Corticostriate Projections from Areas of the “LateralGrasping Network”: Evidence for Multiple Hand-RelatedInput ChannelsMarzio Gerbella1,2, Elena Borra1, Chiara Mangiaracina3, Stefano Rozzi1,and Giuseppe Luppino1
1Dipartimento di Neuroscienze, Sezione di Fisiologia, Università di Parma, Parma 43100, Italy, 2Brain Center forSocial and Motor Cognition, Istituto Italiano di Tecnologia (IIT), Parma 43100, Italy and 3Dipartimento di ScienzeBiomediche, Biotecnologiche e Traslazionali-S.Bi.Bi.T. Università di Parma, Parma 43100, Italy
Address correspondence to Dr Giuseppe Luppino, Dipartimento di Neuroscienze, Sezione di Fisiologia, Università di Parma, Via Volturno 39, I-43100 Parma,Italy. Email: [email protected]
AbstractCorticostriatal projections from the primate cortical motor areas partially overlap in different zones of a large postcommissuralputaminal sector designated as “motor” putamen. These zones are at the origin of parallel basal ganglia-thalamocorticalsubloops involved in modulating the cortical motor output. However, it is still largely unknown how parietal and prefrontalareas, connected to premotor areas, and involved in controlling higher order aspects of motor control, project to the basalganglia. Based on tracer injections at the cortical level, we analyzed the corticostriatal projections of themacaque hand-relatedventrolateral prefrontal, ventral premotor, and inferior parietal areas forming a network for controlling purposeful hand actions(lateral grasping network). The results provided evidence for partial overlap or interweaving of these projections incorrespondence of 2 putaminal zones, distinct from themotor putamen, one located just rostral to the anterior commissure, theother in the caudal and ventral part. Thus, the present data provide evidence for partial overlap or interweaving in specificstriatal zones (input channels) of projections from multiple, even remote, areas taking part in a large-scale functionallyspecialized cortical network. Furthermore, they suggest the presence of multiple hand-related input channels, possiblydifferentially involved in controlling goal-directed hand actions.
IntroductionA series of connectional studies (Rozzi et al. 2006; Borra et al. 2008,2011; Gerbella et al. 2011; Gerbella, Borra, Tonelli, et al. 2013) hasprovided evidence for the involvement of hand-related ventralpremotor (PMv), inferior parietal lobule (IPL), and ventrolateral pre-frontal (VLPF) areas in a cortical network, designated as “lateralgrasping network” (Fig. 1). In this network, the IPL and PMv areasare jointly involved in visuomotor transformations for graspingand the VLPF hand-related areas (Rozzi et al. 2011) possiblycontribute to selecting and organizing hand actions, based on
memory-related information about object properties or identityand information on behavioral goals or behavioral guiding rules.The selected handmotor acts could be then put into action thanksto the connections of PMv area F5p with the primarymotor cortexand subcortical motor centers (Borra et al. 2010).
Cortical control of motor behavior relies on information pro-cessing occurring not only through cortico-cortical connections,but also through the basal ganglia and cerebellar loops
As faras the basal ganglia are concerned, Alexander et al. (1986)have proposed a general model of connectional architecture in
which different cortical regions project to specific striatal territor-ies at the root of largely segregated basal ganglia-thalamocorticalloops. Five major loops were originally identified, designated as“motor,” “oculomotor,” “dorsolateral prefrontal,” “lateral orbito-frontal,” and “limbic,” respectively. Subsequent studies confirmedthis view and, as already suggested by Alexander et al. (1986),showed up a finer modular organization in which each mainloop consists of several largely segregated closed subloops. Ac-cordingly, each subloop originates from, and projects to, individ-ual cortical areas or limited sets of functionally related areas andinvolves distinct, relatively restricted striatal zones, or “inputchannels” (Strick et al. 1995; Middleton and Strick 2000). Basedon their cortical origin and termination, individual subloopscould be then functionally distinct and their definition is thus es-sential for understanding the mode of information processing inthe basal ganglia for different motor and nonmotor functions. Inrecent years, the advent of transneuronal transport of viruseshas greatly contributed to clarify the organization of the basal gan-glia output to the cortical level (see, e.g.,Middleton andStrick 2000;Kelly andStrick 2004). However, incomplete knowledge on the dis-tribution of the corticostriatal projections from every individualarea still precludes a full description of the organization of thebasal ganglia circuitry.
This appears to be the case also for the organization of the“motor” basal ganglia loop. Indeed, it is well assessed (see, e.g.,Nambu 2011; Takada et al. 2013) that agranular frontal and cingu-late motor areas are sources of partially overlapping projectionsto a large putaminal sector located caudal to the level of the anter-ior commissure (AC). In turn, this “motor”putaminal sector is at theroot of parallel, largely segregated reentrant subloops involved inmodulating the cortical motor output for controlling voluntarymotor behavior (see, e.g., Middleton and Strick 2000; Kelly andStrick 2004). However, it is still largely unknown how parietal andprefrontal areas, connected to premotor areas and forming func-tionally specialized networks for controlling higher order aspectsof motor control, project to the basal ganglia. Specifically, it is stillpoorly understood whether projections from these areas convergein specific striatal sectors or involve distinct striatal zones accord-ing to the general topography of the corticostriatal projections.
In the present study, we used the “lateral grasping network”as a model for addressing the issue of how signals related to
higher order aspects of motor control could be conveyed throughthe basal ganglia circuitry. Accordingly, we examined the projec-tions to the basal ganglia labeled after injections of anterogradetracers in the VLPF, PMv, and IPL nodes of the network. Parts ofthis article have been published previously in an abstract form(Gerbella, Borra, Rozzi, et al. 2013).
MethodsSubjects, Surgical Procedures, and Selection ofthe Injection Sites
The present study is based on results from 10 macaque monkeys(6 Macaca mulatta, 2 M. nemestrina, and 2 M. fascicularis), in whichanterograde tracers were injected in VLPF, PMv, and IPL hand-related areas and in the hand field of the primary motor area F1(Table 1). All these cases, except for the tracer injections in Case62l have already been used in previous studies focused on thecortical (Rozzi et al. 2006; Borra et al. 2008, 2011; Gerbella et al.2011; Gerbella, Borra, Tonelli, et al. 2013) and/or corticotectal(Borra et al. 2014) connectivity of the areas under study.
Animal handling, as well as surgical and experimental proce-dures, complied with the European guidelines (86/609/ EEC and2003/65/EC Directives) and Italian laws in force regarding thecare and use of laboratory animals, and were approved by theVeterinarian Animal Care and Use Committee of the Universityof Parma and authorized by the Italian Health Ministry.
Under general anesthesia (Zoletil®, initial dose: 20 mg/kg,i.m.; supplemental: 5–7 mg/kg/h, i.m., or Ketamine, 5 mg/kg i.m.and Medetomidine, 0.08–0.1 mg/kg i.m.) and aseptic conditions,each animal was placed in a stereotaxic apparatus, and an inci-sion was made in the scalp. The skull was trephined to removethe bone overlying the target region, and the dura was opened.The criteria for the selection of the injection sites have beendescribed in detail in previous studies (see Table 1). Briefly, thechoice of the injection sites was based on identified anatomicallandmarks, that is, the principal sulcus, the infraprincipaldimple, the inferior arcuate sulcus, the central sulcus, and theintraparietal sulcus. For the tracer injections in PMv and F1, thearchitectonic map of Belmalih et al. (2009) was also used asframe of reference. In the IPL, the injection sites in anterior intra-parietal (AIP) areawere placed in the lateral bank of the intrapar-ietal sulcus at anteroposterior (AP) stereotaxic levels between −2and 6 (Borra et al. 2008) and those in PFGwere chosen by using asframe of reference the architectonicmap of Gregoriou et al. (2006)referred in terms of stereotaxic coordinates and location of ana-tomical landmarks such as the intraparietal sulcus and the lat-eral fissure. After the tracer injections were placed, the duralflap was sutured, the bone was replaced, and the superficial tis-sueswere sutured in layers. During surgery, hydrationwasmain-tained with saline, and temperature was maintained using aheating pad. Heart rate, blood pressure, respiratory depth, andbody temperature were continuously monitored. Upon recoveryfrom anesthesia, the animals were returned to their homecages and closely monitored. Prophylactic broad-spectrum anti-biotics (Ceftriaxone 80 mg/kg i.m.), cortisonics (Dexamethasone2 mg, i.m.), and analgesics (Ketoprofen, 5 mg/kg, i.m.) were admi-nistered intra- and postoperatively for up to 1 week.
Tracer Injections and Histological Procedures
Once the appropriate site was chosen, biotinylated dextranamine ([BDA] 10 000 molecular weight [MW], 10% 0.1 M phos-phate buffer, pH 7.4; Life Technologies, Eugene, OR, USA), dextran
Figure 1. Summary view of the areas of the lateral grasping network object of the
present study and of their interconnections. C, central sulcus; IA, inferior arcuate
sulcus; IO, inferior occipital; IP, intraparietal sulcus; i12r, intermediate part of area
12r; L, lateral fissure; P, principal sulcus; r46vc, rostral part of ventrocaudal area 46;
SA, superior arcuate sulcus; ST, superior temporal sulcus.
2 | Cerebral Cortex
by guest on June 19, 2015http://cercor.oxfordjournals.org/
conjugated with tetramethylrhodamine (Fluoro-Ruby [FR],10 000 MW, or equal mixture of 10 000 and 3000 MW volumes,10% 0.1 M phosphate buffer, pH 7.4; Life Technologies) or with lu-cifer yellow (Lucifer Yellow Dextrane [LYD], 10 000 MW, 10% in0.1 M phosphate buffer, pH 7.4; Life Technologies), CholeraToxin B subunit, conjugated with Alexa 488 (CTB green, CTBg1% in 0.01 M phosphate buffered saline at pH 7.4, Life Techno-logies) and wheat germ agglutinin-horseradish peroxidase con-jugated (WGA–HRP, 4% in distilled water, Sigma, St Louis, MO,USA) were slowly pressure-injected through a glass micropipette(tip diameter: 50–100 µm) attached to a 1- or 5-µL Hamilton mi-crosyringe (Reno, NV, USA). Table 1 summarizes the locationsof the injections, the injected tracers, and their amounts.
After appropriate periods following the injections (21–28 daysfor BDA, FR, LYD, and CTBg and 2 days forWGA–HRP), each animalwas deeply anesthetized with an overdose of sodium thiopentaland perfused consecutively with saline, 3.5–4% paraformalde-hyde, and 5% glycerol, prepared in 0.1 M phosphate buffer, pH7.4, through the left cardiac ventricle. Each brainwas then blockedcoronally on a stereotaxic apparatus, removed from the skull,photographed, and placed in 10% buffered glycerol for 3 daysand 20%buffered glycerol for 4 days. Finally, all brainswere cut fro-zen into coronal sections of 60-µm thickness, except for Case 62,cut into coronal sections of 50-µm thickness. As in all the subjects,
except Case 62, the controlateral hemisphere was used for inject-ing retrograde fluorescent tracers, for technical reasons, after cut-ting, the 2 hemispheres were separated and only the sectionsthrough the ipsilateral hemisphere were processed for the visual-ization of the BDA, FR, and LYD labeling.
In all cases of BDA injections, one series of each fifth section(each sixth section in Case 62)was processed to visualize BDA (in-cubation 60 h), using a Vectastain ABC kit (Vector Laboratories,Burlingame, CA, USA) and 3,3′-diaminobenzidine (DAB) as achromogen. The reaction product was intensified with cobaltchloride and nickel ammonium sulfate. In all cases of FR, LYD,CTBg injections, in which BDA was also injected in areas understudy or in other cortical areas, one series of each fifth section(each sixth section in Case 62) was processed to visualize FRand BDA, or LYD and BDA, or CTBg and BDA using the double-la-beling 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 BDAwas then stained brown using DAB. The sections were thenincubated overnight in avidin–biotin blocking reagent (VectorSP-2001), for 72 h at 4 °C in a primary antibody solution of rabbitanti-FRor rabbit anti-LY (1:3000; Life Technologies) in 0.3%Triton,or anti-Alexa 488 (1:15 000, Life Technologies), 5% normal goatserum in PBS, and for 1 h in biotinylated secondary antibody
Table 1 Monkey species, localization of the injection sites, and tracers employed in the experiments
Monkey Species (Macaca) Left/right Area Tracer Amount Core sizei (mm)
Case 13 fascicularis R PFGa WGA/HRP 4% 1 × 0.1 µL 1.8 × 1Case 30 nemestrina L AIPb BDA 10% 4 × 1 µL 4 × 1.2
4 × 0.8L F5ag WGA/HRP 4% 1 × 0.1 µL 2.8 × 1.4
Case 31 nemestrina L F5pc FR 10% 1 × 1 µL 3.8 × 2Case 34 fascicularis L F5ac BDA 10% 2 × 1 µL 2 × 0.5
Cortical or corticotectal labeling described in:aRozzi et al. 2006;bBorra et al. 2008;cGerbella et al. 2011;dBorra et al. 2011;fGerbella, Borra, Tonelli, et al. 2013;gBorra et al. 2014.hMix 1/1 of the MW 3000/10 000.iMajor per minor axis of the core of the injection site.
Corticostriate Projections from the Lateral Grasping Network Gerbella et al. | 3
by guest on June 19, 2015http://cercor.oxfordjournals.org/
(1:200, Vector) in 0.3% Triton, 5% normal goat serum in PBS. Final-ly, FR or LYD or CTBg labelingwas visualized using the VectastainABC kit (Vector Laboratories) and the Vector SG peroxidase sub-strate kit (SK-4700, Vector) as a chromogen. With this procedure,BDA labeling was stained brown, and the FR or the LYD or theCTBg labeling was stained blue in the same tissue sections. How-ever, in thismaterial, the BDA labeling, because of the shorter in-cubation period, was less dense than that observed using thestandard 60-h incubation period. In Cases 13r and 30l, one seriesof each fifth section was processed for HRP histochemistry fol-lowing the protocol described by Mesulam (1982). Sections wererinsed in 0.01 M acetate buffer, pH 3.3, and developed at 4°C inan acetate buffer solution, 0.09% sodium ferricyanide, using tet-ramethylbenzidine as chromogen. In all cases, one series of eachfifth section (each sixth section in Case 62) was stained with theNissl method (0.1% thionin in 0.1 M acetate buffer, pH 3.7).
Data Analysis
The distribution of the anterograde labeling in the ipsilateralbasal ganglia was analyzed in sections every 300 µm in all thecases, except for Case 62 in which the analysis was extended tothe contralateral side.
The projection fields in the basal ganglia were typically orga-nized in patches of very dense labeled terminals, surrounded byless densely labeled zones, designated as “focal” and “diffuse”projections, respectively, by Haber et al. (2006). The extent ofthe observed terminal fields and the density of labeled terminalsvaried across cases, even after injections in the same corticalarea. This variability could be accounted for by several factors, in-trinsic to the tract-tracing experimental approach (e.g., differ-ences in amount, spread, and sensitivity of injected tracers),thus precluding any reliable quantitative comparison betweenthe projections originating from the various studied areas.
In the cases of BDA, FR, LYD, and CTBg injections, focal projec-tions were clearly visible even at relatively low magnificationunder bright field illumination (Fig. 2C,D). In these cases, to ob-tain faithful reproductions of the labeling distribution, as inother studies (e.g., Parthasarathy et al. 1992; Calzavara et al.2007; Borra et al. 2013), the distribution of the observed projectionfields was visualized by extracting the labeling from digitalizedphotographs taken with a ×10 objective (Fig. 2A,B). Specifically,using Adobe Photoshop (Adobe Systems Incorporated, San Jose,CA, USA), in each image, the outlines of the basal ganglia andof adjacent structures were delineated on a separate layer.Then, striatal projection fields were selected and converted intoa black-and-white images applying a threshold appropriate to ex-tract the labeling, stained in black or blue, from the lighter back-ground and, in the cases of FR-, LYD-, or CTBg-labeled fields, fromthe brown BDA labeling when present. Comparison with the ori-ginal image ensured that the labeling was accurately extractedand no false positives were included in the image. In Case 62l,to compare the distribution of the FR, LYD, BDA, and CTBg projec-tion fields visualized in adjacent sections, focal projections weredelineated in each section and their outlines superimposed on asingle section, using the borders of the striatal structures andblood vessels for the alignment. In the cases of WGA–HRP injec-tions, in which the labeling was best visualized under dark fieldillumination, the presence of sparse artifacts and uneven bright-ness of the tissue precluded reliable extraction of the projectionsfields with image processing. Accordingly, focal and diffuse pro-jections were plotted qualitatively, together with the outlines ofthe basal ganglia and of adjacent structures, using a computer-based charting system.
Data from individual sections, including outlines of the striat-al structures and of the focal projections, were also imported into3D reconstruction software (Bettio et al. 2001), providing volu-metric reconstructions of the striatum. The distribution in theputamen of the focal projections was then visualized for each in-dividual case in the lateral views of the 3D reconstructions. Inthese views, the caudate labeling was not visualized, as it wasnot possible to distinguish it from the putaminal one. Further-more, as putaminal focal projections typically tended to be orga-nized in an oblique band running from the dorsomedial to theventrolateral direction, 3D reconstructions were sectionedobliquely as follows (Fig. 3). First, in Case 34l, in which at thelevel of the AC the labeling was particularly rich and organizedinto an almost continuous band, a 2-mm-thick oblique slicewas sectioned from the 3D reconstruction to include its entire ex-tent. The plan of this section intersected the midline at 13 mmabove the upper border of the AC with an angle of 35°. Thesesame parameters were then used to obtain oblique slice fromthe 3D reconstructions of all the other cases to visualize the dis-tribution of the focal projections at an as far as possible compar-able level of the striatum. In Case 62l, 1-mm-thick horizontalsections were also obtained from the 3D reconstruction to visual-ize the distribution of the various focal projections fields at differ-ent dorso-ventral levels. Finally, to obtain composite views of theoverall distribution of the focal projections observed after all thetracer injections in a given region, data from coronal and obliquesections taken from every individual case were warped to tem-plate sections by using Adobe Photoshop. As the sections takenat the same level from different cases were very similar, the dis-tortions caused by this warping procedure were generally quitesmall.
ResultsThe upper part of Figure 4 shows a composite view of the locationof the injection sites placed in the hand-related VLPF, PMv, andIPL areas taking part in the lateral grasping network in all thecases except Case 62, shown in the lower part. Photomicrographsof representative injection sites are shown in Supplementary Fig-ure 1. In the VLPF, 2 injection sites involved each of the 2 hand-re-lated sectors (Rozzi et al. 2011), located in the rostral part ofventrocaudal area 46 (r46vc) and in the intermediate part ofarea 12r (i12r), respectively, which display consistent connec-tions with areas F5a and AIP and, for r46vc, also area PFG (Borraet al. 2011; Gerbella, Borra, Tonelli, et al. 2013). In the PMv, 2 injec-tions involved area F5a, a visuomotor hand-related field (Theyset al. 2012) tightly connected to F5p and displaying consistentconnections with r46vc, i12r, and with the IPL areas AIP andPFG (Gerbella et al. 2011). Two other tracer injections were placedin area F5p, a visuomotor hand-related area (Raos et al. 2006) con-nected to F5a, AIP, PFG, and to the handfield of the primarymotorarea F1 (Gerbella et al. 2011) and source of corticospinal projec-tions (He et al. 1993; Borra et al. 2010). In the IPL, 2 tracer injec-tions involved PFG, a visuomotor hand-related area (Rozzi et al.2008) consistently connected to F5p, F5a, and r46vc (Rozzi et al.2006). Two other tracer injections involved AIP, a visuomotorhand-related area (Murata et al. 2000) consistently connected toF5p, F5a, r46vc, and i12r (Borra et al. 2008). In Case 62l, the CTBgand the LYD injections involved both the 2 VLPF and the 2 IPLhand-related areas, respectively, and the BDA injection wasconfined to area F5a. Finally, the FR injection was placed in thelocation of the hand field of area F1. The distribution of the corti-co-cortical labeling observed in this casewas fully congruent withthat expected based on our previous studies (Rozzi et al. 2006;
4 | Cerebral Cortex
by guest on June 19, 2015http://cercor.oxfordjournals.org/
Borra et al. 2008, 2011; Gerbella et al. 2011; Gerbella, Borra, Tonelli,et al. 2013) and from studies of the M1 connectivity (e.g.,Muakkassa and Strick 1979). Specifically, the rich retro-antero-grade labeling observed in F5p, confirmed the involvement ofthe M1 hand field by the injection site (Borra et al. 2010).
Corticostriatal Projections from the Lateral GraspingNetwork
All the studied areas were sources of corticostriatal projections. Ingeneral, as described in other studies (e.g., Haber et al. 2006;Calzavara et al. 2007) terminal fields typically consisted of largelydistinct patches of dense labeled terminals (focal projections),often surrounded by irregularly shaped zones in which labeledterminals were much sparser (diffuse projections). The extent ofthe observed terminal fields and the density of labeled terminalsvaried across cases, even after injections in the same cortical area.
The distribution of the striatal terminal fields observed in allthe cases of tracer injections in each studied area is shown inFigures 5–7 and in Supplementary Figures 2–4. The distributionof the putaminal focal projections is also shown in lateral viewsof the 3D reconstructions of the striatum in Figure 8 and Supple-mentary Figure 5. As it will be described in detail below, the top-ography of the observed striatal terminal fields, in general, variedmostly according to the injected region. In contrast, in spite ofsome differences in the extent and density of the labelingobserved across cases, it tended to be very similar for each pairof areas belonging to the same region.
After the injections in r46vc or i12r, terminal fields were ob-served in both the putamen and the caudate. In the putamen,in all the cases, most of the labeling was located in a sector ex-tending in rostrocaudal direction from about 1 mm caudal toabout 3–4 mm rostral to the level of the AC (sections a–d anda′–d′ in Fig. 5 and Supplementary Fig. 2; upper part of Fig. 8 and
Figure 2. (A) Low-power view of a digitalized photomicrograph, taken with a ×10 objective, showing the distribution of the striatal anterograde labeling at about 1.8 mm
rostral to theAC in Case 34l BDA. Dashed boxes indicate the location of the highermagnification views shown in (C) and (D). (B) Same image as in (A) after the delineation of
the outlines of the nuclear structures and extraction of the labeling using Adobe Photoshop as described inMaterials andMethods section. (C andD) Highermagnification
views of patches of dense labeled terminals (focal projections), relatively sharply demarcated from surrounding zones in which labeled terminals were sparser (diffuse
projections), taken from the section shown in (A). Scale bar in (A) applies also to (B). Scale bar in (C) applies also to (D). Cd, caudate nucleus; ic, internal capsule; Put,
putamen.
Corticostriate Projections from the Lateral Grasping Network Gerbella et al. | 5
by guest on June 19, 2015http://cercor.oxfordjournals.org/
Supplementary Fig. 5). At this level, several, relatively smallpatches of focal projections tended to distribute within a slab ofputaminal tissue, running obliquely from the dorsomedial to theventrolateral direction. Within this zone, focal projections variedin their location across cases, even after tracer injections in thesame area, but tended to involve more constantly the ventrolat-eral part, at about 1.5–3 mm rostral to the AC. After the injectionsin r46vc, patches of focal projections were observed also at morerostral levels of the putamen. More caudally, the putamen wasvirtually devoid of labeling, except for a caudal and ventral rela-tively restricted zone in which focal projections were observed inall the cases (sections f and f′ in Fig. 5 and Supplementary Fig. 2;upper part of Fig. 8 and Supplementary Fig. 5). This zone ex-tended in rostrocaudal direction from about 5 mm to about8 mm caudal to the AC. In the caudate, most of the labeling waslocated in a sector extending in rostrocaudal direction fromabout1 mm caudal to about 3–4 mm rostral to the AC (sections a–d anda′–d′ in Fig. 5 and Supplementary Fig. 2). In this caudate sector,focal projections tended to be located in the lateral part, insome cases invading the bridge of striatal tissue connecting thecaudate to the putamen.
After the injections in the PMv areas F5a or F5p, most of thelabeling was located in the putamen, which was extensively
involved from about 3 mm rostral to the AC up to its caudalmostpart (middle part of Fig. 8 and Supplementary Fig. 5). However, inCase 31l FR, the striatal labeling was more extensive and richerthan that observed in Case 35r BDA even though the injectionsites involved approximately the same portion of area F5p. Thisdifference, at least in part, could be accounted for by the relativelylarger size of the injection site in Case 31l FR, which involved awider portion of the postarcuate bank than in Case 35r BDA(Table 1). In all the cases, more rostrally, focal projections tendedto mostly distribute along an oblique slab of putaminal tissuethat appeared to largely correspond to that observed after theVLPF injections (sections a–d and a′–d′ in Fig. 6 and Supplemen-tary Fig. 3). In Cases 34l BDA and 31l FR, in which the overall stri-atal labeling was especially dense, this zone was extensivelyinvolved by largely confluent focal projections. Caudal to theAC, the focal projections extended almost continuously in the lo-cation of the forelimb representation of themotor putamen up tothe caudal end in the case of those from F5p (sections e, e′, and f′in Fig. 6 and Supplementary Fig. 3). Somevariabilitywas observedacross cases in the mediolateral distribution of the focal projec-tions in this putaminal zone. While in all the cases, the medialpartwas constantly labeled,muchmore variablewas the involve-ment of the lateral part. After the injections in F5a, but not after
Figure 3. Sectioning procedure of the 3D reconstructions of the striatum. Based on the distribution of the labeling, a 2-mm-thick plane was first sectioned from the 3D
reconstruction of Case 34l (upper part of the figure). The lower part of the figure shows the position of this plane in the striatum at 3 different rostrocaudal levels
indicated in the re-sliced section in the upper right part of the figure. At the level of the AC (lower left), this plane intersects the midline at 13 mm above the upper
border of the AC with an angle of 35°. These same parameters were then used to obtain similar oblique slices from the 3D reconstructions of all the other cases. For
the sake of comparison, all the striatum reconstructions and sections in this and subsequent figures are shown as right. AC, anterior commissure; GPe, external
globus pallidus; GPi, internal globus pallidus; LV, Lateral Ventricle. Other abbreviations as in Figure 2.
6 | Cerebral Cortex
by guest on June 19, 2015http://cercor.oxfordjournals.org/
those in F5p, focal projections were also observed in a caudal andventral part of the putamen (sections e and f in Fig. 6 and Supple-mentary Fig. 3). This labeled zone appeared to occupy a locationvery similar to that observed after the VLPF injections, thoughmore extended in rostrocaudal direction (middle part of Fig. 8and Supplementary Fig. 5). After tracer injections in F5p, the la-beling in the caudate was very poor whereas, after tracer injec-tions in F5a, some small patches of focal projections wereobserved in a location very similar to that of the projections ob-served after the tracer injections in VLPF (sections a–d and a′–d′in Fig. 6 and Supplementary Fig. 3).
After the tracer injections in PFG or AIP, the distribution of thelabeling was very similar across cases, though more extensive inCase 13r WGA–HRP (PFG injection), possibly because of differ-ences in uptake and transport of the tracer used in this case. Inall cases, virtually all the labeling was located in the putamen,mostly involving 2 distinct zones (lower part of Fig. 8 and
Supplementary Fig. 5). One more rostral zone extended fromabout 1 mm caudal to 3–4 mm rostral to the AC (sections a–dand a′–d′ in Fig. 7 and Supplementary Fig. 4) and appeared tolargely correspond to the rostral putaminal zone labeled afterthe injections in VLPF or PMv areas. However, after the injectionsin PFG some patches of focal projections were observed also atmore rostral levels. The second zone appeared to partially involvethe same caudal and ventral part of the putamen labeled after thetracer injections in the VLPF or F5a (sections f and f′ in Fig. 7 andSupplementary Fig. 4). Few, small patches of focal projectionswere inconstantly observed in the medialmost part of themotor putamen and in the lateralmost part of the caudatebody, caudal to the level of the AC.
To look for possible additional evidence for the involvement ofspecific striatal zones by projections from all, or almost all, theareas under study, oblique sections were re-sliced from the 3Dreconstructions of the striatum in all the cases at approximately
Figure 4. Location of injection sites. Upper part: composite viewof all the injection sitesmappedona template right hemisphere, except for those in Case 62. Each injection
site is numbered and reported, based on anatomical landmarks and stereotaxic coordinates, in the inferior parietal lobule (A), in the posterior bank of the arcuate sulcus
(B), and in the ventrolateral prefrontal cortex (C). Arrows indicate the location of the injections sites in anterior intraparietal (AIP) in the lateral bank of the inferior parietal
sulcus and in r46vc in the ventral bank of PS. Arrowheads in (A) indicate the rostrocaudal extent of area AIP, which does not extend on the lateral surface. Lower part:
location of the injection sites in Cases 62l (shown as a right hemisphere) shown on dorsolateral views of the injected hemisphere and in coronal sections through the
core (shown in black) and the halo (shown in lighter grey). Cg, cingulate sulcus; LO, lateral orbital sulcus; Lu, lunate sulcus; MO, medial orbital sulcus. Other
abbreviations as in Figure 1. Scale bar in (A) applies also to (B–C). Scale bar in (a) applies also to (b–f ).
Corticostriate Projections from the Lateral Grasping Network Gerbella et al. | 7
by guest on June 19, 2015http://cercor.oxfordjournals.org/
the same level. These sections (Fig. 9 and Supplementary Fig. 6)were quite useful in showing that an obliquely oriented putaminalzone, located mostly just rostral the AC and extending for about4–5 mm in rostrocaudal direction, was a target of focal projectionsfrom all the areas under study. These projections tended, withvariability across cases, to be scattered along the entire mediolat-eral extent of the putamen, thoughmore constantly located in themore lateral (and ventral) part. Furthermore, these sectionsshowed that the focal projections observed in the caudal andventral part of the putamen in all the cases, except for those oftracer injections in F5p, appeared to share a common, relativelyrestricted zone located at about 5–8 mm caudal to the AC.
All together, these data suggested overlap or, at least, interweav-ing of the striatal projections from the various nodes of the lateralgrasping network into at least 2 distinct putaminal zones. These 2zoneswill be here referred to as rostral and caudal lateral graspingnetwork input channel, respectively.
To obtainmore direct evidence for these observations, result-ing from comparison of different cases, in Case 62 3 different an-terograde tracers were injected in the VLPF and IPL hand-relatedregions and in the PMv area F5a, respectively. In general, theresults from this case (Figs 10 and 11) were quite comparablewith those described above, obtained after tracer injections inindividual areas. Furthermore, the superimposition of the focal
Figure 5.Drawings of coronal sections through the striatum showing the distribution of the anterograde labeling observed in Case 44r after the LYD injection in i12r (upper
part) and in Case 52r after the BDA injection in r46vc (lower part). The sections are shown in a rostral to caudal order (a–f and a′–f′) and their AP level is indicated in terms of
distance in mm from the AC. Scale bar in a applies also to (b–f ) and to (a′–f′). Abbreviations as in Figures 1–4.
8 | Cerebral Cortex
by guest on June 19, 2015http://cercor.oxfordjournals.org/
projections delineated in adjacent coronal sections (Fig. 10, upperpart), or in the same horizontal or oblique (Fig. 10, lower part)sections re-sliced from the 3D reconstruction provided clear evi-dence for both overlapping and interweaving of the striatal VLPF,PMv, and IPL projections in both the rostral and the caudal lateralgrasping network input channels (see also Fig. 12). In the caudate,as expected from the results obtained after tracer injections in in-dividual areas, the VLPF focal projectionswere themost extensive.Several spots of PMv focal projections were also observed in thelateral part of the caudate body, which in part overlapped withtheVLPFprojections (Fig. 10, coronal sections b and c, andhorizon-tal section a′). Finally, a small spot of IPL focal projections at aboutthe level of the AC overlapped with those from VLPF and PMv.
In this same case, a fourth anterograde tracer was injected inthe hand field of M1. In agreement with previous studies (Stricket al. 1995; Inase et al. 1996; Takada et al. 1998), M1 striatal projec-tion fields were limited to the intermediate-lateral part of the pu-tamen caudal to theAC, corresponding to the hand representationof the motor putamen (Figs 10 and 11).
Case 62 was the only case of the present study in which it waspossible to look also for projections in the contralateral striatum.Striatal projections were observed after all the tracer injections,but, compared with those observed in the ipsilateral side, wereconsiderably weaker, mostly consisting of few, relatively smallpatches of labeled terminals located in striatal territories sub-stantially corresponding to those labeled in the ipsilateral side.
Figure 6.Drawings of coronal sections through the striatum showing the distribution of the anterograde labeling observed in Case 34l after the BDA injection in F5a (upper
part) and in Case 31l after FR injection in F5p (lower part). Conventions and abbreviations as in Figures 1–5.
Corticostriate Projections from the Lateral Grasping Network Gerbella et al. | 9
by guest on June 19, 2015http://cercor.oxfordjournals.org/
Cortico-subthalamic Projections from the LateralGrasping Network
In all the cases of the present study, anterogradely labeled term-inals were also observed in the subthalamic nucleus. However,these projections compared with those observed in the striatum,were considerably weaker. Specifically, relatively dense clustersof labeled terminalswere observedonlyafter the tracer injectionsin PMv areas F5a and F5p and after the tracer injection in F1(Fig. 13). In agreement with previous observations (Monakowet al. 1978; Nambu et al. 1996), projections fromM1, tended to in-volve mostly the dorsolateral part of the subthalamic nucleus.More diffuse was the labeling observed after the injections in
F5a and F5p. After injections in VLPF and IPL areas, only verysparse terminals, mostly located in the midventral part of thesubthalamic nucleus, were observed. The paucity of labeling ob-served after the VLPF injections is in line with data of Monakowet al. (1978).
DiscussionThe present study provides evidence for partial overlap andinterweaving of corticostriatal projections from the VLPF, PMv,and IPL hand-related areas of the lateral grasping network incorrespondence of 2 putaminal zones, located at different
Figure 7.Drawings of coronal sections through the striatum showing the distribution of the anterograde labeling observed in Case 30l after the BDA injection in AIP (upper
part) and in Case 54r after FR injection in PFG (lower part). Conventions and abbreviations as in Figures 1–5.
10 | Cerebral Cortex
by guest on June 19, 2015http://cercor.oxfordjournals.org/
rostrocaudal levels and distinct from the hand-related zones ofthemotor putamen. These 2 zones have been referred in the pre-sent study to as rostral and caudal lateral grasping network inputchannel (Fig. 14). The rostral lateral grasping network input chan-nel occupies a slab of putaminal tissue, running obliquely fromthe dorsomedial to the ventrolateral direction in the midventralpart of the putamen and extending from about 4 mm rostral toabout 1 mm caudal to the AC. This zone was consistently in-volved by focal projections originating from all the areas of thelateral grasping network object of present study. The caudal lat-eral grasping network input channel occupies a caudal and ven-tral putaminal zone extending from about 5 mm to about 8 mmcaudal to the AC. This zone was consistently involved by focalprojections originating from all the areas of the lateral graspingnetwork of the present study, except for F5p.
Organization of the Corticostriatal Projections and InputChannels in the Motor Putamen
Cortical projections are the major source of input to the basalganglia. As these projections represents the first step in the trans-position of the functional cortical map onto this structure,detailed knowledge of their organization is fundamental fordetermining the nature of the information that is conveyed andintegrated through the various basal ganglia-thalamocorticalloops.
Virtually all cortical areas project to the striatum andtheir terminal fields are typically heterogeneous, consisting ofdense patches of terminals surrounded by more sparse term-inals, which in the frontal plane tend to form diagonal band(see, e.g., Selemon and Goldman-Rakic 1985; Parent and Hazrati1995).
Figure 8. Distribution of the focal projections (indicated by shaded areas) observed in the putamen after tracer injections in the hand-related VLPF (upper part), PMv
(middle part), and in IPL (lower part) areas shown in lateral views of the 3D reconstructions of the striatum. In each reconstruction the arrow marks the AP level of the
AC and the dashed lines indicate the level of the sections showed in Figures 5 (VLPF injections), 6 (PMv injections), and 7 (IPL injections). Scale bar in upper left part applies
to all the 3D reconstructions. Abbreviations as in Figures 2 and 3.
Corticostriate Projections from the Lateral Grasping Network Gerbella et al. | 11
by guest on June 19, 2015http://cercor.oxfordjournals.org/
Early studies, based on fiber degeneration techniques, de-scribed a topographic organization of corticostriatal projectionsin which each cortical area projects to its nearest part of the stri-atum (Kemp and Powell 1970). Subsequent studies, based on an-terograde neural tracers, showed that corticostriatal projectionsare much more extensive in the anteroposterior dimension(Goldman and Nauta 1977; Yeterian and Van Hoesen 1978; VanHoesen et al. 1981). Furthermore, Yeterian and Van Hoesen(1978) first proposed that cortical areas having reciprocal corti-co-cortical connections share common zones of termination inthe striatum. Indeed, evidence from dual-tracer experimentsshowed overlap of terminal fields from tightly or even relativelyweakly connected areas (e.g., Parthasarathy et al. 1992; Flahertyand Graybiel 1993). However, overlap of cortical terminal fieldsin the striatum cannot be predicted simply based on cortical con-nectivity patterns, as in some cases terminal fields from inter-connected areas are simply interdigitated or even segregated(e.g., Selemon and Goldman-Rakic 1985). Based on a large num-ber of tracer injections in different prefrontal (orbitofrontal,ventromedial, and dorsolateral), rostral cingulate, and dorsal pre-motor (PMd) areas, Haber et al. (2006) and Calzavara et al. (2007)observed, mostly in the rostral part of the caudate, interweaving
and convergence of focal projections from areas of the same oreven of different (e.g., rostral PMd and area 9) domains, providingan anatomical substrate for integration between different pro-cessing circuits (see Haber 2010). Furthermore, diffuse projec-tions from one area could overlap with focal projections ofanother area, providing an anatomical substrate for modulationof signals broadcasted by focal projections (see Haber 2010).
Accordingly, based on the organization of corticostriatal pro-jections, different striatal zones, or input channels (Strick et al.1995), can be specified by convergence of specific subsets ofcortical inputs reflecting the parallel output organization of thestriatum (Parthasarathy et al. 1992; Middleton and Strick 2000).
These general principles apply also to the organization of thecorticostriatal projections from the various corticospinal frontaland cingulate motor areas to the motor putamen. Specifically,projection fields from M1 and the supplementary motor area(SMA), which are tightly interconnected, are largely segregated,at least in their focal projections, involving mostly the lateraland the medial part of the motor putamen, respectively (Stricket al. 1995; Inase et al. 1996; Takada et al. 1998). However, projec-tions from PMv and PMd cortex, though segregated, appear topartially overlap with those from the SMA (Takada et al. 1998).
Figure 9.Distribution of the striatal focal projections observed after tracer injections in the hand-related VLPF (upper part), PMv (middle part), and in IPL (lower part) areas
shown in 2-mm-thick oblique sections re-sliced from the 3D reconstruction at the level shown in Figure 3. Conventions and abbreviations as in Figures 2, 3, and 8.
12 | Cerebral Cortex
by guest on June 19, 2015http://cercor.oxfordjournals.org/
Furthermore, the projections from the caudal cingulate motorarea appear to partially overlap with those from M1 (Takadaet al. 2001). Altogether, these data suggest that in the motorputamen there are different input channels specified by conver-gence of specific subsets ofmotor areas. In agreement with theseobservations, the present study showed extensive focal projec-tions in the motor putamen after tracer injections in the 2 PMv
areas F5p and F5a. These focal projections, as previously ob-served by Takada et al. (1998), were located mostly in the medialpart of the putamen, which corresponds to the location of theSMA projection field (Strick et al. 1995; Inase et al. 1996; Takadaet al. 1998). Furthermore, they were also observed more laterallyand ventrally, in a location that is ventral to theM1-recipient one,similar to what observed by Strick et al. (1995).
Figure 10.Distribution of the striatal focal projections observed in Case 62l after tracer injections in hand-related VLPF and IPL regions (green and blue lines, respectively),
in area F5a (red lines), and in the handfield of the primarymotor area F1 (yellow lines), shown in coronal sections and in 1-mm-thick horizontal sections anda 2-mm-thick
oblique section re-sliced from the 3D reconstruction. The horizontal sections are shown in dorsal to ventral order (a′–c′). The level at which the coronal and horizontal
sections were taken is indicated in the 3D reconstruction in the right lower part. Abbreviations as in Figures 2 and 3.
Corticostriate Projections from the Lateral Grasping Network Gerbella et al. | 13
by guest on June 19, 2015http://cercor.oxfordjournals.org/
The present study also showed that these PMv focal projec-tions involve a more rostral and, in the case of F5a, a caudo-ven-tral zone—the rostral and caudal lateral grasping network inputchannels, respectively—which both appear to share commonprojections from hand-related VLPF and IPL areas. Accordingly,the PMv focal projections could partially overlap in differentzones of the striatum with focal projections of different corticalareas, thus taking part to different input channels. Our datamostly based on tracer injections in different subjects can onlystrongly suggest partial overlap or at least interweaving of thefocal projections from the studied areas in these 2 input chan-nels. However, this possibility appears evenmore likely consider-ing that the relative narrowness of our tracer injections, the slightvariability in the location of the patches of focal projectionsobserved even after different tracer injections in the same areaand the common observation that larger is the injection site,more extensive is the striatal terminal field (e.g., Parthasarathyet al. 1992; Flaherty andGraybiel 1994; Takada et al. 1998), suggestthat focal projections observed in our study represent an under-estimate of the entire striatal territory devoted to each corticalarea. Furthermore, in our cases of multiple anterograde tracersinjections (Cases 62l and 30l), the distribution of the focalprojections observed in both the rostral and the caudal lateralgrasping network input channels after injections in the VLPFand the IPL hand-related regions and in F5a, provided directevidence for partial overlapping and interweaving of the striatalprojections from the different cortical regions composing thelateral grasping network.
Connectional Specificity of the Rostral and Caudal LateralGrasping Network Input Channels
The present data leave open the possibility that other areas,moreor less directly involved in cortical control of grasping, project to
the 2 above described input channels. However, a careful examin-ation of the available data on the organization of corticostriatalprojections strongly suggests a relatively high degree of specifi-city of the subset of inputs to these 2 putaminal zones.
Indeed, caudal VLPF eye-related areas, even those (caudal 12rand caudal 46vc) located just caudal to the hand-related VLPFfields, mostly project to the caudate body (Borra et al. 2013),which is the input zone of the so-called oculomotor basal gangliacircuit (e.g., Alexander et al. 1986; Hikosaka et al. 1989), whereasmore rostral VLPF areas (personal observations) or dorsolateralprefrontal areas (Yeterian and Pandya 1991; Ferry et al. 2000;Haber et al. 2006; Calzavara et al. 2007) mostly project more ros-trally and dorsally in the putamen and/or to the caudate headand rostral body.
Furthermore, both the rostral and the caudal lateral graspingnetwork input channels appear to be skipped or only marginallyinvolved by the projections from the various frontal and cingu-late motor areas. Specifically, the 2 PMd areas F2 and F7 andarea F6 (pre-SMA) appear to project dorsally to the rostral lateralgrasping network input channel (Takada et al. 1998; Inase et al.1999; Tachibana et al. 2004; Calzavara et al. 2007). This appearsto be true also for the projections from the rostral and the caudalcingulatemotor areas, at least considering only their focal projec-tions (Takada et al. 2001). In contrast, terminal fields fromM1 andSMA are locatedmostly caudal to the rostral lateral grasping net-work input channel and dorsal to the caudal one (Liles andUpdyke 1985; Inase et al. 1996, 1999; Takada et al. 1998).
Finally, previous studies on corticostriatal projections fromposterior parietal areas showed projection to the location ofboth the lateral grasping network input channels only after injec-tions in the rostral part of the IPL, likely involving PFG and, pos-sibly, AIP, but not after tracer injections in caudal IPL or superiorparietal areas (Cavada and Goldman-Rakic 1991; Yeterian andPandya 1993). Accordingly, these 2 input channels do not appear
Figure 11.Distribution of the focal projections observed in Case 62l after tracer injections in hand-related VLPF and IPL regions, in PMv area F5a, and in primarymotor area
F1 shown in lateral views of the 3D reconstruction of the striatum. Other conventions and abbreviations as in Figures 2, 3, and 8.
14 | Cerebral Cortex
by guest on June 19, 2015http://cercor.oxfordjournals.org/
to be targeted by projections from PMd and caudal superior par-ietal areas forming parietofrontal circuits involved in online con-trol of reaching and graspingmovements (Raos et al. 2004; Fattoriet al. 2012). Some labeling in correspondence or close to both the2 input channels was observed by Yeterian and Pandya (1993)after a large tracer injection in the lateral part of the parietal oper-culum, likely partially involving area SII. Thus, it is possible thatarea SII, which is tightly connected to all the areas under study(Rozzi et al. 2006; Borra et al. 2008, 2011; Gerbella et al. 2011;Gerbella, Borra, Tonelli, et al. 2013) is an additional source of
projections to these 2 putaminal zones. Focal projections in theproximity of, or marginally overlapping with, the caudal lateralgrasping network input channel were observed after tracer injec-tions in inferotemporal areas TE and TEO (Webster et al. 1993;Cheng et al. 1997). Interestingly, AIP and i12r are robustly con-nected to a specific sector of area TEa/m located in the lowerbank of the superior temporal sulcus. Thus, it is possible thatthis inferotemporal sector is an additional source of projectionsto this caudal input channel, which, however, needs empiricaldemonstration.
Figure 12. (A–F) Low-power photomicrographs showing the distribution of striatal anterograde labeling observed in Case 62l after tracer injections in hand-related VLPF
region (A and D), in PMv area F5a (B and E), and in hand-related IPL region (C and F). Sections A–C are adjacent and correspond to an AP level 3.3 mm rostral to the AC.
Sections D–F are adjacent and correspond to an AP level 6.3 mm caudal to the AC. (A′–F′) Higher magnification views, taken from the sections shown in (A–F). Arrows
point to the same blood vessel. In the left and right panels, the labeling was visualized with a double-labeling protocol (see Materials and Methods) in which the CTBg
(inA,A′, D, andD′) and the LYD (in C, C′, F, and F′) labeling was stained blue and the BDA labeling was stained brown. In themiddle panels (B, B′, E, and E′), the BDA labeling
was visualized with the much more sensitive standard protocol. Scale bar in (A) applies also to (B–F). Scale bar in (A′) applies also to (B′–F′). Abbreviations as in Figures 2
and 3.
Corticostriate Projections from the Lateral Grasping Network Gerbella et al. | 15
by guest on June 19, 2015http://cercor.oxfordjournals.org/
The differentiation of the motor putamen in distinct input chan-nels provides the substrate for a parallel processing of differentaspects of motor control through the basal ganglia circuitry. In-deed, electrophysiological data showed, in the lateral part ofthe putamen, movement-related neural activity closely resem-bling that of agonist muscles (Liles 1983; Alexander and Crutcher1990). In contrast, neurons in themedial putamen are active dur-ing the preparatory activity, or in relation to complexmovements(Liles 1983; Alexander and Crutcher 1990). Furthermore,microsti-mulation appears to be much more effective in evoking bodymovements when applied in the lateral than in the medial partof the putamen (Nambu et al. 2002). The present study providesevidence for 2 additional striatal input channels characterizedby rather specific subsets of cortical input from hand-relatedVLPF, PMv, and IPL areas. Indeed, hand-related neural activitywas recorded even rostral to themicroexcitablemotor putaminalregion, likely in the location of the rostral grasping zone (Crutcherand DeLong 1984; Alexander and DeLong 1985). In the light of thewell-established role of the basal ganglia in action selection andmotor learning (DeLong andWichmann 2009), future studies willclarify the possible contribution of these newly identified hand-related striatal zones to the neural mechanisms for controllingpurposeful hand actions. Our connectional data indicate thatinput from F5p, which is the only node of the network connected
to both the primary motor area and the spinal cord, distinguishthe rostral from the caudal input channel. Furthermore, it is pos-
sible that the caudal, but not the rostral input channel is a target
of projections from higher order ventral visual stream areas in-
volved in object recognition (see, e.g., Tanaka 1996). Accordingly,
it seems conceivable that these 2 hand-related input channels
are differentially involved in selecting and generating hand
actions based on object properties, contextual information, and
behavioral goals.Interestingly, our data also showed focal projections to the
lateral part of the caudate body from all the studied areas, moreextensive from VLPF and more restricted from PMv and IPL,which in Case 62 overlapped in correspondence of the AP levelof the AC. Thus, these focal projections could involve the caudatesector targeted by projections from oculomotor frontal and par-ietal areas (Selemon and Goldman-Rakic 1985; Stanton et al.1988; Shook et al. 1991; Parthasarathy et al. 1992; Cui et al. 2003;Borra et al. 2013), hosting neurons displaying saccade-relatedactivity (Hikosaka et al. 1989), and considered to be the striatalregion engaged in the “oculomotor” basal ganglia circuit (Alexan-der et al. 1986). In a previous study, we found that VLPF, PMv, andIPL areas of the lateral grasping network are a source of cortico-tectal projections (Borra et al. 2014). These projections couldbroadcast to this oculomotor structure information relatedto hand action goals and object affordances extraction and
Figure 13.Drawings of coronal sections through the subthalamic nucleus showing the distribution of the anterograde labeling observed in Case 34l after the BDA injection
in F5a, in Case 35r after BDA injection in F5p, and in Case 62l after the FR injections in F1. For each case, the AP level is indicated in terms of distance in mm from the AC.
Small arrows point to the field shown in the photomicrographs in the lower part of the figure.
16 | Cerebral Cortex
by guest on June 19, 2015http://cercor.oxfordjournals.org/
selection for the eye–hand coordination necessary for appropri-ate hand–object interactions. The projections to the caudatebody observed in the present study could represent a furtherpossible substrate for the contribution of this information tooculomotor control.
Final Considerations
To our knowledge, the present study is the first in providing evi-dence for partial overlap in specific striatal zones of projectionsfrom multiple, even remote, areas taking part in a large-scalefunctionally specialized cortical network. This complex connec-tional organization raises the question of how information fromthese input channels is then conveyed back to the cortical levelthrough the basal ganglia-thalamocortical circuitry. The internalglobus pallidus (GPi) and the substantia nigra pars reticulata(SNr), which are the major output stations of the basal ganglia,are organized in largely segregated output channels each ofthem projecting via the thalamus to a specific cortical area(Middleton and Strick 2000; Kelly and Strick 2004). Output chan-nels projecting to PMv areas and to area 46 are located in differentparts of the GPi, whereas those projecting to areas 12 and AIPwere identified in different parts of the SNr (Middleton and Strick2002; Kelly and Strick 2004; Clower et al. 2005). Striatal cells pro-jecting to different output channels directed to closely relatedcortical areas can be intermingled (Saga et al. 2011). Furthermore,different striatal zones where projections from single body partrepresentations of M1 and S1 converge, can in turn project to
the same output channel (Flaherty and Graybiel 1994). Finally, re-stricted striatal zones or even individual striatal neurons appearto project in different parts of both the GPi and the SNr (Hedreenand DeLong 1991; Parent and Hazrati 1994; Lévesque and Parent2005). Thus, it is possible that the output from the varioushand-related input channels could be conveyed back to all, orpart, of the areas of the lateral grasping network in the frame-workof a closed loop organization. If this is the case, then the pre-sent data favor a model of cortical-basal ganglia connectivity inwhich signals from a given area are first sent to different striatalzones, where they are integrated with signals from other func-tionally related areas, and then reconverge to the output channelprojecting back to the same area.
In recent years, several human studies based onmagnetic res-onance connectional techniques have suggested, although withtheir limitations in terms of spatial resolution, overlap of corti-costriatal projections from different cortical region (Draganskiet al. 2008; Choi et al. 2012) or from functionally related areas(Oguri et al. 2013; Jung et al. 2014). Furthermore, functional im-aging evidence in humans showed differential involvement ofrostral versus caudal putamen in different aspects of hand ac-tions planning and execution, suggesting that the cortical grasp-ing network should be expanded to include the basal ganglia (seeProdoehl et al. 2009). The present data providing higher reso-lution views of the possible convergence of corticostriatal projec-tions from different anatomically or functionally related areascould be very helpful for the interpretation of the above-men-tioned human data. Furthermore, they provide additional insight
Figure 14. Composite views of the distribution of the striatal focal projections from VLPF (green lines), PMv (red lines), and IPL (blue lines) hand-related areas obtained by
warping the focal projections observed in each individual case to template 1-mm-thick coronal and 2-mm-thick oblique sections. The sections were taken at the levels
indicated in the 3D reconstructions of the striatum shown in the right part of the figure. Overlap of the focal projections from 3 and 2 regions is shown in darker and lighter
orange, respectively. In the upper part, arrows indicate the sources of projections, identified in the present study, to the 2 input channels. Dashed arrows indicate possible
additional source of projections, based on other studies (Yeterian and Pandya 1993; Webster et al. 1993; Cheng et al. 1997). Abbreviations as in Figures 2 and 3.
Corticostriate Projections from the Lateral Grasping Network Gerbella et al. | 17
by guest on June 19, 2015http://cercor.oxfordjournals.org/
on the neural substrate for the contribution of the basal ganglia tothe generation and control of hand actions which could be usefulfor the interpretation of functional and clinical observations inhumans.
Supplementary MaterialSupplementary material can be found at: http://www.cercor.oxfordjournals.org/.
FundingThework was supported by Ministero dell’Istruzione, dell’Univer-sità e della Ricerca (grant number: PRIN 2010, 2010MEFNF7_005),European Commission Grant Cogsystems FP7–250013, and Interu-niversity Attraction Poles (IAP) P7/11.
NotesConflict of Interest: None declared.
ReferencesAlexander GE, Crutcher MD. 1990. Functional architecture of
basal ganglia circuits: neural substrates of parallel processing.Trends Neurosci. 13:266–271.
Alexander GE, DeLong MR. 1985. Microstimulation of the primateneostriatum. II. Somatotopic organization of striatal microex-citable zones and their relation to neuronal response proper-ties. J Neurophysiol. 53:1417–1430.
Alexander GE, DeLongMR, Strick PL. 1986. Parallel organization offunctionally segregated circuits linking basal ganglia and cor-tex. Annu Rev Neurosci. 9:357–381.
Belmalih A, Borra E, Contini M, Gerbella M, Rozzi S, Luppino G.2009. Multimodal architectonic subdivision of the rostralpart (area F5) of the macaque ventral premotor cortex.J Comp Neurol. 512:183–217.
Bettio F, Demelio S, Gobbetti E, Luppino G, Matelli M. 2001.Interactive 3-D reconstruction and visualization of primatescerebral cortex. Program No 728.4 2001 Neuroscience MeetingPlanner. San Diego (CA): Society of Neuroscience.
Borra E, Belmalih A, Calzavara R, Gerbella M, Murata A, Rozzi S,Luppino G. 2008. Cortical connections of the macaque anter-ior intraparietal (AIP) area. Cereb Cortex. 18:1094–1111.
Borra E, Belmalih A, Gerbella M, Rozzi S, Luppino G. 2010. Projec-tions of the hand field of the macaque ventral premotor areaF5 to the brainstem and spinal cord. J Comp Neurol.518:2570–2591.
Borra E, Gerbella M, Rozzi S, Luppino G. 2011. Anatomicalevidence for the involvement of the macaque ventrolateralprefrontal area 12r in controlling goal-directed actions.J Neurosci. 31:12351–12363.
Borra E, Gerbella M, Rozzi S, Luppino G. 2013. Projections fromcaudal ventrolateral prefrontal areas to brainstem preoculo-motor structures and to basal ganglia and cerebellar oculo-motor loops in the macaque. Cereb Cortex. doi:10.1093/cercor/bht265.
Borra E, Gerbella M, Rozzi S, Tonelli S, Luppino G. 2014. Projec-tions to the superior colliculus from inferior parietal, ventralpremotor, and ventrolateral prefrontal areas involved in con-trolling goal-directed hand actions in the macaque. CerebCortex. 24:1054–1065.
Calzavara R, Mailly P, Haber SN. 2007. Relationship between thecorticostriatal terminals from areas 9 and 46, and those
from area 8A, dorsal and rostral premotor cortex and area24c: an anatomical substrate for cognition to action. Eur JNeurosci. 26:2005–2024.
Cavada C, Goldman-Rakic PS. 1991. Topographic segregation ofcorticostriatal projections from posterior parietal subdivi-sions in the macaque monkey. Neuroscience. 42:683–696.
Cheng K, SaleemKS, Tanaka K. 1997. Organization of corticostria-tal and corticoamygdalar projections arising from the anteriorinferotemporal area TE of the macaque monkey: a Phaseolusvulgaris leucoagglutinin study. J Neurosci. 17:7902–7925.
Clower DM, Dum RP, Strick PL. 2005. Basal ganglia and cerebellarinputs to “AIP”. Cereb Cortex. 15:913–920.
Choi EY, Yeo BT, Buckner RL. 2012. The organization of thehuman striatum estimated by intrinsic functional connectiv-ity. J Neurophysiol. 108:2242–2263.
Crutcher MD, DeLong MR. 1984. Single cell studies of the primateputamen. I. Functional organization. Exp Brain Res.53:233–243.
Cui DM, Yan YJ, Lynch JC. 2003. Pursuit subregion of the frontaleye field projects to the caudate nucleus in monkeys.J Neurophysiol. 89:2678–2684.
DeLongM,Wichmann T. 2009. Update onmodels of basal gangliafunction and dysfunction. Parkinsonism Relat Disord. 15(Suppl 3):S237–S240.
Draganski B, Kherif F, Klöppel S, Cook PA, Alexander DC,Parker GJ, Deichmann R, Ashburner J, Frackowiak RS. 2008.Evidence for segregated and integrative connectivity patternsin the human basal ganglia. J Neurosci. 28:7143–7152.
Fattori P, Breveglieri R, Raos V, Bosco A, Galletti C. 2012. Vision foraction in the macaque medial posterior parietal cortex.J Neurosci. 32:3221–3234.
Ferry AT, Ongür D, An X, Price JL. 2000. Prefrontal cortical projec-tions to the striatum inmacaquemonkeys: evidence for an or-ganization related to prefrontal networks. J Comp Neurol.425:447–470.
Flaherty AW, Graybiel AM. 1994. Input-output organization of thesensorimotor striatum in the squirrel monkey. J Neurosci.14:599–610.
Flaherty AW, Graybiel AM. 1993. Two input systems for body re-presentations in the primate striatal matrix: experimentalevidence in the squirrel monkey. J Neurosci. 13:1120–1137.
Gerbella M, Belmalih A, Borra E, Rozzi S, Luppino G. 2011.Cortical connections of the anterior (F5a) subdivision of themacaque ventral premotor area F5. Brain Struct Funct.216:43–65.
Gerbella M, Belmalih A, Borra E, Rozzi S, Luppino G. 2010. Corticalconnections of the macaque caudal ventrolateral prefrontalareas 45A and 45B. Cereb Cortex. 20:141–168.
Gerbella M, Borra E, Rozzi S, Luppino G. 2013. Connectionalevidence for multiple basal ganglia grasping loops in themacaque monkey. Program No 270.24 2013 NeuroscienceMeeting Planner. San Diego (CA): Society of Neuroscience.
Gerbella M, Borra E, Tonelli S, Rozzi S, Luppino G. 2013. Connec-tional heterogeneity of the ventral part of the macaque area46. Cereb Cortex. 23:967–987.
Goldman PS, Nauta WJ. 1977. An intricately patterned prefronto-caudate projection in the rhesus monkey. J Comp Neurol.72:369–386.
Gregoriou GG, Borra E, Matelli M, Luppino G. 2006. Architectonicorganization of the inferior parietal convexity of themacaquemonkey. J Comp Neurol. 496:422–451.
Haber S. 2010. Integrative networks across basal ganglia circuits.In: Steiner H, Tseng KY, editors. Handbook of basal gangliastructure and function. Oxford: Elsevier. p. 409–427.
18 | Cerebral Cortex
by guest on June 19, 2015http://cercor.oxfordjournals.org/
Haber SN, Kim KS, Mailly P, Calzavara R. 2006. Reward-relatedcortical inputs define a large striatal region in primatesthat interfacewith associative cortical connections, providinga substrate for incentive-based learning. J Neurosci.26:8368–8376.
Hedreen JC, DeLong MR. 1991. Organization of striatopallidal,striatonigral, and nigrostriatal projections in the macaque.J Comp Neurol. 304:569–595.
He SQ, Dum RP, Strick PL. 1993. Topographic organization of cor-ticospinal projections from the frontal lobe: motor areas onthe lateral surface of the hemisphere. J Neurosci. 13:952–980.
Hikosaka O, Sakamoto M, Usui S. 1989. Functional properties ofmonkey caudate neurons. I. Activities related to saccadic eyemovements. J Neurophysiol. 61:780–798.
Inase M, Sakai ST, Tanji J. 1996. Overlapping corticostriatal pro-jections from the supplementary motor area and the primarymotor cortex in the macaque monkey: an anterograde doublelabeling study. J Comp Neurol. 373:283–296.
Inase M, Tokuno H, Nambu A, Akazawa T, Takada M. 1999. Corti-costriatal and corticosubthalamic input zones from the pre-supplementary motor area in the macaque monkey:comparison with the input zones from the supplementarymotor area. Brain Res. 833:191–201.
Jung WH, Jang JH, Park JW, Kim E, Goo EH, Im OS, Kwon JS. 2014.Unravelling the intrinsic functional organization of thehuman striatum: a parcellation and connectivity studybased on resting-state fMRI. PLoS One. 9:e106768.
Kelly RM, Strick PL. 2004. Macro-architecture of basal ganglialoops with the cerebral cortex: use of rabies virus to revealmultisynaptic circuits. Prog Brain Res. 143:449–459.
Kemp JM, Powell TP. 1970. The cortico-striate projection in themonkey. Brain. 93:525–546.
Lévesque M, Parent A. 2005. The striatofugal fiber system inprimates: a reevaluation of its organization based on single-axon tracing studies. Proc Natl Acad Sci U S A.102:11888–11893.
Liles SL. 1983. Activity of neurons in the putamen associatedwithwrist movements in the monkey. Brain Res. 263:156–161.
Liles SL, Updyke BV. 1985. Projection of the digit and wrist area ofprecentral gyrus to the putamen: relation between topog-raphy and physiological properties of neurons in the puta-men. Brain Res. 339:245–255.
Mesulam MM. 1982. Principles of horseradish peroxidase neuro-histochemistry and their applications for tracing neural path-ways. In: Mesulam MM, editor. Tracing neural connectionswith horseradish peroxidase. Chichester (UK): Wiley. p. 1–152.
Middleton FA, Strick PL. 2000. Basal ganglia and cerebellar loops:motor and cognitive circuits. Brain Res Brain Res Rev.31:236–250.
Middleton FA, Strick PL. 2002. Basal-ganglia “projections” to theprefrontal cortex of the primate. Cereb Cortex. 12:926–935.
Monakow KH, Akert K, Künzle H. 1978. Projections of the precen-tral motor cortex and other cortical areas of the frontal lobe tothe subthalamic nucleus in the monkey. Exp Brain Res.33:395–403.
Muakkassa KF, Strick PL. 1979. Frontal lobe inputs to primatemotor cortex: evidence for four somatotopically organized“premotor” areas. Brain Res. 177:176–182.
Murata A, Gallese V, Luppino G, Kaseda M, Sakata H. 2000. Select-ivity for the shape, size, and orientation of objects for graspingin neurons of monkey parietal area AIP. J Neurophysiol.83:2580–2601.
Nambu A. 2011. Somatotopic organization of the primate basalganglia. Front Neuroanat. 5:26.
Nambu A, Kaneda K, Tokuno H, Takada M. 2002. Organizationof corticostriatal motor inputs in monkey putamen.J Neurophysiol. 88:1830–1842.
Nambu A, Takada M, Inase M, Tokuno H. 1996. Dual somatotopi-cal representations in the primate subthalamic nucleus: evi-dence for ordered but reversed body-map transformationsfrom the primary motor cortex and the supplementarymotor area. J Neurosci. 16:2671–2683.
Oguri T, Sawamoto N, Tabu H, Urayama S, Matsuhashi M,Matsukawa N, Ojika K, Fukuyama H. 2013. Overlapping con-nections within the motor cortico-basal ganglia circuit:fMRI-tractography analysis. Neuroimage. 78:353–362.
Parent A, Hazrati LN. 1995. Functional anatomy of the basalganglia. I. The cortico-basal ganglia-thalamo-cortical loop.Brain Res Brain Res Rev. 20:91–127.
Parent A, Hazrati LN. 1994. Multiple striatal representation in pri-mate substantia nigra. J Comp Neurol. 344:305–320.
Parthasarathy HB, Schall JD, Graybiel AM. 1992. Distributed butconvergent ordering of corticostriatal projections: analysisof the frontal eye field and the supplementary eye field inthe macaque monkey. J Neurosci. 12:4468–4488.
Prodoehl J, Corcos DM, Vaillancourt DE. 2009. Basal ganglia me-chanisms underlying precision grip force control. NeurosciBiobehav Rev. 33:900–908.
Raos V, Umiltá MA, Gallese V, Fogassi L. 2004. Functional proper-ties of grasping-related neurons in the dorsal premotor areaF2 of the macaque monkey. J Neurophysiol. 92:1990–2002.
Raos V, UmiltáMA,Murata A, Fogassi L, Gallese V. 2006. Function-al properties of grasping-related neurons in the ventral pre-motor area F5 of the macaque monkey. J Neurophysiol.95:709–729.
Rozzi S, Calzavara R, Belmalih A, Borra E, Gregoriou GG,Matelli M,Luppino G. 2006. Cortical connections of the inferior parietalcortical convexity of the macaque monkey. Cereb Cortex.16:1389–1417.
Rozzi S, Ferrari PF, Bonini L, Rizzolatti G, Fogassi L. 2008. Function-al organization of inferior parietal lobule convexity in thema-caque monkey: electrophysiological characterization ofmotor, sensory and mirror responses and their correlationwith cytoarchitectonic areas. Eur J Neurosci. 28:1569–1588.
Rozzi S, Rodà F, Simone L, Bimbi M, Luppino G, Fogassi L. 2011.Movement-related neurons in the macaque ventrolateralprefrontal cortex. ProgramNo 914.10 2011NeuroscienceMeet-ing Planner. Washington (DC): Society of Neuroscience.
Saga Y, Hirata Y, Takahara D, Inoue K, Miyachi S, Nambu A,Tanji J, Takada M, Hoshi E. 2011. Origins of multisynaptic pro-jections from the basal ganglia to rostrocaudally distinct sec-tors of the dorsal premotor area in macaques. Eur J Neurosci.33:285–297.
Shook BL, Schlag-Rey M, Schlag J. 1991. Primate supplementaryeye field. II. Comparative aspects of connectionswith the thal-amus, corpus striatum, and related forebrain nuclei. J CompNeurol. 307:562–583.
Selemon LD, Goldman-Rakic PS. 1985. Longitudinal topographyand interdigitation of corticostriatal projections in the rhesusmonkey. J Neurosci. 5:776–794.
Stanton GB, Goldberg ME, Bruce CJ. 1988. Frontal eye field effer-ents in themacaquemonkey: I. subcortical pathways and top-ography of striatal and thalamic terminal fields. J CompNeurol. 271:473–492.
Strick PL, Dum RP, Picard N. 1995. Macro-organization of the cir-cuits connecting the basal ganglia with the cortical motorareas. In: Houk G, editor. Models of information processingin the basal ganglia. Boston: MIT Press. p. 117–130.
Corticostriate Projections from the Lateral Grasping Network Gerbella et al. | 19
by guest on June 19, 2015http://cercor.oxfordjournals.org/
Tachibana Y, Nambu A, Hatanaka N, Miyachi S, Takada M. 2004.Input-output organization of the rostral part of the dorsal pre-motor cortex, with special reference to its corticostriatal pro-jection. Neurosci Res. 48:45–57.
Tanaka K. 1996. Inferotemporal cortex and object vision. AnnuRev Neurosci. 19:109–139.
Takada M, Hoshi E, Saga Y, Inoue K, Miyachi S, Hatanaka N,Inase M, Nambu A. 2013. Organization of two cortico-basal ganglia loop circuits that arise from distinct sectorsof the monkey dorsal premotor cortex. In: Barrios FA,Bauer C, editors. Basal ganglia—an integrative view. InTech.p. 103–116.
Takada M, Tokuno H, Hamada I, Inase M, Ito Y, Imanishi M,Hasegawa N, Akazawa T, Hatanaka N, Nambu A. 2001. Organ-ization of inputs from cingulate motor areas to basal gangliain macaque monkey. Eur J Neurosci. 14:1633–1650.
TakadaM, TokunoH,NambuA, InaseM. 1998. Corticostriatal pro-jections from the somatic motor areas of the frontal cortex inthe macaque monkey: segregation versus overlap of inputzones from the primary motor cortex, the supplementary
motor area, and the premotor cortex. Exp Brain Res.120:114–128.
Theys T, Pani P, van Loon J, Goffin J, Janssen P. 2012. Selectivity forthree-dimensional shape and grasping-related activity in themacaque ventral premotor cortex. J Neurosci. 32:12038–12050.
Van Hoesen GW, Yeterian EH, Lavizzo-Mourey R. 1981. Wide-spread corticostriate projections from temporal cortex of therhesus monkey. J Comp Neurol. 199:205–219.
Webster MJ, Bachevalier J, Ungerleider LG. 1993. Subcortical con-nections of inferior temporal areas TE and TEO in macaquemonkeys. J Comp Neurol. 335:73–91.
Yeterian EH, Pandya DN. 1991. Prefrontostriatal connections inrelation to cortical architectonic organization in rhesus mon-keys. J Comp Neurol. 312:43–67.
Yeterian EH, PandyaDN. 1993. Striatal connections of the parietalassociation cortices in rhesus monkeys. J Comp Neurol.332:175–197.
Yeterian EH, Van Hoesen GW. 1978. Cortico-striate projections inthe rhesus monkey: the organization of certain cortico-caud-ate connections. Brain Res. 139:43–63.
20 | Cerebral Cortex
by guest on June 19, 2015http://cercor.oxfordjournals.org/
![Public Input No. 32-NFPA 90A-2012 [ Global Input ]](https://static.documents.pub/doc/80x56/6337281dd63e7c790105aa8b/public-input-no-32-nfpa-90a-2012-global-input-.jpg)
