The Journal of Neuroscience, March 1995, 75(3): 2561-2574

Distribution and Density of Monoamine Receptors in the Primate Visual Cortex Devoid of Retinal Input from Early Embryonic Stages

Pasko Rakic and Michael S. Lidow Section of Neurobiology, Yale University School of Medicine, New Haven, Connecticut 06510

Developmental mechanisms that regulate the areal and neurotransmittersand correspondingreceptors during early de- laminar distribution of various macromolecules, including velopmental stageswas suggestedas one among many compo- neurotransmitter receptors in the cerebral cortex, are not nentsof this complex process,but so far little has been explored known. In the present study, we examined the development experimentally in vivo (seereviews by Mattson, 1988; Lauder, of receptors in the rhesus monkey striate 1993). and peristriate visual cortex in the absence of input from The striate and peristriate cortex in primatescan be usedto the retina. Binocular enucleation was performed between test the role of the periphery on cortical developmentbecause embryonic days E60 and E61, prior to the ingrowth of gen- these two areas display remarkable differences in laminar or- iculocortical fibers into the cortical plate and before gen- ganization, constellationof cell classes,pattern of synaptic con- esis of the granular and supragranular layers of the visual nectivity, and distribution of various macromolecules,including cortex. The animals were delivered at term (E165) and sac- neurotransmitterreceptors (e.g., Lund, 1988; Rakic et al., 1988; rificed at 2 or 12 months of age, and their brains frozen and Felleman and Van Essen, 1991). For example, the color oppo- the occipital lobes cut at 20 pm in the coronal plane. nent system and the so-called “broad band” system of spatial Cortical binding of 3H-, 1251-, 3H-5-HT, information processingin primatesconsist of distinct neuronal 3H-, 3H-, 3H-SCH23390, and 3H- populationsinitially identified physiologically (e.g., Wiesel and that label various monoamine receptors were autoradi- Hubel, 1966; DeValois and DeValois, 1975). The segregationof ographically visualized and quantified using a computer these two parallel systemsis maintainedfrom the retina to the imaging system. All radioligands displayed specific lami- parvocellular (P) and magnocellular (M) layers of the lateral nar patterns in the striate and prestriate areas in both geniculate nucleus, and then via geniculocortical projections to groups of animals. The areal and laminar distribution in the specific cortical layers involved predominately in color and ori- anophthalmic monkeys was similar to that in the controls. entation/motion discrimination (e.g., DeYoe and Van Essen, Significantly, in all enucleated animals, just as in the con- 1988; Livingstone and Hubel, 1988). The sublayer IVCa is the trols, a particularly high density of 3H-clonidine and 3H-pra- magnocellularrecipient zone and IVA and IVCP receive input zosin was observed in the sublayers of layer IV involved in from parvocellular layers of the lateral geniculate nucleus (Liv- color vision. The present results show that the monoamine ingstone and Hubel, 1988). While delineation of their exact receptors in primate visual cortex can establish and main- physiological functions remainsto be elucidated, several lines tain distinct laminar and areal patterns in the absence of of evidence revealed important psychophysiologicaldifferences activity or molecular cues originated from the retina, and betweenM and P systems(Schein and deMonasterio, 1987) that provide new insight into the cortical consequences of sec- can be differentiated by the number of anatomicaland cytolog- ondary congenital anophthalmia. ical criteria (DeYoe and Van Essen,1988; Livingstone and Hu- [Key words: neurotransmitfer receptors, striate cortex, bel, 1988; Lund, 1988; Ts’o and Gilbert, 1988). M and P visual development, rhesus monkey] subsystem-specificcytoarchitectonic features in macaquemon- key display clear and abrupt changesat the border between stri- In recent years, numerousstudies have indicated that develop- ate and extrastriate areas(Lund, 1988). Most importantly, a ma- mental programs operating within cortical neurons, as well as jority of neurotransmitterreceptors in the rhesusmonkey display inductive cues from various afferent systems,play a synergistic region-specificlaminar binding propertiesthat can be correlated role in cortical specification(Wiesel, 1982; Rakic, 1988; Jones, with the pattern of cellular layers and their input-output rela- 1990; Shatz, 1990; McConnell, 1991; O’Leary et al., 1994). tionshipsin visual areas(Rakic et al., 1988). Recently, the time- However, the relative contributions and molecular mechanisms table for the emergenceof thesereceptors and changesin their underlying theseinteractions in laminar and area1specification pattern of distribution during prenatal and postnataldevelopment of the cerebral cortex are still unknown. The interaction between was determined in this species(Lidow and Rakic, 1992, 1994). In the presentstudy, we examined the distribution of the seven Received May 6, 1994; revised Sept. 29, 1994; accepted Dec. 13, 1994. monoaminereceptor-specific ligands in the occipital cortex of This work was supported by the U.S. Public Health Service. Pregnant mon- keys were obtained from Yale University Medical School Breeding Colony and animals that were binocularly enucleated as fetuses, starting N&v England Regional Primate Center, Southboro, MA. from the first third of gestation.Although not engageddirectly Correspondence should be addressed to Pasko Rakic, Section of Neurobiol- in signal transmissionwithin the retino-geniculo-cortical syn- ;;giy University School of Medicine, 333 Cedar Street, New Haven, CT aptic chain, different subtypes of monoamine receptors are Copyright 0 1995 Society for Neuroscience 0270-6474/95/152561-14$05.00/O thought to be involved in modulationof cortical activity by spe- 2562 Rakic and Lidow * Development of Monoamine Receptors

Table 1. Protocols of saturation autoradiographic assays

Site Exposure labeled Cont. (nM) Blank Protocol Buffer Time Temp. time

3H-prazosin 0.2-1.0 Preincubation 0.17 M Tris-HCl 30 min 4°C 3.5 months al 100 PM (PH 7.1) Incubation Same buffer 70 min 4°C Rinse Same buffer 2 X 5 min 4°C ‘H-clonidine Norepinephrine 0.5-4.0 Norepinephrine Preincubation 0.17 M Tris-HCl 30 min room 3.5 months a2 100 PM (pH 7.6), 5 IILM EDTA temp. Incubation 0.17 M Tris-HCl 60 min room (pH 7.6) 1 mM MnCl,, temp. 0.1% ascorbic acid Rinse 0.17 M Tris-HCl 2 X 5 min 4°C (PH 7.6) Norepinephrine 0.025-0.200 Isoproterenol Incubation 0.02 M Tris-HCl 70 min room 2 days B 100 fJ,M (pH 7.4), 100 PM temp. Rinse 0.02 M Tris-HCI 3 X 20 min 4°C (PH 7.4) “H-5-HT 2-32 Serotonin Incubation 0.17 M Tris-HCl 60 min room 2.5 months 5-HT, 10 FM (pH 7.7), 0.4 mu temp. CaCl,, 0.1% ascorbic acid, 1 pM , 10 pM pargyline Rinse 0.17 M Tris-HCl 5 X 1.0 min 4°C pH 7.7) ‘H-ketanserin Serotonin 0.5-3.5 Incubation 0.05 M Tris-HCl 30 min 1 month 5-HT, 10 FM (pH 7.7), 0.1 PM prazosin Rinse 0.05 M Tris-HCl 5 X 0.4 min 4°C 3H-SCH23390 l-10 cis- Incubation 0.05 M Tris-HCl 90 min room 3 months D, 1 PM (pH 7.4), 120 mM temp. NaCl, 5mM KCl, 2 mM CaCl,, 1 UIM MgCl,,O.S )“M Rinse 0.05 M Tris-HCl 2 X 10 min 4°C (PH 7.4) 3H-spiperone Dopamine 0.7-6.6 (-)Butaclomol Incubation 0.17 M Tris-HCl 30 min room 2 months D* 1.0 FM (pH 7.7), 120 mM temp. Serotonin NaCl, 5 mM KCl, 5-HT,,, 5-HT,,, 2 mh4 CaCl, 5-HT, 1 mM M&l,, Norepinephrine 0.1% ascorbic acid a2 Rinse 0.17 M Tris-HCl 5 X 1.0 min 4°C (PH 7.7)

cific afferents (Fallon and Loughlin, 1987). Therefore, one may ceptors (Rakic et al., 1988). Recently we found that several expect that the density and distribution of thesereceptors may monoaminereceptors appear early in the transient embryonic be related to the level of activity in primary visual pathways. zones of the fetal telencephalon,with particularly high density Furthermore, monoaminesoriginating from the brainstem arrive localized in the subplatezone subjacentto the prospectivearea early to the embryonic telencephalonwhere they may have a 17 where it can be detected before area1differences appears in regulatory role on the differentiation of neurons(Lauder et al., the overlaying cortical plate (Lidow and Rakic, 1994). Our ma- 1982; Parnavelasand Blue, 1982; Mattson, 1988; Whitaker-Az- jor goal was to examine how these receptorsdevelop, become matia, 1991; Lauder, 1993). In the cortex of adult primates,there distributed, and are maintainedin the absenceof input from the are significant area1and laminar-specificdifferences in the dis- retina, and thus, to learn about the extent to which input from tribution of monoamines(Brown and Goldman-Rakic, 1977; the retina induces and maintains laminar and area1organization Morrison et al., 1982; Levitt et al., 1984) and monoaminere- of monoaminergicreceptors. The Journal of Neuroscience, March 1995, 75(3) 2563

Figure 1. Cresyl violet-stained sections of visual cortex of normal (A, B) and anophthalmic (C, D) rhesus monkeys, both 2 months old. Note the similarity between the cytoarchitecture of visual cortex in normal controls (A, B) and ophthalmic animals (C, D). Also, the border between striate and peristriate areas (marked by arrows) is sharp in both normal (B) and anophthalmic (D) animals.

Materials and Methods binding for some radioligands at 2 months, may differ from that ob- served in the adult (Rakic et al., 1988; Lidow and Rakic, 1992). How- Experimental procedures. Eight rhesus monkeys (four experimental and ever, it should be underscored that, for the purpose of the present study, four age-matched controls) were used in the present study. The time of comparisons were made between operated and control groups based on pregnancy, which in this species lasts 165 d, was determined as de- animals sacrificed at 2 postnatal months. scribed elsewhere (Rakic, 1972). Bilateral retinal ablation was per- Tissue preparation and binding assays. Animals were anesthetized formed in utero, at embryonic (E) day E60, E63, E67, or E81, as de- with Na-pentobarbital (40 mg/kg) and perfused with ice-cold phosphate- scribed by Rakic and Goldman-Rakic (1985). Following surgery, buffered saline (pH 7.4) followed by 0.1% paraformaldehyde containing fetuses were returned to the uterus and the chorio-alantoic membranes, increasing concentrations of sucrose in buffered saline (Rakic et al., uterine, and abdominal wall were sutured. All animals were delivered 1988). The light fixation allowed brain tissue to be sectioned more at term. Three of the operated animals were sacrificed at the second postnatal month because previous studies indicated that the adult lam- easily without affecting binding parameters (Rakic et al., 1988). The inar, columnar, and area1 pattern in the striate cortex in the macaque brains were rapidly removed, blocked, and immersed in isopentane at monkey is well established by that time (e.g., Hubel et al., 1977; Rakic, -40°C for 5 min before storing at -80°C. Blocks of tissue containing 1977a). In addition, the basic visual functions, including grating acuity, striate and extrastriate cortex were cut in the coronal plane at 20 km. Vernier acuity, and color discrimination, are well developed by the end Sections were mounted on acid-cleaned, chrom alum-subbed slides and of the, second month in this species (Teller and Boothe, 1979). The kept at -80°C until the time of assay, conducted no more than 2 weeks major neurotransmitter receptors in the macaque monkey display an after tissue had been sectioned. elaborate adult pattern by that age, but the level of binding is higher The protocols for the saturation binding assays used in this study than in the adult (Lidow et al., 1991a; Lidow and Rakic, 1992). One have previously been described in Rakic et al. (1988) and Lidow et al. experimental animal in the present study was sacrificed at 1 year of age (1989a,b) and are summarized in Table 1. 3H-prazosin was used to bind when binding is fully differentiated (Lidow and Rakic, 1992). al adrenergic receptors. 3H-clonidine was employed to visualize high Therefore, the pattern of distribution, and in particular, the amount of affinity a2 adrenergic receptors. 1*51-pindolol was used to label B-ad- 2564 Rakic and Lidow l Development of Monoamine Receptors

Figure 2. Typical labeling pattern of 3H-prazosin (al adrenergic receptors) labeling in the visual cortex of normal (A) and anophthalmic (B) monkey, both 2 months old. Arrows indicate the border between striate and peristriate areas. Note the similarity in labeling pattern between the normal and experimental animal, except that in the anophthalmic animal, striate area is interrupted by a segment of cortex that resembles the peristriate area.

renergic sites. To prevent an inclusion of 5HT,-serotonergic receptors in the specific ‘*51-pindolol labeling, nonspecific binding for this radi- Normal animals Anophthalmic animals oligand was determined in the presence of an excess of isoproterenol. 5-HT,-serotonergic sites were identified with 3H-hydroxytryptamine creatinine sulfate (3H-5-HT). Incubation buffer for 3H-5-HT binding contained fluoxetine to prevent binding to serotonin uptake sites, and pargyline to prevent the oxidation of the ligand by monoamine oxidase present in the tissue. 3H-ketanserin was used to label 5-HT,-serotonergic receptors. The binding was done in the presence of prazosin to prevent radioligand binding to al-adrenergic receptors. In addition, the use of methysergide as the displacing agent assures that the specific binding of 3H-ketanserin is not contaminated by cwl-adrenergic, , or unique ketanserin sites associated with dopaminergic terminals. 3H- SCH23390 was employed to label D,-dopaminergic receptors. Labeling was done in the presence of mianserin to prevent radioligand binding to serotonergic and adrenergic sites. The last ligand used in this study

t

Figure 3. Histograms representing the distribution of 3H-prazosin (al receptors) in visual cortex of 2-month-old control and anophthalmic animals. Three histograms are presented for the primary and secondary visual areas of both normal and experimental animals. The upper his- togram shows B,,, with 95% comparison intervals of Gabriel modifi- cation of GT2 test. The middle histogram presents the distribution of B max + SEM (n = 3 for each group). The lower histogram shows the distribution of Kd with their comparison intervals. The roman numerals indicate cortical layers according to Lund (1988) except that sublayers IV& and IVCl3 are labeled IV& and IVCP. If comparison intervals for any two B,,, or Kd overlap, there are no statistically significant differences between them. Note that while the pattern of 3H-prazosin labeling in both groups of animals is similar, the density of labeling in striate area of the experimental monkeys is significantly lower than in the controls. The Journal of Neuroscience, March 1995, 75(3) 2565

Figure 4. Typical labeling pattern of 3H-clonidine (high-affinity (-w2receptors) labeling in the visual cortex of a normal (A) and an anophthalmic (B) monkey, both 2 months old. Arrows show the border between striate and peristriate areas. Note the similarity in labeling pattern between the normal and experimental animal. However, in the anophthalmic animal the striate area is interrupted by cortex that looks more as the peristriate

was ‘H-spiperone. Originally, we expected it to allow us to visualize weight for each cortical layer. The diffuse optical densities used in this D,-dopaminergic receptors. However, we found later that under the con- study were between 0.08 and 0.80 on all autoradiograms. Within this ditions used in this study, 3H-spiperone binds receptor subtypes, which, range, optical densities of 3H-sensitive ultrafilm are linearly related to in addition to D, sites, include 5-HT,-, 5-HT,,-, 5-HT,,-serotonergic variations in tissue radioactivity (Lidow et al., 1988). and cx2-adrenergic sites (Lidow, et al., 1991b). Therefore, 3H-spiperone The analysis of saturation binding utilized nonlinear, curve-fitting in the present study was used as the useful molecular probe for marking computer programs KINETIC/EDBA/LIGAND/LOWRY from Elsevier-BIO- laminar and area1 differences between anophthalmic and control animals SOFT Co. (Cambridge, UK). The analyses were based on concentra- rather than as the means to determine the precise distribution of func- tions of radioactive ligands specifically bound to tissue labeled with five tional D, receptor subtype, which can be labeled by more specific ra- different concentrations of free ligands in incubating solutions. The dioligand (Lidow et al., 1991b). number of data points employed for quantitative analysis of saturation Quantitative densitometry. The identification of area 17 (striate cor- binding is a compromise between the desired accuracy and the avail- tex) and area 18 (peristriate cortex), as well as the definition of their ability of appropriate tissue. Five concentrations of free ligand in in- lamination, was made according to the scheme of Brodmann (1909). cubating solution is a minimal number that allows a relatively accurate The layers were identified according to the classification of Lund estimation of B,,, and Kd for a one-site receptor model (Brooksbank et (1973). The autoradiograms were analyzed with a computer-imaging al., 1978; Lidow et al., 1989a). B,,, and K,, values obtained from dif- system described previously (Rakic et al., 1988). The system allows the ferent layers of each neocortical area were compared with Gabriel’s overlay on a computer screen of the digitized images of cresyl violet- modification of the GT2-method (Sokal and Rohlf, 1981). This method stained sections and corresponding autoradiograms to identify histolog- utilizes the data provided by KINETIC/EDBA/LICAND/OWRY and produc- ically the autoradiographic image of each cortical layer and sublayer. es 95% comparison intervals for each Kd or B,,,, Kd or B,,, with over- In some instances, autoradiograms revealed that individual cytoarchi- lapping intervals considered statistically identical; Kd or B,,, with in- tectonic layers can be further subdivided into two bands (e.g., layer IIIA tervals that do not overlap are considered statistically different. Plots of and IIIB for 3H-clonidine or laver IVBa and IVBb for 1251-nindolol). these comparison intervals make it very easy to appreciate the degree The computer system also subtracts film images of sections with non- of variation in Kd or B,,, within each cortical area as well as to identify specific binding from film images of adjacent sections with total bind- the layers for which these variations are statistically significant. For ing, thus allowing the investigator to observe directly on the screen the most ligands, binding was repeated on three experimental and three images representing specific binding. The program then compares op- control ‘animals of 2 months of age. However, in the case of 3H- tical densities of film images of individual cortical layers with those of SCH23390 and Y-pindolol, the ligands were introduced late in the SH or lzsI standards, which were apposed to the film along with tissue study. Thus, binding of 3H-SCH23390 was conducted on only one ex- sections. The optical densities from autoradiographs of labeled tissue perimental and one control animal of 2 months of age. In the case of were converted into concentrations of labeled compounds per tissue wet lZSI-pindolol, the quantitative data were obtained only from the animals 2666 Rakic and Lidow l Development of Monoamine Receptors

[3H]clonidine Normal animals Anophthalmic animals

Figure 5. Histogramsrepresenting the distributionof ‘H-clonidine (high-affinitya2 receptors)in visualcortex of controland anophthalmic 2-month-oldmonkeys. Conventions are the sameas in Figure3. Note the similarityin both densityand distributionof ‘H-clonidinelabeling Figure 6. Typical labelingpattern of 1251-pindolol(p receptors) in the in both groupsof animals. visualcortex of normal(A) andanophthalmic (B) l-year-oldmonkeys. The cortex in both photographsis representedtwice, once aboveand oncebelow the calcarinefissure (asterisks). Arrows indicatethe sharp of 1 year of age.Two-month-old animals provided qualitative data for borderbetween striate and peristriate areas in the upperlip of the fissure. this radioligand. The photographsillustrate the similarity in labelinglaminar pattern in Quenchingcorrection for TH-generatedautoradiographic images of the normaland experimentalanimals, although layer VI in B is cut layer IVB in the striatecortex wasmade as describedearlier (Lidow et tangentiallydue to the developmentof moreelaborate convolutions in al., 1988).The measurementsobtained from 2-month-oldanimals were the anophthalmics. not correctedfor quenchingsince no significantamount of myelin is developedby that time in layer IVB (seeLidow andKakic, 1992,their Fig. 1). than the surface occupied by this area in control animals. In Results some specimens,the striate area appearsto be barely one-third Cytoarchitectonic pattern. The surface of the occipital lobe in of the normal size in both surface and number of nerve cells. all anophthalmicanimals was more convoluted than in the con- This dramatic decreasein the size of the striate areais in agree- trols and commonly contained a deep horizontal sulcus running ment with our quantitative analysis based on the frozen and acrossthe occipital operculum(Rakic, 1988). The striate cortex celloidin-embedded,cresyl violet-stained serial sections(Rak- (area 17) and the adjacent peristriate cortex (area 18) were, in ic, 1988; Rakic et al., 1991; I. Suner and I? Rakic, unpublished each case, first identified in cresyl violet-stained preparations observation). The division of the peristriate cortex into areas (Fig. 1). In agreement with previous observations (Rakic, 18, 19, or unusual archetype (area X of Rakic et al., 1991) was 1988), the striate area in the prenatally enucleated animals more difficult to assess.The light fixation of the tissue, re- could be recognized by its characteristic cytoarchitectonic lam- quired for receptor autoradiography, did not allow sufficient inar pattern. In particular, one can easily distinguish sublayers resolution to delineate theseareas on cresyl violet sectionsand IVA, IVB, IV&, and IVCP that reveal the segregationof the no variations in the receptor binding patterns within prestriate M and P system in the cortex. The thickness of each layer was cortex were recorded for any of radioligands usedin this study. approximately equal to that of the age-matchedcontrols, except Thus, we concentrated the present analysison the striate cortex for layer I, which was usually lo-15% wider than in the con- and refer to the regions outside its border in the occipital lobe trol. The border of the striate cortex with adjacent areas was as the peristriate area. always very sharp and easily discernible. A cursory examina- 3H-prazosin binding. The al-specific binding of 3H-prazosin tion of the tissue sectionsalso revealed that the surface of the had a similar distribution in both 2-month-old prenatally oper- striate area in the early enucleates was significantly smaller ated animalsand age-matchednormal controls. In the striate area The Journal of Neuroscience, March 1995, 15(3) 2567

Normal animals Anophthalmic animals

Figure 7. Histograms representing the distribution of ‘251-pindolol (p receptors) in visual cortex in the l-year-old control and anophthalmic animals. Conventions are as in Figure 3. For this particular ligand, the Figure 8. Typical labeling pattern of ‘H-5-HT (5-HT, receptors) in the data were collected from a single normal and experimental animal, both visual cortex of a normal (A) and an anophthalmic (B) monkey, both 2 of which display the similarity in the distribution of 1251-pindolol bind- months old. Note that the cortex is represented twice, above and below ing. the calcarine fissure (asterisks). The band of heavily labeled deep layers in the striate cortex in the upper lip of the sulcus stops abruptly at the border to peristriate cortex in both specimens (arrows). Labeling pattern of all specimens,the pattern of binding was characterizedby the of both area in the normal and experimental animal is similar. two additional darker bands-one thin, densely labeled and as- sociated with sublayer IVA and the other much less densecor- respondingto layer VI (Fig. 2). Although it is difficult to discern in both groups of animals,the density of binding was consis- in the individual photograph, a medium-denseband is present tently 35-40% lower in the experimental group comparedto the in layer VI, which is confirmed by densitometry (see below). normal controls (Fig. 3). This reduction was proportional in all While the superficial, broad band in layers I and II continues cortical layers, so that the overall pattern of receptordistribution over the entire occipital surface, the dark bands corresponding remainedpreserved. In addition, there were no differencesin the to deeperlayers are presentonly in the striate area. The border affinity of this radioligandbetween control and experimentaltis- of the striate and peristriate cortex can be easily discernedas sues(Fig. 3). the site where the dark bandscorresponding to layers IVA and 3H-clonidine binding. The cw2-specificlabeling of ?H-clonidine VI abruptly disappear(Fig. 2A,B). However, in contrast to the in the striate cortex of normal 2-month-old monkeys displayed binding in normal tissue,the striate pattern of “H-prazosin bind- three dark bands(Fig. 4A). The superficial band corresponded ing in the depth of the calcarine fissureof the enucleatedanimals to layer I. The domain of the middle band included layers IIIB was often interrupted by the insertsof segmentswith a pattern and IVA, while the deepestsituated band correspondedto cy- of receptor binding similar to that of the peristriatecortex (Fig. toarchitectonic layers IVC and V (Fig. 4A). The samethree-layer 2B). Since only 20 sectionswere used for the binding of this pattern was observed in the 2-month-old animalsenucleated as ligand in each case,we could not determinewhether the segment fetusesbetween E60 and E67 (Fig. 4B). The border of the striate displaying this deviation from the characteristic striate pattern to peristriate area in both groups of animals was easily recog- of binding representsan “island” encircled by the normal-look- nizable by the abrupt lossof the denseband presentcorrespond- ing cortex or a “peninsula” that is anteriorly continuous with ing to layers IIIB/IVA. However, the higher levels of binding the adjacent area. present in layers I, IVCB and V of the striate cortex continued Histogramsof the B,,, values of 3H-prazosinbinding across without interruption acrossthe border into layers I, IV and V of cortical layers in normal and enucleated animals confirmed the peristriate area. As in the case of ?H-prazosinbinding, 3H- quantitatively the visual impressionof the similarity in the pat- clonidine binding acrossthe anterior portion of calcarine fissure tern of laminar distribution of this radioligand in the striate cor- of anophthalmicspecimens showed a short segmentof the ex- tex in both groups of animals (Fig. 3). The highest level of trastriate pattern (Fig. 4B), which was not present in age- binding in IVA of the striate cortex is evident in all animals,as matched controls (Fig. 4A). However, again due to the gap in is the small, statistically nonsignificant increase in layer VI. serial sections, we could not determine whether this segment However, although the laminar pattern was remarkably similar representsan isolated “island” or a “peninsula.” 2566 Rakic and Lidow l Development of Monoamine Receptors

Normal animals Anophthalmic animals

Figure 9. Histogramsrepresenting the distributionof 3H-5-HT(5-HT, receptors)in the visualcortex of controland anophthalmic 2-month-old monkeys.Conventions are as in Figure 3. Note the similarityof both densityand distribution on ?H-5-HTlabeling in bothgroups of animals.

The B,,, values for 3H-clonidinewere similar in control and experimentalgroups acrossall layers in both striate and peris- Figure IO. Typical labelingpattern of 3H-ketanserin(5-HT, receptors) in the visualcortex of a normal(A) andan anophthalmic(B) monkey, triate cortex (Fig. 5). However, unlike al receptor,a2 sitesdis- both2 monthsold. The calcarinefissure is marked with asterisk. Arrows play a high density in cell-sparselayer I that is similar to the pointto the borderbetween striate and peristriate areas, which is visible more cellular layers IIIb, IVA. Furthermore, no difference in Kd dueto the relatively abruptdisappearance of heavy labelin the deep valuescould be discernedbetween enucleates and control in any corticallayers. Despite the changein positionof the border,the laminar of the layers (Fig. 5). patternof labelingis similarin both specimens. ‘251-pindolol binding. 12SI-pindololwas used for analysisof the distribution of P-adrenergicreceptors. In the striate cortex of normal l-year-old animals,the binding of this radioligand was the two groups of animals (Fig. 7). There were no differences characterized by a prominently high density in layers I, II, III, in Kd values (Fig. 7). While we had only a limited number of the superficial half of IVB and V and low density in sublayers tissue sectionsfrom 2-month-old animals labeled with lZ51-pin- IVA, the deephalf of IVB and IVC (Fig. 6A). The high binding dolol, they showedsimilar patterns of labeling in experimental density in the superficial half of layers IVB and V as well as and control animals (not shown). low density in sublayersIVA, the deep half of IVB, and IVC 3H-5-HT binding. In the striate area of normal 2-month-old stoppedabruptly at the border of the striate area. However, the specimens,S-HT, specific binding of 3H-5-HT had a dark band binding density in superficial layers I, II, and III continued into in layer IVB, which mergeswith a more moderate binding in the prestriatecortex (Fig. 6A). The samelaminar pattern as well IVC. The borderline of striate and prestriateareas was sharply as an abrupt change at the area1border was present in the op- delineated;radiolabeling of layer IV is much lighter in the per- erated animals.Comparison of the photographsin Figure 6, A istriate area (on the left side in Fig. 8A). Similarly, the sharp andB, showsa remarkablesimilarity in the laminar composition border and the samearea1 and laminar pattern can be observed betweenoperated and controls, although the oblique cut, due to in the enucleated2-month-old animals (Fig. 8B). Quantitative the more elaborateconvolutions in anophthalmics,renders layer analysisrevealed a similar laminar distribution and densitiesof VI as distorted and wider than it is (Fig. 6B). Particularly dra- 3H-5-HT binding in both areas(Fig. 9). There were no differ- matic is abrupt stop of the low-density bandthe striate/peristriate encesin affinity of 3H-5-HT binding in the two groups of ani- border in both specimens. mals (Fig. 9). B,,, values of ‘*-Y-pindololbinding in all layers in the striate -‘H-ketanserin binding. 5-HT,-specific binding of 3H-ketan- and peristriate areas were basically indistinguishablebetween serin was particularly high in layers IVB, IVC, and V in the The Journal of Neuroscience, March 1995, 75(3) 2569

Normal animals Anophthalmic animals

Figure 12. Typicallabeling pattern of 3H-SCH23390(D, receptors)in the striatevisual cortex of a normal(A) andan anophthalmic(B) mon- key, both2 monthsold. Notethe similarityin labelingpattern between the normaland experimental animals.

Figure 11. Histograms representing the distribution of 3H-ketanserin certainty concerning the validity of the B,,, valuesgiven for 3H- (5HT, receotors) in the visual cortex of control and anophthalmic an-.._ \ A L imals 2 months bld display similarity in binding pattern: Conventions SCH23390. are as in Figure 3. 3H-spiperone binding. The labelingof 3H-spiperonein the stri- ate cortex of the normal 2-month-oldmonkeys was characterized by a wide dark band correspondingto layers IVC and V, which striate areas of the normal 2-month-old monkeys. Although the stops at the border with the peristriate areas(Fig. 14A). In the same set of layers (IV and V) were also intensely labeled in the latter cortical region, the labeling was largely homogeneous(Fig. peristriate area, here the intensity of labeling was visibly lower 14A). The distribution of the 3H-spiperonewas similar between than in area 17, making the area1 border easily recognizable the normal and enucleatedmonkeys in both cytoarchitectonic (arrow in Fig. 1OA). An essentially similar pattern was observed areasexamined (Fig. 14B). The B,, and Kd valuesfor all layers in the operated animals (Fig. 10B). Quantitative analysis showed in both areaswere similar in all specimens(Fig. 15). no differences in B,,, and Kd values between experimental and control animals(Fig. 11). Discussion 1H-SCH23390 binding. In the striate cortex of the 2-month- The distinct difference in neuronal organization between striate old monkey, the D,-specific 3H-SCH23390binding had three and peristriate areasin primatesdevelops as a result of syner- dark, high-density bandscorresponding to layers I, II, IVB and gistic interactionsbetween the intrinsic capacity of cortical cells V, VI, respectively (Fig. 12A). Only the middle band, localized and information from the periphery via thalamocorticalconnec- to layer IVB, disappearedabruptly at the border to the peristriate tions (Rakic, 1988, 1991; Rakic et al., 1991; Kennedy and De- area.This changein binding pattern, nevertheless,makes a clear hay, 1993). The present study addsto the growing list of mo- demarcationmarker for identification of the area1border (Fig. lecular and cellular features of the cerebral cortex that can 12B). Essentiallythe samefindings were obtained for both cy- develop and be maintained independently of modality-specific toarchitectonic areasin the operated and age-matchedcontrol cuesoriginated in the peripheralreceptors and transferredto the animals. cortical neuronsvia thalamocortical pathways (Berry and Hol- The quantitative analysisof autoradiogramssupported our vi- lingsworth, 1973; Kaiserman-Abramov et al., 1983; Dehay et sual observation that 3H-SCH23390had a similar pattern of la- al., 1989; Kuljis and Rakic, 1989; Kennedy et al., 1990; Rakic beling in both experimentaland control groupsof animals.How- et al., 1991; Arimatsu et al., 1992; Barbe and Levitt, 1993;Ferri ever, the B,,, values for this ligand in all layers of both striate and Levitt, 1993; Kind et al., 1993; Cohen-Tannoudjiet al., and peristriate areasof the experimental specimenswere about 1994; Schlaggaret al., 1994). However, to evaluate the devel- 15% higher than in the control (Fig. 13). It should be under- opmental significance of the present findings, one has to take scoredthat, as we mentionedin the Materials and Methods, only into account the embryonic stage when enucleation has been one experimentaland one control animal were processedfor the performed. analysis of this particular ligand. This fact may raise some un- The timing of enucleation. Retinal ablation was performed 2570 Rakic and Lidow - Development of Monoamine Receptors

Anophthalmic animals 12 z * * * * 8 * i f s 4

s oJ - Figure 13. Histograms representing the distribution of 3H-SCH23390 (D, receptors) in the visual cortex of the 2-month-old control and an- ophthalmic animals. Conventions are as in Figure 3. Note the similarity on distribution of 3H-clonidine labeling in both animals. However, al- though the pattern of ‘H-SCH23390 labeling in both animals is the same, the density of labeling in the experimental monkey is significantly higher than in the control. Figure 14. Pattern of 3H-spiperone labeling in the visual cortex of a normal (A) and an anophthalmic (B) monkey, both 2 months old, shows during the first half of the 165 d gestationof the macaquemon- similarity. Arrows indicate the border between striate and peristriate key. At the beginning of that time (E60, E63, E67 specimens), areas. lessthan one-third of the photoreceptorsand only about half of the retinal ganglion cells have been generated (LaVail et al., 1991). At the end of this period (E81) the genesisof ganglion glion cells and the diencephalon,but, importantly, the photore- cells has beencompleted, but, importantly, synaptic contactsbe- ceptorsare not connectedto the ganglion cells, synaptic contacts tween photoreceptorsand ganglion cells via bipolar cells have in the primary visual pathway have not been established,and not been established(Nishimura and Rakic, 1985). The mosaic layer IV, the main recipient of the geniculateinput, has not been map of rods and three wave-length sensitive cones has not generatedyet. emergedat the time of enucleation (Wikler and Rakic, 1991, Bilateral enucleation performed during the first half of ges- 1994). Although all neuronsdestined for the lateral geniculate tation of the macaquemonkey results in a significant reduction nucleushave been generated at the time of enucleation in the in size of the lateral geniculate nucleus(Rakic, 1988; Dehay et youngestthree specimens(Rakic, 1977b), only the infragranular al., 1989). The nucleusdevelops neither the layers nor the mag- layers in the visual cortex have been formed (Rakic, 1974). no- and parvocellular division that normally comprise a relay Therefore, neuronsof layer IV, the main recipient of thalamic station in the M and P system (seeintroductory remarks).The input, as well as layers III and II, are generatedonly after the geniculate nucleus retains only 400,000 and 800,000 neurons animalshave been enucleated.At the time of early enucleation, from over 2,000,OOO(Williams and Rakic, 1988) presentwhen some ganglion cells have extended their axons to the lateral the surgery was performed, around E60 and E80, respectively geniculate nucleus (Rakic and Riley, 1983; Williams et al., (Rakic et al., 1991; Suner and Rakic, unpublishedobservation). 1991),but did not as yet form synapses(Hendrickson and Rakic, Therefore, in the presentstudy, the developing visual cortex was 1977). Although some geniculate axons already have invaded deprived of all information from the retina, and could receive the wall of telencephalic vesicle (Rakic, 1976), they are sus- only partial, presumably undifferentiated information from the pendedin the transient subplatezone and have not entered the grossly diminished lateral geniculate nucleusthat did not seg- cortical plate itself (Rakic, 1977a). Therefore, at the time of regate into M and P layers. These facts should be taken into enucleationthere is an early axonal link betweenthe retinal gan- account in the interpretation of monoaminereceptor develop- The Journal of Neuroscience, March 1995, 75(3) 2571

[3H]spiperone mals. Since thesesublayers normally receive input from the col- Normal animals or opponent P system all the way from the retina via the par- vocellular layer of the lateral geniculate body (seeintroductory remarks), the present results in the anophthalmicanimals indi- cate that the density of c-w2-adrenergicreceptors and their high distribution in cortical sublayerssubserving color vision are not induced and/or maintained by impulse activity or some other information from the wavelength-sensitivephotoreceptors. Thus, unlike the formation of ocular dominance columns, which are induced by competition between inputs subserving two eyes (Rakic, 1976; Hubel et al., 1977), the segregationof cortical layers devoted to color and noncolor visual functions can form in the absenceof activity pattern originated in the retina. For example, removal or even occlusion of a single eye prevents formation of the ocular dominancecolumns (Hubel et al., 1977; Rakic, 1981), suggestingthat activity-dependent binocular com- petition, initiated from the retina, is an essentialfactor (Shatz, 1990). The present results, as well as the nearly normal devel- opment of cytochrome oxidasepatches in prenatally bienucleat- ed animals (Kuljis and Rakic, 1989), indicate that their cyto- and chemoarchitectonicfeatures can form independently of the activity pattern in the retina. While the role of signalsoriginated from the retinal photo- receptorson the distribution of any of the ligandstested in the present study can be excluded, it is not possibleto determine what role, if any, geniculocortical input may play in this process. In macaque monkey fetus, adrenergic receptors assumetheir laminar pattern only in the secondhalf of gestation(Lidow and Rakic, 1994) when axons originating from the geniculatenucle- Figure 15. Histograms representing the distribution of ‘H-spiperone us begin to invade the cortical plate (Rakic, 1976, 1977a). This in visual cortex of control and anophthalmic 2-month-old monkeys. timing of developmentalevents, theoretically, allowsfor the pos- Nonspecific binding was determined in the presence of an excess of sibility that thalamic input exerts its influence on the monoamine (+)butaclomol. Conventions are as in Figure 3. Note the similarity of receptor distribution in the cortex. However, since, in the ano- both density and distribution on 3H-spiperone labeling in both groups phthalmic monkeys, the remaining neuronsin the lateral genic- of animals. ulate nucleus do not segregateinto M- and P-related moieties (Rakic, 1988; Suner and Rakic, unpublishedobservation), it is ment and distribution in the cortex of prenatally induced ano- inexplicable that neuronal targets of the geniculate axon termi- phthalmia. nals in the cortex display such a clear cyto- and chemoarchitec- Distribution of monoaminergicreceptors in the anophthalmic tonic segregationinto M- and P-related sublayers.One possibil- animals.As has been shown in previous primate studiesinvolv- ity is that the occipital cortex in primates has an intrinsic ing prenatal binocular enucleation, the basic cytoarchitectonic developmental program essentialfor forming some of the spe- features, including lamination of the visual cortex, were rela- cies-specific cytoarchitectonic features, including formation of tively well preserved, although the size of the striate area was M and P sublayersin area 17 and distribution pattern of mono- significantly smaller than in the age-matchedcontrols (Rakic, amine receptors. An alternative possibility is that geniculate neu- 1988; Dehay et al., 1989; Rakic et al., 1991; Suner and Rakic, rons generate such a program independently of the cues from unpublished observation). In addition, the present analysis the retinal photoreceptors,and induce formation of thesecortical showedthat all seven monoaminereceptor subtype-specific ra- featuresvia the remaining thalamocorticalinput. The appropriate dioligands examined display a normal pattern of distribution in tests of thesetwo possibilitieshave not been performed in any anophthalmics.Five radioligandshad the samedensity as in the primates. However, relevant studies were recently done in the controls, and only two receptor subtypesdisplayed a statistically somatosensorysystem of developing rodents.In this species,the significant difference in their concentration, as expressedin B,,, block of impulse activity, originated from the ventroposterolat- values. For example, although the density of al-labeling with era1 thalamocortical afferents by NMDA antagonists, does not “H-prazosinin anophthalmicswas lower than in the controls, the disrupt the emergenceof a normal barrel pattern characteristic pattern of labeling was neverthelessremarkably similar to in of somatosensorycortex (Schlaggaret al., 1994). both groupsof animals.Since the lower density of binding can Comparison with previous studies. The present work is the be attributed to the total absenceof visual stimulation from the first quantitative autoradiographicanalysis of the distribution of retina, the initial formation of the normal laminar pattern, char- monoamine receptors in the cerebral cortex of experimentally acteristic for the striate cortex, in the absenceof this stimulation induced anophthalmic animals.Earlier reports on the effect of from the early embryonic stagesis even more significant. In extrinsic factors on neurotransmitterreceptor maturation in the particular, we find intriguing that normally high o2-specific neocortex dealt mainly with the effect of visual deprivation, such binding of 3H-clonidine in sublayersIVA and IVCP (Rakic et as eyelid suture,dark rearing, or monocularenucleation at post- al., 1988) was present also in the cortex of anophthalmic ani- natal ages(Schliebs et al., 1982a,b;Wilkinson et al., 1983; Aoki 2572 Rakic and Lidow - Development of Monoamine Receptors et al., 1986; Aurich and Bigl, 1988; Shaw and Cynader, 1988; anisms of secondary anophthalmia in which retinas in the em- Chalmers and McCulloch, 1989; Reynolds and Bear, 1989). Fur- bryo degenerate soon after the eye primordium has been formed. thermore, most of these studies were carried on in nonprimate Both in secondary congenital anophthalmia in humans (Bolton, species in which the primary visual cortex has different connec- 1900; Brunquell et al., 1984) and in prenatally enucleated mon- tions, area1 division, laminar pattern, and synaptic organization. keys (Rakic, 1988; Rakic et al., 1991), the laminar pattern of For example, the border of area 17, which is very sharp and the striate cortex appears to develop normally, but the overall prominent in primates, is either faint or nonexistent in carnivores size of the striate area is considerably smaller. The present re- and rodents (Lund, 1988). Furthermore, the visual cortex in the sults show, in addition, that the striate cortex of the occipital rhesus monkey has a more elaborate area1 and laminar distri- lobe can develop and maintain a normal pattern and density of bution of neurotransmitter-specific innervation than in any sub- major monoamine receptor subtypes in the absence of activity primate species examined thus far (for review, see Rakic et al., or contact-mediated transfer of molecular cues from the photo- 1988). There is some evidence that the cellular mechanisms un- receptors of the retina. derlying differentiation of the visual cortex in primate are initi- What is the function of the monoamine receptors in the visual ated earlier and display a more autonomous course than in the cortex deprived of its normal retinal input? There is evidence nonprimate species (Rakic, 1988; Kennedy and Dehay, 1993). that the presumptive visual cortex is electrically and metaboli- It is therefore possible that the chemoarchitecture of this highly tally active. For example, the occipital pole, occupied normally differentiated cortical subsystem in primates had acquired a new by the primary visual cortex in congenitally blind humans, dis- developmental strategy during evolution. However, despite plays a considerable level of metabolic activity in the absence species-specific differences in timing and sequence of develop- of visual stimulation (Veraat et al., 1990). In contrast, the same mental cellular events, and the use of different experimental pro- region in individuals who lose their sight during adulthood dis- cedures, and in many instances different radioligands, a plays low activity under the very same conditions. The high comparison between the present results and the data of receptor metabolic activity may be sustained by the aberrant synapses bindings in enucleates and dark-reared animals in other species developed and maintained in the absence of visual stimulation. is instructive. For example, the density of synapses per unit volume of neuropil Review of the literature shows that dark rearing does not usu- in the striate cortex attains and retains normal levels in ano- ally interfere with development of monoaminergic receptors in phthalmic monkeys (Bourgeois and Rakic, 1994). Finally, for- the visual cortex. For example, several investigators report no mation of synapses in the striate cortex in premature monkeys changes in the density of P-adrenergic receptors in visually de- in which visual stimulation begins at an earlier age proceeds on prived rats and cats (Schliebs et al., 1982a,b; Wilkinson et al., the same schedule as in controls, suggesting that the initial 1983; Aoki et al., 1986; Aurich and Bigl, 1988). It was also phases of synaptogenesis in the visual cortex may be indepen- shown that monocular deprivation after birth does not alter either dent of visual stimulation (Bourgeois et al., 1989). Taken to- al- or a2-adrenergic receptors in rat visual cortex (Aurich et gether, these findings suggest that certain species-specific, area- al., 1989). Finally, while Aurich et al. (1985) observed an in- specific aspects of synaptic and chemical architecture of the crease in density of 5HT,-serotonergic sites in dark-reared rats, cerebral cortex can develop and be maintained independently of Murphy et al. (1980) were unable to detect changes in these visual experience. Therefore, the affect of impulse activity from sites in similar conditions. Since previous studies of monoamine the periphery may have an effect mostly on the selection and receptor distribution involved less drastic, later sustained, and stabilization of already formed synapses. shorter lasting manipulations of visual input, we expected that enucleation performed before midgestation, and involving both References retinas, would have a more significant affect on monoamine re- Aoki C, Kaufman D, Rainbow TC (1986) The ontogeny of the laminar ceptors. The results, however, revealed a remarkably normal dis- distribution of B-adrenergic receptors in the visual cortex of cats, tribution and density of monoamine-specific ligands examined. normally reared and dark-reared. Dev Brain Res 27: 109-l 11. A normal developmentof monoaminereceptors observed in Arimatsu Y, Mixamoto M, Nihonmatsu I, Hirata K, Uratani Y, Hatanka visual deprivation studies stands in contrast to significant Y, Takiguchi-Hoyashi K (1992) Early regional specification for a molecular neuronal phenotype in the rat neocortex. Proc Nat1 Acad changesin the density and distribution that were reported for Sci USA 89:8879-8883. several other classesof neurotransmitterreceptors. Thus, it has Aurich M, Big1 V (1988) A critical period of the development of B-ad- been shown that eyelid suture in newborn kittens reduces sub- renergic receptor binding in the visual system of rat during visual stantially the concentrationof GABA receptors(Shaw and Cy- deprivation. Int J Dev Neurosci 6:35 l-357. nader, 1988). Likewise monocularenucleation in the rat causes Aurich M, Schliebs R, Big1 V (1985) Serotonergic receptors in the visual system of light-deprived rats. Int J Dev Neurosci 3:285-290. a reduction in jH-glutamate binding in both superficial and deep Aurich M, Schliebs R, Stewart MC, Rudolph E, Fischer H-D, Big1 V layers of the rat and cat visual cortex (Chalmersand McCulloch, (1989) Adaptive changes in the central noradrenergic system in mo- 1989; Reynolds and Bear, 1989). Unilateral eye suture in the nocular deprived rats. Brain Res Bull 22:173-180. developing rat can result in changesin the pattern and decrease Barbe ME Levitt P (1993) The early commitment of fetal neurons to in density of distribution of muscarinicACh receptors(Walsh et limbic cortex. J Neurosci 11:519-533. Berrv M, Hollingsworth J (1973) Develonment of isolated neocortex. al., 1989). Finally, normal laminar distribution of M, and M, Experientia 29:204-207. ~ ’ I muscarinic ACh receptor subtypesin the cat visual cortex can Bolton JS (1900) the exact histological localization of the visual area be prevented by elimination of cortical afferents in early post- of the human cerebral cortex. Philos Trans R Sot (Lond) 193: 165- natal life (Prusky and Cynader, 1990). Thus, it is likely that some 202. other neurotransmittersystems may dependmore on visual stim- Bourgeois J-P, Rakic P (1994) Changing of synaptic density in the primary visual cortex of the rhesus monkey from fetal to adult stage. ulation for their developmentand maintenance. J Neurosci 13:2801-2820. Implicationsfor congenital anophthalmia. The presentresults Bourgeois J-R Jastreboff PJ, Rakic P (1989) Synaptogenesis in visual are relevant to understandingthe timing and pathogenesismech- cortex of normal and preterm monkeys: evidence for intrinsic regu- The Journal of Neuroscience, March 1995, 1.5(3) 2573

lation of synaptic overproduction. Proc Nat1 Acad Sci USA 86:4297- neurotransmitter receptors with primate neocortex. Cereb Cortex 4301. 2:401-415. Brodmann K (1909) Vergleichende Lokalisationslehre der Grosshirnin- Lidow MS, Rakic P (1994) Unique profiles of the al-, a2- and B-ad- de. Leipzig: Barth. renergic receptors in the developing cortical plate and transient em- Brooksbank BWL, Martinez M, Atkinson DJ, Balas R (1978) Bio- bryonic zones of the rhesus monkey. J Neurosci 14:4064&4078. chemical development in the human brain. I. Some parameters of Lidow MS, Goldman-Rakic PS, Rakic P Gallager DW (1988) Differ- chofinergic system. Dev Neurosci 1:267-284. ential quenching and limits of resolution in autoradiograms of brain Brown RM, Goldman-Rakic PS (1977) Catecholamines in neocortex tissue labeled with 3H-lz51- and 14C-compounds. Brain Res 459:105- of rhesus monkeys: regional distribution and ontogenetic develop- 119. ment. Brain Res 124:550-576. Lidow MS, Goldman-Rakic PS, Gallager DW, Rakic P (1989a) Quan- Brunquell IJ, Papale JH, Horton JC, Williams RS, Zgrabik MJ, Albert titative autoradiographic mapping of serotonin 5-HT, and 5-HT, re- DM, Hedley-Whyte T (1984) Sex-linked hereditary bilateral ano- ceptors and serotonin uptake sites in neocortex of rhesus monkey. J phthalmos. Pathologic and radiologic correlation. Arch Ophthalmol Comp Neural 280:27-42. (Copenh) 102: 108-l 13. Lidow MS, Goldman-Rakic PS, Gallager DW, Geschwind DH, Rakic P Chalmers DT, McCulloch J (1989) Reduction in [)H] glutamate binding (1989b) Distribution of major neurotransmitter receptors in the pri- in the visual cortex after unilateral orbital enucleation. Neurosci Lett mary motor and somatosensory cortex of rhesus monkey. Neurosci- 97:298-304. ence 32:609-627. Cohen-Tannoudji M, Babinet C, Wassef M (1994) Early intrinsic re- Lidow MS, Goldman-Rakic PS, Rakic P (1991a) Synchronized over- gional specification of the mouse somatosensory cortex. Nature 368: production of neurotransmitter receptors in diverse regions of the 460-463. primate cerebral cortex Proc Nat1 Acad Sci USA 88:10218-10221. Dehay C, Horsburgh 0, Berland M, Killackey H, Kennedy H (1989) Lidow MS, Goldman-Rakic PS, Gallager DW, Rakic P (1991b) Dis- Maturation and connectivity of the visual cortex in monkey is altered tribution of dopaminergic receptors in the primate cerebral cortex: by prenatal removal of retinal input. Nature 337:265-267. quantitative autoradiographic analysis using [‘Hlraclopride, r3H]spip- Dehay C, Horsburgh G, Berland M, Killackey H, Kennedy H (1991) erone and [3H]SCH23390. Neuroscience 40:657-671. The effects of bilateral enucleation in the primate fetus on the par- Livingstone M, Hubel DH (1988) Segregation of form, color, move- cellation of visual cortexes. Dev Brain Res 62:137-141. ment, and depth: anatomy, physiology and perception. Science 240: DeValois RL, DeValois KK (1975) Neural coding of color. In: Hand- 740-749. book of perception: seeing (Carterette EC, Freedman MP, eds). New Lund J (1973) Organization of neurons in the visual cortex of area 17 York: Academic. of the monkey (Macaca mulatta). J Comp Neural 147:455496. DeYoe EA, Van Essen DC (1988) Concurrent processing streams in Lund J (1988) Anatomical organization of macaque monkey visual monkey visual cortex. Trends Neurosci 11:2 19-226. cortex. Annu Rev Neurosci 11:253-288. Fallon JH, Loughlin SE (1987) Monoamine innervation of cerebral Mattson MP (1988) Neurotransmitters in the regulation of neuronal cortex and theory of the role of monoamines in cerebral cortex and cytoarchitecture. Brain Res Rev 13:179-212. basal ganglia. In: Cerebral cortex, Vol 6 (Jones EG, Peters A, eds), McConnell SK (1991) The generation of cortical diversity in the cen- pp 41-128. New York: Plenum. tral nervous system. Annu Rev Neurosci 14:269-300. Felleman DJ, Van Essen DC (1991) Distributed hierarchical processing Morrison JH, Foote SL, O’Connor D, Bloom FE (1982) Laminar, tan- in the primate cerebral cortex. Cereb Cortex 1: l-47. gential and regional organization of noradrenergic innervation of Ferri RT, Levitt P (1993) Cerebral cortical progenitors are fated to monkey cortex: dopamine B-hydroxidase immunocytochemistry. produce area-specific neuronal population. Cereb Cortex 3:187-198. Brain Res Bull 9:309-319. Hendrickson A, Rakic P (1977) Histogenesis and synaptogenesis in the Murphy S, Uzbekow MG, Rose SP (1980) Changes in serotonin re- dorsal lateral geniculate nucleus (LGd) of the fetal monkey brain. ceptors in different brain regions after exposure to dark-reared rats. Anat Ret 187:602. Neurosci Lett 17:317-321. Hubel DH, Wiesel TN, LeVay S (1977) Plasticity of ocular dominance Nishimura Y, Rakic P (1985) Development of the rhesus monkey ret- columns in monkey striate cortex. Philos Trans R Sot Lond [Biol] ina. I. Emergence of the inner plexiform layer and its synapses. J 278:377409. Comp Neural 24 1:420-434. Jones EG (1990) Determinants of the cytoarchitecture of the cerebral O’Leary DDM, Schlaggar BL, Tuttle R (1994) Specification of neo- cortex. In: Signals and sense. Local and global order in perceptual cortical areas and thalamocortical connections. Annu Rev Neurosci maps (Edelman EG, Gall WE, Cowan MW, eds), pp 3-49. New York: 17:419439. Wiley. Parnavelas JG, Blue ME (1982) The role of noradrenergic system in Kaiserman-Abramov I, Graybiel A, Nauta WH (1983) Thahunic pro- formation of synapses in the visual cortex of the rat. Dev Brain Res jection to area 17 in a congenitally anophthalmic mouse strain. Neu- 3:140-144. roscience 5:41-52. Peters A, Jones EG, eds (1984) Cerebral cortex, Vol 1, Cellular com- Kennedy H, Dehay C (1993) Cortical specification of mice and men. ponents of the cerebral cortex. New York: Plenum. Cereb Cortex 3:171-186. Prusky G, Cynader M (1990) The distribution of M, and M, muscarinic Kennedy H, Dehay C, Horsburg G (1990) Striate cortex periodicity. acetylcholine receptor subtypes in developing cat visual cortex. Dev Nature 384:494. Brain Res 56:1-12. Kind P, Murphy DM, Sengpiel E Blakemore C (1993) Disruption of Rakic P (1972) Mode of cell migration to the superficial layers of fetal generation of layer 4 in cat area 17 retards geniculate axon segre- monkey neocortex. J Comp Neurol 145:61-84. gation but not the emergence of cytochromoxidase “blobs.” Sot Rakic P (1974) Neurons in rhesus monkey visual cortex: systematic Neurosci Abstr 19:893. relation between time of origin and eventual disposition. Science 183: Kuljis R, Rakic P (1989) Hypercolumns in primate visual cortex de- 425427. velop in the absence of cues from photoreceptors. Proc Nat1 Acad Rakic P (1976) Prenatal genesis of connections subserving ocular dom- Sci USA 875303-5306. inance in the rhesus monkey. Nature 261:467-471. Lauder JM (1993) Neurotransmitters as growth regulator signals: role Rakic P (1977a) Prenatal development of the visual system in the rhe- of receptors and second messengers. Trends Neurosci 16:233-240. sus monkey. Philos Trans R Sot Lond [Biol] 278:245-260. Lauder JM, Wallace JA, Krebs H, Petrusz P McCarthy K (1982) In Rakic P (1977b) Genesis of the lateral geniculate nucleus in the rhesus vivo and in vitro development of serotonergic neurons. Brain Res monkey. J Comp Neurol 176:23-52. Bull 9:605-625. Rakic P (1988) Specification of cerebral cortical areas. Science 241: LaVail MM, Rapaport DH, Rakic P (1991) Cytogenesis in the monkey 170-176. retina. J Comp Neural 309:8&l 14. Rakic P (1991) Radial unit hypothesis of cerebral cortical evolution. Levitt PR, Rakic P, Goldman-Rakic PS (1984) Region-specific distri- Exp Brain Res [Suppl] 21:25+3. bution of catecholamine afferents in primate cerebral cortex: a fluo- Rakic P Goldman-Rakic PS (1985) Use of fetal neurosurgery for ex- rescence histochemical analysis. J Comp Neural 225:1-14. perimental studies of structural and functional brain development in Lidow MS, Rakic P (1992) Postnatal development of monoaminergic nonhuman primates. In: Prenatal neurology and neurosurgery 2574 Rakic and Lidow - Development of Monoamine Receptors

(Thompson RA, Green JR, Johnson SD, eds), pp 1-15. New York: Veraat C, DeVolder AG, Wanet-Defalque MC, Bol A, Michel C, Gof- Spectrum. finet AM (1990) Glucose utilization in human visual cortex is ab- Rakic P Riley KP (1983) Overproduction and elimination of retinal normally elevated in blindness of early onset but decreased in blind- axons in the fetal rhesus monkey. Science 209:1441-1444. ness of late onset. Brain Res 510:115-164. Rakic P Gallager D, Goldman-Rakic PS (1988) Area1 and laminar Walsh C, Schliebs R, Bigi V (1989) Effect of early visual pattern distribution of major neurotransmitter receptors in the monkey visual deprivation on development and laminar distribution of cholinergic cortex. J Neurosci 83670-3690. markers in rat visual cortex. In: Central cholinergic synaptic trans- Rakic l?, SuRer I, Williams RW (1991) A novel cytoarchitectonic area mission (Frotscher M, Misgeld V, eds), pp 295-304. Basel: Birkhau- induced experimentally within the primate visual cortex. Proc Nat1 ser. Acad Sci USA 88:2083-2087. Whitaker-Azmatia P (1991) Role of serotonin and other neurotrans- Reynolds I, Bear MF (1989) NMDA receptor development in the visual mitter receptors in brain development: basis for developmental phar- cortex of cats. Sot Neurosci Abstr 15:2. macolony. Pharmacol Rev 43:553-561. Schein ST, deMonasterio FM (1987) Mapping of retinal and geniculate Wiesel Tg (1982) Postnatal development of the visual cortex and in- neurons onto striate cortex of macaque. J Neurosci 7:996-1009. fluence of environment. Nature 299:583-591. Schlaggar BL, Fox K, O’Leary DDM (1994) Postsynaptic control of Wiesel TN, Hubel D (1966) Spatial and chromatic interactions in the plasticity in developing somatosensory cortex. Nature 364:623-626. lateral geniculate body of the rhesus monkey. J Neurophysiol 29: Schliebs R, Big1 V, Beisold D (1982a) Development of muscarinic 11150-11156. receptor binding in the visual system of monocularly deprived and Wikler KC, Rakic P (1991) Emergence of the photoreceptor mosaic dark-reared rats. Neurosci Res 7:1181-l 198. from a protomap of early-differentiating cones in the primate retina Schliebs R, Burgoyne RD, Big1 V (1982b) The effect of visual depri- Nature 35 1:397400. Wikler KS, Rakic P (1994) An array of early-differentiating cones vation of B-adrenergic receptors in the visual centers of the rat brain. precedes the emergence of the photoreceptor mosaic in the fetal mon- J Neurochem 38:1038-1043. key retina. Proc Nat1 Acad Sci USA 91:6534-6538. Shatz CJ (1990) Impulse activity and patterning of connections during Wilkinson M, Shaw C, Khan I, Cynader M (1983) Ontogeny of B-ad- CNS development. Neuron 5:745-756. renergic binding sites in kitten visual cortex and the effects of visual Shaw C, Cynader M (19.88) Unilateral eyelid suture increases GABA, deprivation. Dev Brain Res 7:349-352. receptors in visual cortex. Dev Brain Res 40: 148-153. Williams RW, Rakic P (1988) Elimination of neurons in the rhesus Sokal RR, Rohlf FJ (1981) Biometry. New York: Friedman. monkey’s lateral geniculate nucleus during development. J Comp Teller DY, Boothe R (1979) Development of vision in infant primates. Neurol 272~424-436. Trans Ophthalmol Sot UK 99:333-337. Williams RW, Borodkin M, Rakic P (1991) Growth cone distribution Ts’o DY, Gilbert CD (1988) The organization of chromatic and spatial patterns in optic nerve of fetal monkeys: implication for mechanisms interactions in the primate striate cortex. J Neurosci 8:1712-1727. of axonal guidance. J Neurosci 11:1081-1094.