Proc. Nail. Acad. Sci. USA Vol. 87, pp. 449-453, January 1990 Neurobiology Innervation patterns of single physiologically identified geniculocortical axons in the striate cortex of the tree shrew (geniculocortical afferents/visual cortex/on- and off-center pathways/binocular convergence) DAVID FITZPATRICK* AND DENiS RACZKOWSKI Department of Neurobiology, Duke University, Durham, NC 27710 Communicated by Irving T. Diamond, October 10, 1989 ABSTRACT We examined the termination patterns of MATERIALS AND METHODS single geniculocortical axons in the striate cortex of the tree shrew by using intracellular recording and horseradish per- We obtained the data for this report from adult tree shrews oxidase staining methods. Axons were classified by whether raised in our breeding colony. At the outset ofan experiment, they responded to light onset (ON center) or light offset (OFF we anesthetized the tree shrew with ketamine and xylazine, center) and whether they were driven by the ipsi- or contra- cannulated a vein for infusion of a dextrose/saline solution lateral eye. Afferents with ON-center responses end in the containing sodium thiopental (delivered at a rate of 3 upper part oflayer IV (IVa) whereas afferents with OFF-center mgkg-1hr-1 for the remainder of the experiment), per- responses end in the lower part of layer IV (IVb). Within each formed a tracheotomy, and placed a catheter in the bladder. tier, axons driven by the ipsilateral and contralateral eye The animal was placed in a stereotaxic apparatus, paralyzed overlap. These results suggest that binocular convergence with pancuronium bromide to prevent eye movements and occurs within layer IV without mixing the information from the ventilated with a respirator to maintain expired CO2 between ON- and OFF-center pathways and we consider the significance 2 and 3%. Body temperature was maintained at 37.50C by of this arrangement for visual cortical function. using a thermostatically controlled heating blanket. The pupils were dilated with atropine and contact lenses were The tree shrew is a good species in which to study the used to protect the corneas and to focus the eyes on visual organization of the visual system because it shows a remark- stimuli presented on a tangent screen. able degree of segregation of visual pathways in the lateral We encountered geniculocortical axons in the deep layers geniculate nucleus and in the striate cortex. In most mam- of the cortex or in the adjacent white matter by using mals, the inputs from the two eyes are segregated into fine-tipped glass micropipettes filled with HRP. Axons were different layers of the lateral geniculate nucleus, and the tree characterized by their response to the onset or offset ofa light shrew is no exception. But, in addition to ocular segregation, stimulus presented to the ipsilateral or contralateral eye. lamination in the tree shrew lateral geniculate nucleus seg- Then, we attempted to impale the axon with the micropipette. regates the projections of ganglion cells that respond to light If successful, we rechecked its receptive field properties to onset (ON-center) from those that respond to light offset ensure that it was the same fiber and injected HRP by passing (OFF-center). Conway and Schiller (1) showed that the two current. All injected axons had response properties entirely most medial layers, 1 and 2, contain mostly ON-center consistent with those recording from neurons in the lateral neurons and layers 4 and 5 contain mostly OFF-center geniculate nucleus (1, 7): they were monocular, nonoriented, neurons. and had receptive field centers that responded to stationary The segregation of the ON and OFF pathways appears to flashing stimuli. be maintained by the projections of the lateral geniculate Several hours after the last intracellular injection, the nucleus to layer IV of striate cortex. Layer IV in the tree animal was deeply anesthetized with barbiturate and per- shrew consists of two horizontal tiers separated by a cell- fused with aldehyde fixatives. We removed the brain, sparse cleft and the tract tracing studies of Harting et al. (2) trimmed it to a block containing the visual cortex, and and Conley et al. (3) showed that the ON-center geniculate refrigerated this block overnight in a solution of 5.0% (wt/ layers project to the upper tier, IVa, whereas the OFF-center vol) sucrose in 0.1 M sodium phosphate, pH 7.4. The next layers project to the lower tier, IVb. The studies of Norton day, 80-,.m-thick sections were cut in the coronal plane using and colleagues (4, 5) provided support for these anatomical a vibratome and these were reacted with 3,3'-diaminoben- experiments by showing that neurons in IVa have ON zidine tetrahydrochloride as the chromogen with nickel and responses whereas neurons in IVb have OFF responses. cobalt intensification (8). Here we tested the idea that the ON and OFF pathways Sixteen axons that were labeled by intraaxonal injection of terminate in separate tiers by combining intracellular record- HRP and recovered for morphological analysis form the basis ing and horseradish peroxidase (HRP) staining methods to for our conclusions. Twelve of these axons (6 ON center and visualize the termination patterns of individual physiologi- 6 OFF center) were reconstructed using a microscope cally characterized geniculocortical axons. The results ob- equipped with 63x and 100x oil-immersion objectives (NA tained confirm that the axons of individual ON- and OFF- 1.40 and 1.32, respectively) and a drawing tube. The relation center geniculate neurons end in distinct subtiers of layer IV of axon terminal arbors to the layers of the visual cortex was and we consider the significance of this segregation for the assessed by counterstaining selected sections with cresyl function of visual cortex. Some of these results have ap- violet. peared in abstract form (6). Abbreviations: ON, light onset; OFF, light offset; HRP, horseradish peroxidase. The publication costs of this article were defrayed in part by page charge *To whom reprint requests should be addressed at: Department of payment. This article must therefore be hereby marked "advertisement" Neurobiology, Box 3209 Medical Center, Duke University, in accordance with 18 U.S.C. §1734 solely to indicate this fact. Durham, NC 27710. 449 Downloaded by guest on October 3, 2021 450 Neurobiology: Fitzpatrick and Raczkowski Proc. Natl. Acad. Sci. USA 87 (1990) RESULTS cesses arborized in layer IlIb and in the cleft that separates layer IVa and IVb. A few processes also ended in layer IVb. Fig. 1 shows a cross section of the striate cortex in the tree Neither this nor any other geniculocortical axon in our sample shrew stained for Nissl substance. Layer IV stands out as a had processes that arborized in layer VI. prominent cell-rich band that is divided into two tiers by a The laminar projections of a representative OFF-center cell-sparse cleft. The upper tier is designated IVa and the geniculocortical axon driven by the ipsilateral eye is illus- lower tier is IVb. trated in Fig. 4. The first bifurcation point occurred just Fig. 2 illustrates the morphological features seen in the below where the axon arborized densely in the lower tier of terminal arbors of several geniculocortical axons in layer IV; layer IV (layer IVb). In addition to the cell's main arbor, a few the two classes, ON and OFF, are virtually indistinguishable boutons ended in layer V just below its border with layer IV. from one another. For both, the arbors consist mainly of very Only a few labeled terminal boutons occupied the cleft fine axon collaterals, less than 0.5 ,um in diameter, that are between layers IVa and IVb. studded with many boutons en passant and that end with a In both tiers, axons driven by the contralateral eye and single bouton or terminal. Only occasionally do these bou- those driven by the ipsilateral eye were distributed across the tons occur in clusters. The boutons vary in diameter from 0.5 depth of the tier. Since there is no evidence for columnar to 3.0 ,um. segregation of the inputs from the two eyes in layer IV of the The main point of this report is that all of the ON-center tree shrew's striate cortex (2, 9-11), axons driven by the geniculocortical axons, whether driven by the left or right eye, contralateral and ipsilateral eyes must overlap extensively. had terminal fields largely confined to cortical layer IVa, while There are, however, differences in the spatial distribution of all of the OFF-center axons had terminal fields mainly re- ipsilateral and contralateral axons. For example, axons stricted to IVb. The terminal arborization pattern of a repre- driven by the ipsilateral eye have a larger lateral extent than sentative ON-center geniculocortical axon driven by the con- those driven by the contralateral eye. Also, ipsilateral axons tralateral eye is illustrated in Fig. 3. Traced from the site of display more boutons near the borders of layer IV (i.e., the injection in the white matter (data not illustrated), the parent upper part of IVa and the lower part of IVb) than in the middle axon coursed unbranched until it neared its termination in the of the layer. These differences will be considered in more striate cortex. There, the axon turned sharply to enter into the detail elsewhere. gray matter where it bifurcated several times to form a dense collection of preterminal axon branches and terminal boutons DISCUSSION in layer IV. For this ON-center axon, most ofthe bouton-laden What might the segregation of the ON and OFF pathways in processes arborized densely in the upper tier oflayer IV (layer tree shrew layer IV tell us about the functional organization IVa). Besides the main arborization in layer IVa, some pro- of striate cortex? To address this question it is first necessary .