The pathways governing signal transduction in the mesocortical and nigrostriatal dopamine systems of the brain are of central importance in a variety of drug actions and neurological diseases. We have analyzed the regional, cellular, and subcellular distribution of the closely related D1 and D5 subtypes of dopamine receptors in the cerebral cortex and selected subcortical structures of rhesus monkey using subtype specific antibodies. The distribution of D1 and D5 receptors was highly differentiated in subcortical structures. In the neostriatum, both D1 and to a lesser extent D5 antibodies labeled medium spiny neurons, while only D5 antibodies labeled the large aspiny neurons typical of cholinergic interneurons. In the caudate nucleus, D1 labeling was concentrated in the spines and shafts of projection neurons, whereas D5 antibodies predominantly labeled the shafts, and less commonly, the spines of these cells. The D1 receptor was abundantly expressed in the neuropil of the substantia nigra pars reticulata while the D5 antibodies labeled only a few scattered cell bodies in this structure. Conversely, D5 antibodies labeled cholinergic neurons in the basal forebrain more intensely than D1 antibodies. Within the cerebral cortex and hippocampus, D1 and D5 antibody labeling was prominent in pyramidal cells. Double-label experiments revealed that the two receptors were frequently coexpressed in neurons of both structures. Ultrastructurally, D1 receptors were especially prominent in dendritic spines whereas dendritic shafts were more prominently labeled by the D5 receptor. The anatomical segregation of the D1 and D5 receptors at the subcellular level in cerebral cortex and at the cellular level in subcortical areas suggest that these closely related receptors may be preferentially associated with different circuit elements and may play distinct regulatory roles in synaptic transmission.

We investigated the convergence of somatosensory and auditory inputs in within subregions of macaque auditory cortex. Laminar current source density and multiunit activity profiles were sampled with linear array multielectrodes during penetrations of the posterior superior temporal plane in three macaque monkeys. At each recording site, auditory responses to binaural clicks, pure tones, and band-passed noise, all presented by earphones, were compared with somatosensory responses evoked by contralateral median nerve stimulation. Subjects were awake but were not required to discriminate the stimuli. Borders between A1 and surrounding belt regions were identified by mapping best frequency and stimulus preferences and by subsequent histological analysis. Regions immediately caudomedial to A1 had robust somatosensory responses co-represented with auditory responses. In these regions, both somatosensory and auditory response profiles had "feedforward" patterns; initial excitation beginning in Lamina 4 and spreading to extragranular laminae. Auditory and somatosensory responses displayed a high degree of temporal overlap. Anatomical reconstruction indicated that the somatosensory input region includes, but may not be restricted to, the caudomedial auditory association cortex. As was earlier reported for this region, auditory frequency tuning curves were broad and band-passed noise responses were larger than pure tone responses. No somatosensory responses were observed in A1. These findings suggest a potential neural substrate for multisensory integration at an early stage of auditory cortical processing.

Antibodies to the Di dopamine receptor were used to localize this protein in several areas of human and monkey cerebral cortex with light and electron microscopy. In addition to cell body labeling in monkeys, all areas of humans and monkeys had a neuropil label with a laminar distribution predicted by previous Di receptor autoradiography studies.Using electron microscopy, this neuropil label was seen in numerous dendritic spines, in dendritic shafts, and in occasional axon terminals. While labeled spines were common, they represented only a subset of all cortical spines. Serial sectioning through labeled spines showed that the diaminobenzidine reaction product was usually not at postsynaptic densities but instead was displaced to the side of the large asymmetric (presumed glutamatergic) synapse (12) or by local diffusion of dopamine in the neuropil (13,14).The recent production of receptor subtype-specific antibodies (15)(16)(17)47) provides tools to identify the cell types and neural processes that contain dopamine receptors. Initial studies in rats have demonstrated the expected high density of D1 and D2 immunoreactivity in the basal ganglia (15)(16)(17)47). In the primate cerebral cortex, mRNA expression studies (18)(19)(20)(21)(22) and ligand binding studies (23,24,48) predict that D1, D4, and D5 receptors are comparatively dense, with less D2 receptors and perhaps no D3 receptors. Here we demonstrate the distribution ofD1 dopamine receptor immunoreactivity in the human and monkey cerebral cortex. Aside from the potential clinical relevance, primate cortex is advantageous for these studies because it has a high density of dopamine innervation compared to rodent cortex (25). MATERIALS AND METHODSTwo adult and one juvenile (14 months) macaque monkeys (Macaca mulatta) were perfused with 4% paraformaldehyde/0.08% glutaraldehyde/0.2% picric acid in phosphate buffer (PB) (pH 7.4) and processed as described (1) Two primary antibodies were used, a rat monoclonal and a rabbit polyclonal antibody, both directed to the C-terminal 97 amino acids of the human D1 dopamine receptor fused to a polypeptide fragment of glutathione S-transferase (GST) (17). Immunolocalization was done with ABC-Elite kits from Vector Laboratories visualized by exposure to 0.03% diaminobenzidine, 0.01% H202 or with the diaminobenzidine glucose oxidase reaction (26). For double labeling, tissue already labeled for dopamine receptors was treated with anti-tyrosine hydroxylase (TH) antibody, which was visualized with silverenhanced 1 nM gold-conjugated secondary antibodies (Amersham) as described (27). Rat anti-D1 receptor antibody was visualized in the same tissue as rabbit anti-TH antibody (Pel-Freez Biologicals) and rabbit anti-D1 antibody was combined with mouse anti-TH antibody (Chemicon).Receptor labeling was processed simultaneously with the following controls: (i) Primary antibodies were omitted. (ii) Primary antibodies were preadsorbed with the D1-GST fusion protein (0.5 mg/ml) conjugated to Affi-Gel beads (BioRad) (17). (iii) Pri...

The prevailing hierarchical model of cortical sensory processing holds that early processing is specific to individual modalities and that combination of information from different modalities is deferred until higher-order stages of processing. In this paper, we present physiological evidence of multisensory convergence at an early stage of cortical auditory processing. We used multi-neuron cluster recordings, along with a limited sample of single-unit recordings, to determine whether neurons in the macaque auditory cortex respond to cutaneous stimulation. We found coextensive cutaneous and auditory responses in caudomedial auditory cortex, an area lying adjacent to A1, and at the second stage of the auditory cortical hierarchy. Somatosensory-auditory convergence in auditory cortex may underlie effects observed in human studies. Convergence of inputs from different sensory modalities at very early stages of cortical sensory processing has important implications for both our developing understanding of multisensory processing and established views of unisensory processing.

The caudal medial auditory area (CM) has anatomical and physiological features consistent with its role as a first-stage (or "belt") auditory association cortex. It is also a site of multisensory convergence, with robust somatosensory and auditory responses. In this study, we investigated the cerebral cortical sources of somatosensory and auditory inputs to CM by injecting retrograde tracers in macaque monkeys. A companion paper describes the thalamic connections of CM (Hackett et al., J. Comp. Neurol. [this issue]). The likely cortical sources of somatosensory input to CM were the adjacent retroinsular cortex (area Ri) and granular insula (Ig). In addition, CM had reliable connections with areas Tpt and TPO, which are sites of multisensory integration. CM also had topographic connections with other auditory areas. As expected, connections with adjacent caudal auditory areas were stronger than connections with rostral areas. Surprisingly, the connections with the core were concentrated along its medial side, suggesting that there may be a medial-lateral division of function within the core. Additional injections into caudal lateral auditory area (CL) and Tpt showed similar connections with Ri, Ig, and TPO. In contrast to CM injections, these lateral injections had inputs from parietal area 7a and had a preferential connection with the lateral (gyral) part of Tpt. Taken together, the findings indicate that CM may receive somatosensory input from nearby areas along the fundus of the lateral sulcus. The differential connections of CM compared with adjacent areas provide additional evidence for the functional specialization of the individual auditory belt areas.

Deposits of diffuse beta-amyloid (Abeta) may exist in the brain for many years before leading to neuritic degeneration and dementia. The factors that contribute to the putative transformation of the Abeta amyloid from a relatively inert to a pathogenic state remain unknown and may involve interactions with additional plaque constituents. Matching brain sections from 2 demented and 4 nondemented subjects were processed for the demonstration of Abeta immunoreactivity, butyrylcholinesterase (BChE) enzyme activity, and thioflavine S binding. Additional sections were processed for the concurrent demonstration of two or three of these markers. A comparative analysis of multiple cytoarchitectonic areas processed with each of these markers indicated that Abeta plaque deposits are likely to undergo three stages of maturation, ie, a "diffuse" thioflavine S-negative stage, a thioflavine S-positive (ie, compact) but nonneuritic stage, and a compact neuritic stage. A multiregional analysis showed that BChE-positive plaques were not found in cytoarchitectonic areas or cortical layers that contained only the thioflavine S-negative, diffuse type of Abeta plaques. The BChE-positive plaques were found only in areas containing thioflavine S-positive compact plaques, both neuritic and nonneuritic. Within such areas, almost all (>98%) BChE-containing plaques bound thioflavine S, and almost all (93%) thioflavine S plaques contained BChE. These results suggest that BChE becomes associated with amyloid plaques at approximately the same time that the Abeta deposit assumes a compact beta-pleated conformation. BChE may therefore participate in the transformation of Abeta from an initially benign form to an eventually malignant form associated with neuritic tissue degeneration and clinical dementia.

Dopamine projections to the cerebral cortex have been implicated in normal and pathological cognitive processes, notably, Parkinson's disease and schizophrenia. To help elucidate the function of these dopamine axons, they were characterized by serial section electron microscopy in individual layers of monkey prefrontal cortex. Dopamine immunoreactivity was visualized with a silver precipitation technique that allowed clear resolution of the internal structures and cell membranes of labeled axons. Apart from the occasional large microtubule-filled axon, dopamine axons were thin and varicose with many clear synaptic vesicles and fewer dense-core vesicles. With few exceptions, dopamine synapses were symmetric and quite small, seen in only one to three serial sections. A determination of the "synaptic incidence" showed that only 39% of labeled varicosities formed identifiable synapses. However, it is certain that some small synapses could not be visualized even in serial sections, and it is possible that the vast majority if not all varicosities form synapses. Except for one soma, dendritic spines and shafts were the recipients of dopamine synapses. Many postsynaptic shafts were small and spiny, indicating that they were distal pyramidal dendrites. However, some postsynaptic shafts especially in supragranular layers had distinctly nonpyramidal features. These lacked spines, had a high density of synaptic inputs, and often had a strikingly varicose morphology. The data suggest that the majority of dopamine synapses in all layers are on pyramidal cells, but that a significant fraction are on presumed GABAergic nonpyramidal cells.

Recent studies of macaque monkey auditory cortex have revealed convergent auditory and somatosensory activity in the caudomedial area (CM) of the belt region. In the present study and its companion (Smiley et al., J. Comp. Neurol. [this issue]), neuroanatomical tracers were injected into CM and adjacent areas of the superior temporal plane to identify sources of auditory and somatosensory input to this region. Other than CM, target areas included: A1, caudolateral belt (CL), retroinsular (Ri), and temporal parietotemporal (Tpt). Cells labeled by injections of these areas were distributed mainly among the ventral (MGv), posterodorsal (MGpd), anterodorsal (MGad), and magnocellular (MGm) divisions of the medial geniculate complex (MGC) and several nuclei with established multisensory features: posterior (Po), suprageniculate (Sg), limitans (Lim), and medial pulvinar (PM). The principal inputs of CM were MGad, MGv, and MGm, with secondary inputs from multisensory nuclei. The main inputs of CL were Po and MGpd, with secondary inputs from MGad, MGm, and multisensory nuclei. A1 was dominated by inputs from MGv and MGad, with light multisensory inputs. The input profile of Tpt closely resembled that of CL, but with reduced MGC inputs. Injections of Ri also involved CM but strongly favored MGm and multisensory nuclei, with secondary inputs from MGC and the inferior division (VPI) of the ventroposterior complex (VP). The results indicate that the thalamic inputs of areas in the caudal superior temporal plane arise mainly from the same nuclei, but in different proportions. Somatosensory inputs may reach CM and CL through MGm or the multisensory nuclei but not VP.