 12r are also sources of cortico-pontine projections. Thus, cortico-pontine projections appear to represent a characterizing connectional feature that distinguishes the caudal from the mid-rostral VLPF (Schmahmann and Pandya 1997; Strick et al. 2009; Borra et al. 2012). Our data showed that cortico-pontine projections from caudal VLPF areas/fields involved the dorsomedial and dorsolateral parts of the PN, that is those parts of the PN hosting saccade-related neurons and targets of projections originating from area 8/FEF, the SEF, and area LIP, as well as from the SC, and projecting to oculomotor regions of the cerebellar cortex (see, e.g., May, 2006; Thier and Möck 2006). These PN sectors also host neurons with smooth-pursuit related activity and, in Cebus monkeys, are targets of projections from the pursuit subregion of the FEF (FEFsem; Lynch and Tian, 2006). Thus, our data strongly suggest that caudal VLPF areas/fields are involved jointly with other cortical oculomotor areas in the cerebellar oculomotor loops for controlling saccadic and smooth pursuit eye movements.
The present study also showed that caudal VLPF areas/fields are sources of robust projections to the caudate. According to the general framework of the organization of the cortico-basal ganglia-cortical loops (basal ganglia circuits; Alexander et al. 1986), the VLPF should project to the caudate head, and thus participate in the “dorsolateral” and the “lateral orbitofrontal” circuits. Interestingly, our data showed that caudal VLPF areas/fields do not project to the caudate head, but to a sector of the caudate body that largely overlaps the target of projections from area 8/FEF, the FEFsem, the SEF, and area LIP (Selemon and Goldmann-Rakic 1985; Stanton et al. 1988a; Shook et al. 1991; Parthasarathy et al. 1992; Cui et al. 2003). This caudate zone hosts neurons displaying saccade-related activity (Hikosaka et al. 1989), and is considered to be the striatal region engaged in the “oculomotor” basal ganglia circuit. Accordingly, our data clearly suggest that caudal VLPF areas/fields, unlike mid-rostral VLPF, participate in the oculomotor basal ganglia circuit. Our data do not allow us to make any firm conclusion on the possible degree of segregation of the striatal projection zones of the caudal VLPF areas and area 8/FEF. Indeed, our data were obtained from different animals and, being based on tracer injections not involving the entire extent of the target area, likely do not show its entire striatal projection fields (see for similar considerations Calzavara et al. 2007). Areas 45A, caudal 12r, and 45B (only one case) also showed projections to the caudate tail. Interestingly, recent data showed that this striatal sector, which is a target of projections from the temporal cortex (Van Hoesen et al. 1981), host neurons involved in guiding saccades based on “what” and “where” object-related visual information (Yamamoto et al. 2012). Furthermore, our data showed that not only area 8/FEF but also areas 45B and 45A (only one case) project to the ventral and caudal parts of the subthalamic nucleus where saccade-related neurons were recorded (Matsumura et al. 1992; Isoda and Hikosaka 2008), suggesting engagement in the oculomotor basal ganglia circuits also through the “hyperdirect” basal ganglia pathway of these areas (Nambu et al. 1996). The present study also showed that areas 8/FEF, 45B, and caudal 12r are sources of projections to the SNpc. To our knowledge, our data are the first to report a direct cortical projection to the SNpc in the macaque. As labelled terminals observed in the present study were mostly well within the limits of this nucleus and all the SNpc neurons are dopaminergic (Halliday et al. 2012), it seems conceivable that these projections directly contact this type of neuronal population, providing evidence for a possible modulatory role of cortical areas on the dopaminergic system.
In conclusion, the present study provides further support for the idea that the macaque caudal VLPF hosts an oculomotor cortical domain in which different areas or fields may be differentially involved in controlling voluntary gaze shifts, not only through connections to other eye-movement-related cortical areas, but also through projections to brainstem oculomotor structures and to the basal ganglia and cerebellar oculomotor loops.
The work was supported by Ministero dell'Istruzione, dell'Università e della Ricerca (grant number: PRIN 2010, 2010MEFNF7_005) and European Commission Grant Cogsystems FP7-250013.
Conflict of Interest: None declared.