Medium-sized Aspiny Interneurons Research Articles

NNOS(+) striatal neurons, a subpopulation spared in Huntington's Disease, possess functional NMDA receptors but fail to generate mitochondrial ROS in response to an excitotoxic challenge

Huntington's disease (HD) is a neurodegenerative condition characterized by severe neuronal loss in the cortex and striatum that leads to motor and behavioral deficits. Excitotoxicity is thought to be involved in HD and several studies have indicated that NMDA receptor (NMDAR) overactivation can play a role in the selective neuronal loss found in HD. Interestingly, a small subset of striatal neurons (less than 1% of the overall population) is found to be spared in post-mortem HD brains. These neurons are medium-sized aspiny interneurons that highly express the neuronal isoform of nitric oxide synthase (nNOS). Intriguingly, neurons expressing large amounts of nNOS [hereafter indicated as nNOS(+) neurons] show reduced vulnerability to NMDAR-mediated excitotoxicity. Mechanisms underlying this reduced vulnerability are still largely unknown and may shed some light on pathogenic mechanisms involved in HD. One untested possibility is that nNOS(+) neurons possess fewer or less functioning NMDARs. Employing single cell calcium imaging we challenged this hypothesis and found that cultured striatal nNOS(+) neurons show NMDAR-evoked responses that are identical to the ones observed in the overall population of neurons that express lower levels of nNOS [nNOS(−) neurons]. NMDAR-dependent deregulation of intraneuronal Ca2+ is known to generate high levels of reactive oxygen species of mitochondrial origin (mt-ROS), a crucial step in the excitotoxic cascade. With confocal imaging and dihydrorhodamine (DHR; a ROS-sensitive probe) we compared mt-ROS levels generated by NMDAR activation in nNOS(+) and (−) cultured striatal neurons. DHR experiments revealed that nNOS(+) neurons failed to produce significant amounts of mt-ROS in response to NMDA exposure, thereby providing a potential mechanism for their reduced vulnerability to excitotoxicity and decreased vulnerability in HD.

Distribution and chemical coding pattern of somatostatin immunoreactivity in the dorsal striatum of the guinea pig

The present study provides a detailed description of somatostatin (SOM) distribution and the colocalization pattern of SOM, neuropeptide Y (NPY) and nitric oxide synthase (NOS) in the dorsal striatum (caudate-putamen complex) of the guinea pig. Within the dorsal striatum, SOM is found in a population of medium-sized aspiny interneurons. We found that 97% of all SOM-IR neurons expressed NPY simultaneously, while 98% of all NPY-ergic perikarya was simultaneously SOM-IR. On the other hand, while 98% of all SOM-IR cells were simultaneously NOS-IR, only 91% of all NOS-containing neurons exhibited SOM-immunoreactivity. Irrespective of their chemical coding, both types of SOM-IR neurons were scattered throughout the dorsal striatum, sometimes in the form of small, loosely arranged clusters of 2-4 cells. While SOM-IR and NPY-IR nerve fibers were present in all of the studied regions, they were more numerous in the ventro-medial part of the studied structure, with the exception of its caudal portion, where SOM-IR and NPY-IR fibers additionally formed a dense network in the part corresponding to the caudate nucleus. A low expression of staining for NOS-IR fibers was seen throughout the entire dorsal striatum. In some fibers, SOM and NPY were co-expressed. Fibers expressing both SOM and NOS were not found.

Colocalization of Somatostatin Receptors with DARPP-32 in Cortex and Striatum of Rat Brain

Somatostatin (SST)-positive medium-sized aspiny interneurons are selectively spared in excitotoxicity. The biological effects of SST are mediated via five different receptors, namely somatostatin receptor (SSTR)1-5; however, SSTR subtype spared in excitotoxicity and involved in neuroprotection is not known. Dopamine- and cAMP-regulated phosphoprotein (DARPP-32) is predominantly expressed in medium-sized projection neurons that are most vulnerable in excitotoxicity. In the present study, we determined the colocalization of SST and SSTRs with DARPP-32 in rat brain cortical and striatal regions using immunofluorescence immunohistochemistry. We also determined the expression of DARPP-32 in SSTR1-5 immunoprecipitate prepared from cortex and striatum. SST-positive neurons in cortex and striatum are devoid of colocalization with DARPP-32. However, in cortical and striatal brain regions, three different neuronal populations either expressing SSTRs and DARPP-32 alone or displaying colocalization were identified. Quantitative analysis reveals that in cortex and striatum, SSTR1 and 5 are most predominant receptor subtypes colocalized with DARPP-32 followed by SSTR4, 2, and 3 in cortex whereas SSTR2, 4, and 3 in striatum. Importantly, DARPP-32 is expressed in SSTR1-5 immunoprecipitate prepared from cortex and striatum. Taken together, these results provide the first evidence that the SSTR-positive neurons lacking colocalization with DARPP-32 might be spared in excitotoxicity.

Somatostatin in Medium-Sized Aspiny Interneurons of Striatum is Responsible for Their Preservation in Quinolinic Acid and N-Methyl-d-Asparate-Induced Neurotoxicity

Somatostatin (SST) is a multifunctional peptide and involves in several neurodegenerative diseases. N-Methyl-D-asparate (NMDA) receptor agonist quinolinic acid (QUIN)-induced neurotoxicity mimics an experimental model of Huntington's disease that is characterized by the selective preservation of medium-sized aspiny interneurons and degeneration of medium-sized spiny projection neurons in striatum. In QUIN- and NMDA-induced neurotoxicity, increased expression of SST and messenger RNA levels along with SST release in culture medium is generally observed. However, the molecular mechanisms and the functional consequences of increased SST are still obscure. In the present study, the role of SST was determined using immunoneutralization and immunoblockade of SST in cultured striatal neurons upon QUIN- and NMDA-induced neurotoxicity. The immunoblockade of SST with antisense oligonucleotides and immunoabsorption of released SST with specific antibodies potentiate QUIN- and NMDA-induced neuronal cell death. NADPH-diaphorase positive neurons that are selectively spared in several processes of neurodegeneration result in severe damage upon immunoblockade or immunoabsorption of SST. In addition, exogenous SST along with QUIN and NMDA provides selective preservation of projection neurons, which are selectively susceptible in excitotoxicity. Neuroprotective effect of SST is completely blocked by pertussis toxins, suggesting the role of somatostatin receptors. Taken together, these results provide first evidence that the presence of SST is a unique feature for the selective sparing of medium sized aspiny interneurons in excitotoxicity.

Distribution of GABAergic neurons in the striatum of amygdala-kindled rats: An immunohistochemical and in situ hybridization study

A large body of experimental evidence suggests that the basal ganglia circuitry may be part of a remote control system modulating the spread of epileptic seizures. In the kindling model of temporal lobe epilepsy, this endogenous inhibitory control mechanism seems to be impaired. Neurochemical and neurophysiological studies have indicated that the activity of the GABAergic projection from the striatum to the substantia nigra pars reticulata is reduced in kindled rats, but the exact mechanisms involved in this observation are not known. Possible explanations include a kindling-induced loss of striatal GABAergic projection neurons to the substantia nigra or enhanced inhibition of these neurons by GABAergic interneurons. In the present experiments, the GABAergic system of the striatum (caudate–putamen) of amygdala-kindled rats and controls was studied immunohistochemically with a monoclonal antibody to GABA and with nonisotopic in situ hybridization with cRNA probes selective for glutamic acid decarboxylase 65 (GAD65) and GAD67, respectively. Compared to sham controls, an increased density of neurons heavily labeled for GAD67 mRNA was observed in the anterior striatum of kindled rats when cells were counted 6 weeks after the last kindled seizure. This subgroup of striatal GABAergic neurons has been suggested previously to correspond to the medium-sized aspiny interneurons in the striatum, indicating that kindling is associated with an increased activity of these neurons. Our previous finding of reduced GAD and GABA levels in synaptosomes isolated from the substantia nigra of kindled rats together with the present observation of increased density of GABAergic striatal interneurons in such rats suggest that kindling affects the regulation of the GABAergic projections from the striatum to the substantia nigra rather than directly damaging GABAergic neurons in the striatum.

Morphological features of neurons containing calcium‐binding proteins in the human striatum

An immunohistochemical approach was used to characterize the morphological phenotype of neurons containing the calcium-binding proteins calretinin (CR), parvalbumin (PV), or calbindin-D28k (CB) in the normal human striatum. The protein CR occurs in at least four morphologically distinct types of neurons. Apart from the numerous medium-sized aspiny interneurons and the less abundant giant aspiny interneurons, CR also labels some medium-sized spiny neurons morphologically identical to striatal projection neurons. This finding indicates that CR is not only confined to striatal interneurons but also may be involved in the function of certain projection neurons. Some small and peculiar bushy-like aspiny neurons also are enriched with CR. These neurons could correspond to the dwarf or neurogliform neurons first described by Ramón y Cajal (1911). Three types of PV-immunoreactive striatal neurons can be visualized in the human striatum: 1) the common medium-sized aspiny leptodendritic neurons, 2) some smaller and profusely arborized aspiny neurons, and 3) a few large and intensely stained neurons with conspicuously beaded and poorly branched dendrites. The protein CB labels virtually all medium-sized spiny projection neurons located in the striatal matrix but also identifies a small subset of large and more intensely immunostained aspiny neurons. The latter finding indicates that CB is not entirely confined to striatal projection neurons but also may play a role in local circuit neurons. These normative data should help our understanding of the chemical anatomy of the human striatum in both health and disease.

Morphological features of neurons containing calcium‐binding proteins in the human striatum

An immunohistochemical approach was used to characterize the morphological phenotype of neurons containing the calcium-binding proteins calretinin (CR), parvalbumin (PV), or calbindin-D28k (CB) in the normal human striatum. The protein CR occurs in at least four morphologically distinct types of neurons. Apart from the numerous medium-sized aspiny interneurons and the less abundant giant aspiny interneurons, CR also labels some medium-sized spiny neurons morphologically identical to striatal projection neurons. This finding indicates that CR is not only confined to striatal interneurons but also may be involved in the function of certain projection neurons. Some small and peculiar bushy-like aspiny neurons also are enriched with CR. These neurons could correspond to the dwarf or neurogliform neurons first described by Ramón y Cajal (1911). Three types of PV-immunoreactive striatal neurons can be visualized in the human striatum: 1) the common medium-sized aspiny leptodendritic neurons, 2) some smaller and profusely arborized aspiny neurons, and 3) a few large and intensely stained neurons with conspicuously beaded and poorly branched dendrites. The protein CB labels virtually all medium-sized spiny projection neurons located in the striatal matrix but also identifies a small subset of large and more intensely immunostained aspiny neurons. The latter finding indicates that CB is not entirely confined to striatal projection neurons but also may play a role in local circuit neurons. These normative data should help our understanding of the chemical anatomy of the human striatum in both health and disease. J. Comp. Neurol. 390:552–563, 1998. © 1998 Wiley-Liss, Inc.

Distribution and neurochemical character of substance P receptor (SPR)-immunoreactive striatal neurons of the macaque monkey: accumulation of SP fibers and SPR neurons and dendrites in "striocapsules" encircling striosomes.

The striatal distribution of the substance P receptor (SPR) protein was examined in relation to its ligand, the neuro-peptide SP, as well as to the neurochemical and compartmental composition of the neostriatum in rhesus monkeys (Macaca mulatta) in immunohistochemical experiments. About 2% of striatal neurons, displaying varicose, virtually spine-free dendrites characteristic of large and medium-sized aspiny interneurons, expressed SPR immunoreactivity. SPR/choline acetyltransferase, SPR/somatostatin, SPR/GABA, SPR/calbindin D28k, and SPR/parvalbumin double immunolabeling experiments demonstrated that SPR-positive cells are either cholinergic or somatostatinergic. Comparison of SP and SPR immunoreactivities in double-labeled and adjacent single-labeled sections revealed compartment-specific match and mismatch between the densities of the peptide and receptor. A matching high density of SP fibers and SPR cells and dendrites was only observed in the rim of the striosome compartments. To our knowledge, this is the first evidence for an anatomical border comprised of dendritic processes that separate striatal compartments. We have termed these zones "striocapsules," because they encircle and encapsulate striosomal cell islands. In the striatal matrix, an abundance of SPR-labeled profiles was complemented with light SP staining. By contrast, in the core of the striosomes, SPR labeling was sparse and SP staining intense. SP-positive axon-like puncta frequently contacted SPR-positive dendrites in all striatal compartments. The SP receptor/ligand match indicates a sharp increase in the efficacy of SP action in the striocapsules, and suggests that the influence of SP might be heightened in this striatal subcompartment.

Presynaptic and postsynaptic subcellular localization of substance P receptor immunoreactivity in the neostriatum of the rat and rhesus monkey (Macaca mulatta).

The substance P receptor (SPR) gene is expressed at high levels in basal ganglia, but the paucity of information about localization of the encoded receptor protein has limited our understanding of this peptide's involvement in cellular and subcellular mechanisms in this region. Morphological evidence in the rodent striatum indicates that SPRs are expressed in postsynaptic neuronal elements, while pharmacological studies suggest the existence of presynaptic SPRs in this structure. We have examined the issue of subcellular distribution of this receptor protein in rat and primate neostriatal tissue, employing an antiserum raised against SPR. Electron microscopic analysis revealed that SPR immunoreactivity is present in presynaptic and postsynaptic neuronal elements in both species. In agreement with earlier studies, SPR immunoreactivity was found predominantly in perikarya and dendrites of a small subset of striatal neurons, the large and medium-sized aspiny interneurons. In addition, a small but significant proportion of the immunoreaction product was localized in presynaptic profiles, both in axons and axon terminals. The majority of SPR immunoreactive boutons formed asymmetric synapses with dendrites and dendritic spines. The association of SPRs with asymmetric synapses provides a morphological substrate for peptidergic modulation of excitatory neurotransmission of extrastriatal origin. A minor proportion of immunolabeled axons established symmetric synaptic junctions with unlabeled dendrites. The presence of SPRs in these synapses suggests a presynaptic peptidergic modulation of intrinsic striatal transmitter systems. The observations in this study also indicate that SPR mediates a complex combination of postsynaptic and presynaptic effects on acetylcholine release in the mammalian striatum.

Differential Effects of Suicide Transport Lesions of the Striatonigral or Striatopallidal Pathways on Subsets of Striatal Neurons

In the basal ganglia, centrally active suicide transport agents produce apparently selective lesions of the striatopallidal and striatonigral pathways based on receptor binding and neuropeptide mRNA studies. In the present study we sought to determine the selectivity of suicide transport lesions for specific subsets of striatal neurons. Using immunohistochemical methods, the neostriata of adult rats were examined 10 days after an injection of volkensin into the substantia nigra or an injection of OX7-saporin into the globus pallidus, Ricin, a suicide transport agent active in the peripheral but not the central nervous system, was injected into each target as a control. Adjacent sections were processed for (1) Nissl stain to assess neuronal density, both overall and for large interneurons. (2) NADPH-diaphorase (NADPH-d) histochemistry, to mark medium-sized aspiny interneurons, (3) eukephalin immunocytochemistry, to label striatopallidal neurons, or (4) substance P immunocythochemistry, to label striatonigral neurons. Ricin injections produced no change in the densities of these subsets of striatal cells. In animals receiving volkensin or OX7-saporin injections, analyses of Nissl material revealed that the striata ipsilateral to the toxin injections appeared normal and did not exhibit shrinkage or gliosis; however, a quantitation analysis revealed a moderate decrease in cell density (12-16% loss, P < 0.01). The densities of both large and NADPH-d-containing striatal interneurons were unchanged after lesions in either target. Following nigral injections with volkensin, the density of striatal substance P-labeled cells decreased (26% loss, P < 0.01), while the density of enkephalin-labeled cells did not decrease significantly (11% decrease, P > 0.1). After pallidal injections with OX7-saporin, the density of striatal enkephalin-labeled cells decreased (20% loss, P < 0.01), while that of substance P-labeled cells remained unchanged. These data show that nigral volkensin and pallidal OX7-saporin injections differentially lesion striatonigral and striatopallidal projection neurons and spare striatal interneurons. This study provides further evidence for the selectivity, specificity, and utility of suicide transport agents to study brain structure and function.

Amino acid levels and gamma-aminobutyric acidA receptors in rat neostriatum, cortex, and thalamus after neonatal 6-hydroxydopamine lesion.

The amino acid gamma-aminobutyric acid (GABA) is the main inhibitory neurotransmitter in brain, and GABAergic neurons have been proposed to play a major role in basal ganglia physiology. In the neostriatum (caudate putamen), medium-sized aspiny interneurons, as well as neostriatal output neurons that project to several brain regions, use GABA as their neurotransmitter. Dopamine fibers arising from the substantia nigra represent a major input to the neostriatum where, besides their classic neurotransmitter role, they are seemingly involved in the regulation of amino acid neurotransmitter release. To further characterize the nature of some of the amino acid/dopamine interactions, selective dopaminergic deafferentations were produced in neonatal rats (3 days postnatal) by intraventricular administration of the neurotoxin 6-hydroxydopamine (6-OHDA); the noradrenergic neurons were protected by prior administration of desmethylimipramine. After a 3-month survival, levels of catecholamines, indoleamines, and amino acids were determined in cingulate cortex, thalamus, and neostriatum. In addition, GABAA receptors were measured in membrane preparations from these three regions, using the specific agonist [3H]muscimol. In the 6-hydroxydopamine-lesioned rats, levels of dopamine and its metabolites homovanillic acid, 3,4-dihydroxyphenylacetic acid, and 3-methoxytyramine were decreased, as expected, in cortex and neostriatum, but remained unmodified in thalamus. In all three regions, serotonin content was increased; its metabolite, 5-hydroxyindole-3-acetic acid, was also elevated, but only in cortex and neostriatum. The levels of GABA were increased in neostriatum and thalamus, but remained unmodified in cortex. Glycine was increased in all three regions examined. There were also increases of phosphatidylethanolamine and serine in thalamus, and of aspartic acid and alanine in neostriatum. The density of GABAA binding sites was increased in neostriatum, but remained unchanged in cortex and thalamus. The changes in amino acid levels and [3H]muscimol binding sites induced by a neonatal 6-hydroxydopamine treatment differ from those found after similar lesions in adult animals, possibly because of the plastic and synaptic rearrangements that can still occur during early postnatal development. The present results also demonstrate that adaptations occur in response to a dopaminergic deafferentation at an early age and that these exhibit a regional specificity.

The neostriatal mosaic: compartmental distribution of calcium-binding protein and parvalbumin in the basal ganglia of the rat and monkey.

Calcium-binding protein (CaBP) and parvalbumin are two proteins that are expressed in brain and bind calcium in the micromolar range. The immunohistochemical distribution of these two proteins was examined in the basal ganglia of rats and rhesus monkeys. In the striatum, CaBP immunoreactivity is localized to a subset of striatonigral projection neurons; CaBP-positive neurons are distributed in areas containing somatostatin-immunoreactive fibers and not in the complementary areas containing dense mu opiate-receptor binding. These biochemical labels mark, respectively, the matrix and patch compartments of the striatum. Previous studies have shown that striatal matrix neurons project to the substantia nigra pars reticulata, whereas striatal patch neurons project to the substantia nigra pars compacta. Consistent with the restricted localization of CaBP in the matrix projection neurons is the confinement of CaBP-immunoreactive afferent fibers to the pars reticulata. CaBP is also localized to a portion of dopaminergic and a few nondopaminergic neurons in the substantia nigra pars compacta and in most dopaminergic neurons in the ventral tegmental area. Parvalbumin immunoreactivity is localized to a subset of substantia nigra pars reticulata neurons and their axons. In the lateral striatum, some medium-sized aspiny interneurons are also parvalbumin immunoreactive. The distinct distributions of CaBP and parvalbumin in the basal ganglia are discussed in terms of their possible roles as intracellular calcium buffer systems related to the physiologic response properties of the neurons in which they are contained.

Striatal neurons containing both somatostatin- and avian pancreatic polypeptide (APP)-like immunoreactivities and NADPH-diaphorase activity: a light and electron microscopic study.

Striatal neurons containing both somatostatin- and avian pancreatic polypeptide (APP)-like immunoreactivities have been shown to be selectively stained by the histochemical method for nicotinamide adenine dinucleotide phosphate (NADPH)-diaphorase activity. In the present study, we have utilized this histochemical technique to examine the morphology of these striatal neurons at the light and electron microscopic levels. Our results indicate that the striatal somatostatin/APP/NADPH-diaphorase neurons occur throughout the striatum and have long, aspiny dendrites, oval, invaginated nuclei with prominent nucleoli, and receive few axosomatic contacts. These cells appear to correspond to a population of medium-sized aspiny interneurons reported previously in Golgi and electron microscopic studies of the striatum.