Abstract Two functionally different cortical areas are lo-cated in the rostral part of the intraparietal sulcus (IP): theventral intraparietal area (VIP), along the fundus of thesulcus, and the anterior intraparietal area (AIP), rostral inthe lateral bank. VIP and AIP have functional propertiescomparable to those of the ventral premotor areas, F4 andF5, respectively. The aim of this study was to establishwhether these intraparietal and premotor areas have directand specific anatomical connections. Neural tracers wereinjected in F4, F5, and AIP in three macaque monkeys.The results showed that F4 and F5 are targets of strongprojections from VIP and AIP, respectively, and that thelinkage between F5 and AIP is highly selective. Thesedata support the notion that parietofrontal connections se-lectively link areas displaying similar functional proper-ties and form largely segregated anatomical circuits. Eachof these circuits is possibly dedicated to specific aspectsof sensorimotor transformations. In particular, the AIP-F5circuit should play a crucial role in visuomotor transfor-mation for grasping, the VIP-F4 circuit is possibly in-volved in peripersonal space coding for movement.

Key words Anatomical connections · Primates ·Visuomotor transformations · LIP

Introduction

The inferior sector of area 6 of Brodmann (1909), gener-ally referred to as the “ventral premotor cortex” (PMv),

is formed by two histochemical areas (Matelli et al.1985), a more caudal one, F4, and a more rostral one, F5,which markedly differ in their functional properties.

Area F4 contains a representation of face, neck, andproximal arm movements and more than half of its neu-rons have bimodal, somatosensory and visual, responses.These neurons have tactile receptive fields (RFs) locatedon the body, arm, or face and 3D visual RFs, limited tothe peripersonal space, in register with the tactile ones(Gentilucci et al. 1988; Graziano et al. 1994; Fogassi etal. 1996). The location of the visual RFs remains un-changed, regardless of the gaze direction and, therefore,it has been suggested that F4 neurons are involved incoding peripersonal space in a body-parts centered frameof reference for the visual guidance of movements(Graziano et al. 1994; Fogassi et al. 1996).

Area F5 contains a distal movement representation ofthe hand and the mouth. F5 neurons, typically, code spe-cific types of grasping or manipulation movements(Rizzolatti et al. 1988; Murata et al. 1997) and may alsohave visual responses. Visual F5 neurons may fall intotwo classes, possibly segregated into different parts of F5(see Rizzolatti et al. 1998). Neurons of one class are most-ly located within the posterior bank of the inferior arcuatesulcus and discharge to the presentation of 3D objects,even when no action upon the object is required (Murataet al. 1997). Very likely, these neurons are involved incoding the 3D features of the objects and in selecting themost appropriate grasping or manipulation movement.

This differential functional role of F4 and F5 could beaccounted for by a differential pattern of their corticaland subcortical connections (see Rizzolatti et al. 1998).One major source of afferents to PMv is the rostral partof the inferior parietal lobule (IPL; Godschalk et al.1984; Matelli et al. 1986; Kurata 1991), generally re-ferred to as area 7b (Vogt and Vogt 1919) or PF (Pandyaand Seltzer 1982). However, in these studies, tracer in-jections were placed along the inferior limb of the arcu-ate sulcus (mainly in F5), and, at present, there is no de-scription about a possible differential parietal connectivi-ty of these PMv areas.

G. Luppino (✉) · M. MatelliIstituto di Fisiologia Umana, Università di Parma,Via Volturno 39, I-43100 Parma, Italy,e-mail: luppino@ipruniv.cce.unipr.it,Tel.: +39-521-903879, Fax: +39-521-903900

A. MurataDepartment of Physiology, Nihon University,Oyaguchi-kaminachi 30-1, Itabashi-ku, 173-8610, Tokyo, Japan

P. GovoniIstituto di Istologia ed Embriologia, Università di Parma,Via Volturno 39, I-43100 Parma, Italy

Exp Brain Res (1999) 128:181–187 © Springer-Verlag 1999

R E S E A R C H N O T E

Giuseppe Luppino · Akira Murata · Paolo GovoniMassimo Matelli

Largely segregated parietofrontal connections linking rostralintraparietal cortex (areas AIP and VIP) and the ventral premotorcortex (areas F5 and F4)

Received: 03 August 1998 / Accepted: 16 March 1999

Furthermore, recent evidence showed that, in the lat-eral bank of the intraparietal sulcus (IP), there are twoareas, functionally independent of area PF. One area, re-ferred to as the “ventral intraparietal area” (VIP; Colbyet al. 1993) is located along the fundus of IP, in its mid-dle third. VIP contains neurons with sensory properties,for several aspects comparable to those of F4 (Colby etal. 1993; Duhamel et al. 1997, 1998). The other area, re-ferred to as the “anterior intraparietal area” (AIP), occu-pies the rostral part of the lateral bank of IP, with the ex-ception of its rostralmost part. AIP neurons have func-tional properties similar to those of F5 (Taira et al. 1990;Sakata et al. 1995).

The aim of this study was to assess whether these in-traparietal and PMv areas, sharing comparable functionalproperties, have direct and, eventually, specific anatomi-cal connections.

Materials and methods

Three macaque monkeys (two Macaca nemestrina and one Maca-ca fuscata) were used in these experiments. All the experimentalprocedures complied with Italian laws on the care and use of ani-mals.

Two monkeys (cases 16 and 15) were anesthetized with ket-amine chloridrate (15 mg/kg i.m., every 30 min). Under asepticconditions, a craniotomy was performed and the dura opened. Incase 16, wheat-germ agglutinin conjugated with horseradish per-oxidase (WGA-HRP; 4% in distilled water, 0.08 µl) was injectedin F4, 1.5 mm below the cortical surface, using a glass micropi-pette (tip diameter 50–100 µm) attached to a 1 µl Hamilton micro-syringe. Intracortical microstimulation was used in order to identi-fy the injection site within F4, according to the procedure and cri-teria described by Gentilucci et al. (1988). In case 15, DiamidinoYellow (DY, 2% in 0.2 M phosphate buffer, three injections, 0.2 µleach) and True Blue (TB, 5% in distilled water, three injections,0.2 µl each) were injected in F5 and F4, respectively, 1.5 mm be-low the cortical surface. Case 17 was prepared for chronic record-ing experiments according to the procedures described in Taira etal. (1990). The lateral bank of IP was then systematically exploredin both hemispheres, in order to define the location and extent ofarea AIP (for further details on recording procedure and experi-mental paradigm, see Taira et al. 1990 and Sakata et al. 1995). Atthe end of the physiological investigation, the animal was anesthe-tized and Fast Blue (FB, 3% in distilled water) and DY were in-jected in area AIP of the left hemisphere. Three injections of0.2 µl each of FB and DY were placed within the lateral bank of IP,at a depth between 2.5 and 4 mm from the cortical surface in therostral and in the caudal part of AIP, respectively. In the same hemi-sphere, TB (two injections, 0.2 µl each) was then injected more cau-dally at a depth of 4 mm within the lateral intraparietal area (LIP),identified as defined by Andersen et al. (1990). Ten days later,WGA-HRP (seven injections, 0.1 µl each) was injected in AIP inthe right hemisphere. Injections were placed at different rostro-cau-dal levels between 3 and 4.5 mm below the cortical surface.

After a survival period of 2 days following WGA-HRP injec-tions and 12 days for fluorescent-tracer injections, the animalswere deeply anesthetized with Nembutal and perfused transcardi-ally with 1.25% glutharaldehyde and 1% paraformaldehyde (case16) or with 3.5% paraformaldehyde (cases 15 and 17). The brainswere then blocked stereotactically and cut frozen in coronal sec-tions (thickness 60 µm). One section of each five was mountedand quickly coverslipped for fluorescent microscopy (cases 15 and17) or processed for HRP histochemistry (cases 16 and 17) by us-ing tetrametilbenzidine as a chromogen (TMB; Mesulam 1982).Adjacent sections were Nissl-stained or processed for cytochromeoxidase histochemistry (Matelli et al. 1985).

The fluorescent material was viewed at ×400 with a ZeissUniversal epi-fluorescence microscope equipped with a narrowband excitation filter (BP 365/11), a dichroic mirror (FT 395),and a barrier filter (LP 395). Under UV illumination and with theaid of a longpass barrier filter, which allows wavelengths greaterthan 395 nm to be visualized, the fluorescent neurons were identi-fied as follows: FB-labeled neurons by a sky-blue fluorescence inthe cytoplasm; TB-labeled neurons by a violet homogeneous flu-orescence in the cytoplasm; DY-labeled neurons by a yellow-green fluorescent nucleus. The HRP-processed material was stud-ied under both brightfield and darkfield illumination at low mag-nification.

In each section, every 600 µm, the outer and inner cortical bor-ders and the location of labeled neurons were plotted with the aidof inductive displacement transducers, mounted on X and Y axesof the microscope stage. The transducer signals were digitized andstored using a software developed in our laboratory that allows thevisualization of section outlines, of gray-white matter borders, andof labeled cells. Using the same software, the cortex was then un-folded at the level of the midpoint between the outer and innerborders, in order to obtain 2D reconstructions of the cortical sur-face or of the IP and arcuate sulcus. The unfolded sections werethen aligned and the labeling distributed along the space betweentwo consecutive plotted sections (600 µm).

Results

Figure 1A shows the distribution of the retrogradely la-beled neurons in four representative coronal sectionsthrough IP, observed in case 16 following injection ofWGA-HRP in F4. The labeling was mostly concentratedin two regions. The first one was located in the rostral-most part of the lateral bank of the IP (section a), rostralto area AIP, possibly in the face region of area 7b(Leinonen and Nyman 1979). The other was locatedmore caudally, close to and within the fundus of theIP (sections b, c, d). Its location appears to closelycorrespond to area VIP, as functionally defined byColby et al. (1993). Following TB injections in F4, incase 15, (Fig. 1B, upper right) the distribution of the la-beling within IP was remarkably similar to that ob-served in case 16 and was particularly dense in areaVIP. In contrast, following DY injections in F5, the dis-tribution of the labeling within IP was markedly differ-ent (Fig. 1B, lower right). DY-labeled neurons weremostly located in the rostral part of the lateral bank,with the exception of its rostralmost part and, therefore,well within area AIP, as functionally defined by Sakataet al. (1995).

Figure 2 shows the results of the WGA-HRP injec-tions placed in area AIP, physiologically identified, inthe right hemisphere of case 17. In the agranular frontalcortex, retrogradely labeled neurons (upper right part)were almost completely located in the rostralmost partof PMv, the densest labeling being located along theposterior bank of the inferior arcuate sulcus and, there-fore, almost completely within area F5. The antero-grade labeling in the agranular frontal cortex was com-pletely limited to PMv (lower right part of Fig. 2), andits location closely matched that of the retrograde label-ing. The fluorescent tracers injections made in the lefthemisphere of the same animal confirmed these results.

182

183

Fig. 1 A Left Dorso-lateral view of the left hemisphere of case 16,showing the location of the wheat-germ agglutinate horseradischperoxidase (WGA-HRP) injection in F4. Core and halo of the in-jection site are depicted as a black area surrounded by a shadedarea, respectively. Dotted lines mark borders of agranular areasF1, F4 and F5. AI Inferior arcuate sulcus, AS superior arcuate sul-cus, C central sulcus, IP intraparietal sulcus, L lateral sulcus, Pprincipal sulcus, STS superior temporal sulcus. Right (a–d) Repre-sentative coronal sections through the IP, showing the distributionof the labeled neurons. Each dot represents one labeled neuron.

Arrows on the hemisphere mark the levels of the sections on theright. B Left Dorso-lateral view of the left hemisphere of case 15,showing the location of True Blue (TB, upper part) and Diami-dino Yellow (DY, lower part) injections in F4 and F5, respective-ly. Right Unfolded reconstructions of the IP, showing the distribu-tion of TB- (upper part) and DY- (lower part) labeled neurons.Dashed line marks the fundus. Each dot represents one labeledneuron. Arrows mark the rostro-caudal extent of the densest label-ing in the ventral intraparietal area (VIP, upper part) or in anteriorintraparietal area (AIP, lower part)

Figure 3 (lower left part) shows the distribution withinthe arcuate sulcus of the neurons labeled with the threedifferent injected tracers. Following injections of FBand DY in the rostral and caudal part of AIP, respec-tively, the distribution of labeled neurons was very sim-ilar to that of the WGA-HRP labeling observed in theright hemisphere. Again, the densest labeling wasfound along the posterior bank of the inferior arcuatesulcus. No differential distribution was observed be-tween the FB and the DY labelings, which almost com-pletely overlapped. Conversely, following TB injection

in area LIP, virtually no labeling was observed in thepostarcuate cortex and, in full agreement with the re-sults of Andersen et al. (1990), labeled neurons weremostly located in the dorsal part of the prearcuate cor-tex.

Discussion

Present data indicate that the two PMv areas, F4 andF5, are selectively connected with two areas locatedwithin the IP, VIP and AIP. These connections link ar-eas that appear to have comparable functional proper-ties. It should be noted, however, that areas locatedwithin IP lack a clear anatomical definition and havebeen, so far, defined mainly on the basis of functionalcriteria. For this reason, the projections of AIP werestudied only after physiological mapping of this region,and the results confirmed that AIP and F5 are selective-ly and reciprocally linked. On the other hand, the attri-bution of the labeling, observed following injections inPMv, to either VIP or AIP, could be made only on thebasis of the location of VIP and AIP neurons, as report-ed by Colby et al. (1993) and Sakata et al. (1995), re-

184

Fig. 2 Left Location of wheat-germ agglutinin conjugated withhorseradisch peroxidase (WGA-HRP) injections in case 17-right,shown in a dorso-lateral view of the hemisphere (upper part) andin an unfolded reconstruction of the intraparietal sulcus (IP, lowerpart). Empty circles mark penetrations from which anterior intra-parietal area (AIP) neurons were recorded. Arrows mark therostro-caudal extent of the injection site. The dashed frame marksthe cortical sector shown enlarged in the middle part of the figure.Middle Reconstructions of the framed area, showing the distribu-tion of retrograde and anterograde labeling. Right Unfolded recon-structions of the arcuate sulcus showing the distribution of retro-grade and anterograde labeling. Dark gray areas represent the lat-eral bank of the spur, light gray areas the posterior bank of the in-ferior arcuate sulcus. Other conventions as in Fig. 1

185

spectively, or on the basis of the physiological data col-lected in this study.

Area F4 is the target of strong projections originatingfrom area VIP and of weaker, but consistent, projectionsoriginating from the rostralmost part of the lateral bankof the IP. The VIP projection to F4, in particular, appearsto be rather selective, because consistent labeling in VIPwas neither observed following tracer injections in F5,nor following injections in the dorsal and the mesial area

6 sectors (Luppino et al. 1993; Matelli et al. 1998).Functional studies have shown that neurons in both theseparietal areas projecting to F4 (VIP and rostral part ofthe lateral bank of IP) have sensory properties compara-ble, in several aspects, to those of F4. Area VIP was ob-ject of several studies showing that, in this area, there areneurons with visual or bimodal, visual and tactile, re-sponses. Bimodal neurons have tactile RFs typically lo-cated around the face and visual RFs located within theperipersonal space, in register with the tactile fields. Inmany of them, the location of the visual RFs remains un-changed, regardless of where the gaze is directed. It hasbeen suggested, therefore, that, as in F4, in VIP, there areneurons involved in space coding in egocentric coordi-nates, centered on the head (Duhamel et al. 1997, 1998;see also Colby 1998). Much less is known about thefunctional properties of the rostralmost part of the lateralbank of the IP. However, according to Leinonen andNyman (1979), this cortical sector contains neurons with

Fig. 3 Upper left and middle Location of fluorescent tracers injec-tions in case 17-left, shown in a dorso-lateral view of the hemisphereand in an unfolded reconstruction of the intraparietal sulcus (IP).Empty circles mark penetrations from which anterior intraparietal area(AIP) neurons were recorded. Arrows mark the rostro-caudal extentof injection sites. Right Representative coronal sections through theinjection sites. The level of the section is marked on the view of thehemisphere. Lower left Unfolded reconstructions of the arcuate sulcusshowing the distribution of Fast-Blue (FB), Diamidino-Yellow (DY),and True-Blue (TB) labeled neurons. Conventions as in Fig. 2

both tactile (mainly around the face) and visual RFs and,therefore, should be considered as an extension of area7b within the rostral tip of the IP.

F5 is target of strong projections originating from ar-ea AIP. Injections in this parietal area showed that theanterograde and retrograde labelings in the agranularfrontal cortex was almost completely confined to F5and, therefore, the anatomical linkage between thesetwo areas is highly selective and reciprocal. In addition,the differential distribution of the labeling observed inthe present study following injections in AIP and LIP, inagreement also with data of Andersen et al. (1990) onLIP connections, support the physiological evidencethat AIP is an independent field within the lateral bankof the IP. AIP neurons have been studied with an experi-mental paradigm virtually identical to that more recentlyemployed by Murata et al. (1997) in F5. These studiesshowed that in AIP, as in F5, there are also neurons withmotor responses coding specific kinds of grasping ormanipulation movements and/or visual responses thatappear to be related to the coding of 3D object charac-teristics (Taira et al. 1990; Sakata et al. 1995). There-fore, as F5, AIP also appears to be involved in visuomo-tor transformations for grasping (see Jeannerod et al.1995).

It is now largely accepted that parietofrontal connec-tions are organized in a parallel, although overlappingfashion (see Wise et al. 1997). In particular, it has beenpointed out that regions of the superior parietal lobuleand of the dorsal premotor cortex, displaying similarfunctional properties, tend to be preferentially linkedthrough parallel sets of connections (Johnson et al. 1996).

Present data, along with data on the parietal connec-tions of mesial and dorsal premotor cortex (Luppino etal. 1993; Matelli et al. 1998), clearly indicate that thereare cases in which parietofrontal connections are highlyselective. All together, they support the notion that themain connections between parietal and frontal areasform a series of largely segregated circuits (see Rizzol-atti et al. 1997, 1998).

The possible functional correlate of this anatomicalorganization is that each of these parietofrontal circuitsis dedicated to a particular aspect of sensorimotortransformation. In particular, functional evidence, aswell as evidence from inactivation experiments, indi-cate that the AIP/F5 circuit plays a crucial role in thevisual guidance of hand grasping and manipulationmovements (Jeannerod et al. 1995; Gallese et al. 1997).On the other hand, the VIP/F4 circuit is possibly in-volved in encoding peripersonal space and in trans-forming object locations into appropriate movementstoward them (Duhamel et al. 1997; Rizzolatti et al.1997; Colby 1988).

Acknowledgements This research was supported by Biomed 2,contract n-BMH4-CT95-0789, Human Frontieres and Murst.

References

Andersen RA, Asanuma C, Essick G, Siegel RM (1990) Cortico-cortical connections of anatomically and physiologically de-fined subdivisions within the inferior parietal lobule. J CompNeurol 296:65–113

Brodmann K (1909) Vergleichende Lokalisationslehre der Gross-hirnrinde. Barth, Leipzig (Reprinted 1925)

Colby CL (1998) Action-oriented spatial references frames in cor-tex. Neuron 20:15–24

Colby CL, Duhamel J-R, Goldberg ME (1993) Ventral intraparie-tal area of the macaque: anatomic location and visual responseproperties. J Neurophysiol 69:902–914

Duhamel J-R, Bremmer F, BenHamed S, Graf W (1997) Spatialinvariance of visual receptive fields in parietal cortex neurons.Nature 389:845–848

Duhamel J-R, Colby CL, Goldberg ME (1998) Ventral intraparie-tal area of the macaque: congruent visual and somatic proper-ties. J Neurophysiol 79:126–136

Fogassi L, Gallese V, Fadiga L, Luppino G, Matelli M, RizzolattiG (1996) Coding of peripersonal space in inferior premotorcortex (area F4). J Neurophysiol 76:141–157

Gallese V, Fadiga L, Fogassi L, Luppino G, Murata A (1997) Aparietal-frontal circuit for hand grasping movements in themonkey: evidence from reversible inactivation experiments.In:Thier P, Karnath H-O (eds) Parietal lobe contributions toorientation in 3D space. Springer, Berlin Heidelberg NewYork, pp 255–270

Gentilucci M, Fogassi L, Luppino G, Matelli M, Camarda R,Rizzolatti G (1988) Functional organization of inferior area 6in the macaque monkey. I. Somatotopy and the control ofproximal movements. Exp Brain Res 71:475–490

Godschalk M, Lemon RN, Kuypers HGJM, Ronday HK (1984)Cortical afferents and efferents of monkey postarcuate area: ananatomical and electrophysiological study. Exp Brain Res56:410–424

Graziano MSA, Yap GS, Gross CG (1994) Coding of visual spaceby premotor neurons. Science 266:1054–1057

Jeannerod M, Arbib MA, Rizzolatti G, Sakata H (1995) Graspingobjects: the cortical mechanisms of visuomotor transforma-tion. Trends Neurosci 18:314–320

Johnson PB, Ferraina S, Bianchi L, Caminiti R (1996) Corticalnetworks for visual reaching: physiological and anatomical or-ganization of frontal and parietal lobe arm regions. Cereb Cor-tex 6:102–119

Kurata K (1991) Corticocortical inputs to the dorsal and ventralaspects of the premotor cortex of macaque monkeys. NeurosciRes 12:263–280

Leinonen L, Nyman G (1979) Functional properties of cells in an-terolateral part of area 7 associative face area of awake mon-keys. Exp Brain Res 34:321–333

Luppino G, Matelli M, Camarda R, Rizzolatti G (1993) Cortico-cortical connections of area F3 (SMA-Proper) and area F6(Pre-SMA) in the macaque monkey. J Comp Neurol338:114–140

Matelli M, Luppino G, Rizzolatti G (1985) Patterns of cytochromeoxidase activity in the frontal agranular cortex of macaquemonkey. Behav Brain Res 18:125–137

Matelli M, Camarda R, Glickstein M, Rizzolatti G (1986) Afferentand efferent projections of the inferior area 6 in the macaquemonkey. J Comp Neurol 251:281–298

Matelli M, Govoni P, Galletti C, Kutz DF, Luppino G (1998) Su-perior area 6 afferents from the superior parietal lobule in themacaque monkey. J Comp Neurol 402:327–352

Mesulam M-M (1982) Principles of horseradish peroxidase neuro-histochemistry and their applications for tracing neural path-ways. In: Mesulam M-M (ed) Tracing neural connections withhorseradish peroxidase. Wiley, Chichester, pp 1–152

Murata A, Fadiga L, Fogassi L, Gallese V, Raos V, Rizzolatti G(1997) Object representation in the ventral premotor cortex(area F5) of the monkey. J Neurophysiol 78:2226–2230

186

Sakata H, Taira M, Murata A, Mine S (1995) Neural mechanismsof visual guidance of hand action in the parietal cortex of themonkey. Cereb Cortex 5:429–438

Taira M, Mine S, Georgopulos AP, Murata A, Sakata H (1990) Pa-rietal cortex neurons of the monkey related to the visual guid-ance of hand movement. Exp Brain Res 83:29–36

Vogt O, Vogt C (1919) Ergebnisse unserer Hirnforschung. J Psy-chol Neurol 25:277–462

Wise SP, Boussaoud D, Johnson PB, Caminiti R (1997) Premotorand parietal cortex:corticocortical connectivity and combinato-rial computations. Annu Rev Neurosci 20:25–42

187

Pandya DN, Seltzer B (1982) Intrinsic connections and architec-tonics of posterior parietal cortex in the rhesus monkey. JComp Neurol 204:196–210

Rizzolatti G, Camarda R, Fogassi M, Gentilucci M, Luppino G,Matelli M (1988) Functional organization of inferior area 6 inthe macaque monkey. II. Area F5 and the control of distalmovements. Exp Brain Res 71:491–507

Rizzolatti G, Fogassi L, Gallese V (1997) Parietal cortex: fromsight to action. Curr Opin Neurobiol 7:562–567

Rizzolatti G, Luppino G, Matelli M (1998) The organization of thecortical motor system: new concepts. Electroencephalogr ClinNeurophysiol 106:283–296