Differential distribution of nicotinamide adenine dinucleotide phosphate-diaphorase and neural nitric oxide synthase in the rat choroid plexus. A histochemical and immunocytochemical study.

This study used NADPH diaphorase (NADPHd) histochemistry and neuronal nitric oxide synthase immunocytochemistry to examine the localization of nitric oxide synthase in the choroid plexus of the lateral ventricles and the fourth ventricle of rat brain. That the NADPHd reaction product in choroid plexus was specific to nitric oxide synthase was evaluated: (i) by comparison to immunocytochemical labelling for nitric oxide synthase; and (ii) by comparing NADPHd histochemical staining in choroid plexus and brain (rich in nitric oxide synthase-positive and NADPHd-positive neurons) in the presence or absence of iodonium diphenyl or dichlorophenolindophenol, two potent albeit non-selective inhibitors of nitric oxide synthase activity. In brain, NADPHd histochemistry homogeneously stained neuronal cell bodies, axons and dendrites, while it produced particulate cytoplasmic staining of all epithelial cells in the choroid plexuses of the lateral and fourth ventricles. Within the choroid plexus of the lateral ventricles, NADPHd-positive nerve fibres were also observed around blood vessels and coursing among the epithelial cells. The distribution of immunoreactivity for nitric oxide synthase in brain and in nerve fibres in the choroid plexuses of the lateral ventricles resembled the distribution of histochemical labelling for NADPHd. Choroid plexus epithelial cells were, however, devoid of nitric oxide synthase immunoreactivity. Consistent with this, iodonium diphenyl and dichlorophenolindophenol (0.1 mM) inhibited NADPHd histochemical staining in brain neurons and in choroid plexus nerve fibres, but not in choroid plexus epithelial cells. These results demonstrate that the choroid plexus of the lateral ventricles in rat brain is innervated by nitric oxide synthase-positive nerve fibres. These nitric oxide synthase-positive nerve fibres may have an important role in the regulation of cerebrospinal fluid balance. Although choroid plexus epithelial cells contain an enzyme with NADPHd activity, this enzyme is not nitric oxide synthase.

Calretinin is largely localized to a unique population of striatal interneurons in rats.

Previous studies have reported the presence of the calcium binding protein calretinin in neurons in the striatal part of the basal ganglia in rats and primates. In the present study, immunofluorescence double-labeling techniques and immunofluorescence combined with retrograde labeling were used in rats to determine whether calretinin is found in any of the known types of striatal neurons. The results showed that a small fraction of the calretinin-containing neurons (< 10%) contain parvalbumin, but none of the calretinin-containing striatal neurons contained markers for the other two major types of striatal interneurons (i.e., choline acetyltransferase-containing cholinergic neurons and somatostatin-containing neurons). Additionally, calretinin was not found in projection neurons, using either calbindin or DARPP32 as immunofluorescent markers of striatal projections neurons in general, or using retrograde labeling to specifically identify either striatonigral or striatopallidal neurons. Thus, calretinin appears to be largely found in a unique population of striatal interneurons in rats. This population appears to be about one third the abundance of any of the previously identified populations of striatal interneurons.

Differential abundance of superoxide dismutase in interneurons versus projection neurons and in matrix versus striosome neurons in monkey striatum.

To investigate whether differences in vulnerability to free radicals might underlie differences among striatal neurons in their vulnerability to neurodegenerative processes such as occur in ischemia and Huntington's disease, we have analyzed the localization of superoxide free radical scavengers in different striatal neuron types in normal rhesus monkey. Single- and double-label immunohistochemical experiments were carried out using antibodies against the enzymes copper, zinc superoxide dismutase (SOD1), or manganese superoxide dismutase (SOD2), and against markers of various striatal cell types. Our results indicate that the striatal cholinergic and parvalbumin interneurons are enriched in SOD1 and/or SOD2, whereas striatal projection neurons and neuropeptide Y/somatostatin (NPY+/SS+) interneurons express only low levels of both SOD1 and SOD2. We also found that projection neurons of the matrix compartment express significantly higher levels of SOD than those in the striosome compartment. Since projection neurons have been reported to be more vulnerable than interneurons and striosome neurons more vulnerable than matrix neurons to neurodegenerative processes, our results are consistent with the notion that superoxide free radicals are at least partly involved in producing the differential neuron loss observed in the striatum following global brain ischemia or in Huntington's disease.

Colocalization of somatostatin, neuropeptide Y, neuronal nitric oxide synthase and NADPH-diaphorase in striatal interneurons in rats.

The neuropeptides somatostatin (SS), neuropeptide Y (NPY), the enzyme neuronal nitric oxide synthase (nNOS) and enzymatic activity for NADPH diaphorase (NADPHd) are extensively colocalized in striatal interneurons, which has led to the widespread tendency to operationally treat all four substances as being completely colocalized within a single class of striatal interneurons. We have explored the validity of this assumption in rat striatum using multiple-labeling methods. Conventional epi-illumination fluorescence microscopy was used to examine tissue triple labeled for SS, NPY and nNOS, or double-labeled for SS and nNOS or for SS and NPY. In tissue double-labeled for SS and nNOs, confocal laser scanning microscopy (CLSM) images of SS and nNOS labeling were compared to subsequent NADPHd labeling. We found that SS, NPY and nNOS co-occurred extensively, but a moderately abundant population of neurons containing SS and nNOS but not NPY was also observed, as were small populations of SS only and nNOS only neurons. About 80% of SS+ neurons contained NPY, and no NPY neurons were devoid of SS or nNOS. All neurons containing nNOS in rat striatum were found to contain NADPHd. Combining our various quantitative observations, we found that of those striatal neurons containing any combination of SS, NPY, nNOS and NADPHd in rats, about 73% contained all four, 16% contained SS, nNOS and NADPHd, 5% contained SS only, and 6% contained only nNOS and NADPHd. These results indicate that while there is a large population of striatal neurons in which SS, NPY, nNOS and NADPHd are colocalized in rats, there may be smaller populations of striatal neurons devoid of NPY in which SS or nNOS/NADPHd are found individually or together.

Brainstem motoneuron pools that are selectively resistant in amyotrophic lateral sclerosis are preferentially enriched in parvalbumin: evidence from monkey brainstem for a calcium-mediated mechanism in sporadic ALS.

Some brainstem motoneuron groups appear more resistant to the process of neurodegeneration in ALS (for example, oculomotor, trochlear, and abducens nuclei) than others (for example, trigeminal, facial, ambiguus, and hypoglossal nuclei). The possibility that the differential presence of the calcium-chelating protein parvalbumin might underlie this difference in vulnerability was examined immunohistochemically as a way to determine whether a calcium-mediated mechanism might be involved in ALS. In normal monkey brainstem, we found that the abundance of parvalbumin-containing neurons in the oculomotor, trochlear, and abducens nuclei was approximately 90% of the abundance of choline acetyltransferase (CHAT)-containing motoneurons. In contrast, the abundance of parvalbumin-containing neurons in the other brainstem motor nuclei innervating skeletal muscle (trigeminal, facial, ambiguus, and hypoglossal) was only about 30-60% of the abundance of CHAT-containing motoneurons. Since some of these motoneuron pools contain nonmotoneuron internuclear neurons that might be parvalbumin-containing, we also carried out double-label studies to specifically determine the percentage of cholinergic motoneurons that contained parvalbumin in each of these motoneuron pools. We found that 85-100% of the oculomotor, trochlear, and abducens motoneurons were parvalbumin-containing. In contrast, only 20-30% of the trigeminal, facial, ambiguus, and hypoglossal motoneurons were parvalbumin-containing. These results raise the possibility that motoneuron death in sporadic ALS is related to some defect that promotes cytosolic calcium accumulation in motoneurons. This excess calcium entry may promote cell death via an excitotoxic pathway. Motoneurons rich in parvalbumin may resist the deleterious effects of this putative calcium gating defect because they are better able to sequester the excess calcium.

Age-dependent differences in survival of striatal somatostatin-NPY-NADPH-diaphorase-containing interneurons versus striatal projection neurons after intrastriatal injection of quinolinic acid in rats.

Some authors have reported greater sparing of neurons containing somatostatin (SS)-neuropeptide Y (NPY)-NADPH-diaphorase (NADPHd) than projection neurons after intrastriatal injection of quinolinic acid (QA), an excitotoxin acting at NMDA receptors. Such findings have been used to support the NMDA receptor excitotoxin hypothesis of Huntington's disease (HD) and to claim that intrastriatal QA produces an animal model of HD. Other studies have, however, reported that SS/NPY/NADPHd interneurons are highly vulnerable to QA. We examined the influence of animal age (young versus mature), QA concentration (225 mM versus 50 mM), and injection speed (3 min versus 15 min) on the relative SS/NPY/NADPHd neuron survival in eight groups of rats that varied along these parameters to determine the basis of such prior discrepancies. Two weeks after QA injection, we analyzed the relative survival of neurons labeled by NADPHd histochemistry, SS/NPY immunohistochemistry, or cresyl violet staining (which stains all striatal neurons, the majority of which are projection neurons) in the so-called lesion transition zone (i.e., the zone of 40-60% neuronal survival). We found that age, and to a lesser extent injection speed, had a significant effect on relative SS/NPY/NADPHd interneuron survival. The NADPHd- and SS/NPY-labeled neurons typically survived better than projection neurons in young rats and more poorly in mature rats. This trend was greatly accentuated with fast QA injection. Age-related differences may be attributable to declines in projection neuron sensitivity to QA with age. Since rapid QA injections result in excitotoxin efflux, we interpret the effect of injection speed to suggest that brief exposure to a large dose of QA (with fast injection) may better accentuate the differential vulnerabilities of NADPHd/SS/NPY interneurons and projection neurons than does exposure to the same total amount of QA delivered more gradually (slow injection). These findings reconcile the discordant results found by previous authors and suggest that QA injected into rat striatum does reproduce the neurochemical traits of HD under some circumstances. These findings are consistent with a role of excitotoxicity in HD pathogenesis, and they also have implications for the basis of the more pernicious nature of striatal neuron loss in juvenile onset HD.

Differential abundance of glutamate transporter subtypes in amyotrophic lateral sclerosis (ALS)-vulnerable versus ALS-resistant brain stem motor cell groups.

Previous studies have suggested that defective high-affinity glutamate uptake, due mainly to a major loss of the astroglial-specific GLT-1 glutamate transporter, underlies the selective motoneuron degeneration observed in sporadic ALS (24, 28). If a defect in glutamate transport underlies the pathogenesis of sporadic ALS, the glutamate transporter subtype found to be lost in sporadic ALS should be present in abundance in the affected motor nuclei under normal conditions. To investigate this, we used immunohistochemical methods to analyze the localization of two subtypes of high-affinity glutamate transporters in the cranial motor nuclei of normal monkey brain stem: GLT-1, localized to astroglia; and EAAC1, localized to neurons. Our results indicated that all motor cell groups of monkey brain stem are rich in the GLT-1 glutamate transporter, which is localized to astroglial cells and processes that surround and envelop motoneuron cell bodies and dendrites. Image analysis indicated that the abundance of GLT-1 immunoreactive astroglial elements in ALS-vulnerable motor cell groups (i.e., the trigeminal, facial, and hypoglossal motor cell groups) is higher than in ALS-resistant motor cell groups (i.e., the oculomotor, trochlear, and abducens motor cell groups), and statistical analysis showed that this difference is significant. Our results also indicated that both ALS-vulnerable and ALS-resistant motor cell groups of monkey brain stem are relatively poor in EAAC1 immunoreactivity. Therefore, in the case of a loss in the GLT-1 glutamate transporter in sporadic ALS, glutamate may increase in the vicinity of motoneurons in all brain-stem motor cell groups, but especially in the ALS-vulnerable motor cell groups, which are normally richer in GLT-1. Increased extracellular glutamate could lead to excess entry of Ca2+ into motoneurons via glutamate-gated or voltage-activated Ca2+ channels and produce degeneration of those motoneurons unable to resist the insult. Since motoneurons in the ALS-resistant motor cell groups of the brain stem are enriched in the Ca2+ buffering protein parvalbumin, they should be better able to resist the damage than the majority of motoneurons in the ALS-vulnerable motor cell groups, which lack parvalbumin (20).

Relative survival of striatal projection neurons and interneurons after intrastriatal injection of quinolinic acid in rats.

An excitotoxic process mediated by the NMDA type glutamate receptor may be involved in striatal neuron death in Huntington's disease (HD). To explore this possibility, we have injected an NMDA-receptor-specific excitotoxin, quinolinic acid (QA), into the striatum in adult rats and 2-4 months postlesion explored the relative patterns of survival for the various different types of striatal projection neurons and interneurons and for the striatal efferent fibers in the different striatal projection areas. The perikarya of specific types of striatal neurons were identified by neurotransmitter immunohistochemical labeling or by retrograde labeling from striatal target areas, while the striatal efferent fiber plexuses were identified by neurotransmitter immunohistochemical labeling. The pattern of survival for the perikarya of each neuron type as a function of distance from the center of the injection site was determined, and the relative survival of each type was compared. For the fibers in target areas, computer-assisted image analysis was used to determine the degree of fiber loss for each projection target. In the study of perikaryal vulnerability, we found that the somatostatin-neuropeptide Y (SS/NPY) interneurons were the most vulnerable to QA and the cholinergic neurons were invulnerable to QA. The perikarya of all projection neuron types (striatopallidal, striatonigral, and striato-entopeduncular) were less vulnerable than the SS/NPY interneurons and more vulnerable than the cholinergic interneurons. Among projection neuron perikarya, there was evidence of differential vulnerability, with striatonigral neurons appearing to be the most vulnerable. Examination of immunolabeled striatal fibers in the striatal target areas indicated that striato-entopeduncular fibers better survived intrastriatal QA than did striatopallidal or striatonigral fibers. The apparent order of vulnerability observed in this study among projection neurons and/or their efferent fiber plexuses and the invulnerability observed in this study of cholinergic interneurons is similar to that observed in HD. The vulnerability of the SS/NPY interneurons to QA is, however, in stark contrast to their invulnerability in HD. The results thus suggest that although the excitotoxin hypothesis of striatal neuron death in HD has merit, QA injections into adult rat striatum do not strictly mimic the outcome in HD. This suggests that either adult rats are not a completely suitable subject for mimicking HD or the HD excitotoxic process does not involve a freely circulating excitotoxin such as QA.

Relative resistance of striatal neurons containing calbindin or parvalbumin to quinolinic acid-mediated excitotoxicity compared to other striatal neuron types.

To evaluate the relative ability of those striatal neuron types containing calbindin or parvalbumin to withstand a Ca(2+)-mediated excitotoxic insult, we injected the NMDA receptor-specific excitotoxin quinolinic acid (QA) into the striatum in mature adult rats and 2 months later examined the relative survival of striatal interneurons rich in parvalbumin and striatal projection neurons rich in calbindin. To provide standardization to the survival of striatal neuron types thought to be poor in Ca2+ buffering proteins, the survival was compared to that of somatostatin-neuropeptide Y (SS/NPY)-containing interneurons and enkephalinergic projection neurons, which are devoid of or relatively poorer in such proteins. The various neuron types were identified by immunohistochemical labeling for these type-specific markers and their relative survival was compared at each of a series of increasing distances from the injection center. In brief, we found that parvalbuminergic, calbindinergic, and enkephalinergic neurons all showed a generally comparable gradient of neuronal loss, except just outside the lesion center, where calbindin-rich neurons showed significantly enhanced survival. In contrast, striatal SS/NPY interneurons were more vulnerable to QA than any of these three other types. These observed patterns of survival following intrastriatal QA injection suggest that calbindin and parvalbumin content does not by itself determine the vulnerability of striatal neurons to QA-mediated excitotoxicity in mature adult rats. For example, parvalbuminergic striatal interneurons were not impervious to QA, while cholinergic striatal interneurons are highly resistant and SS/NPY+ striatal interneurons are highly vulnerable. Both cholinergic and SS/NPY+ interneurons are devoid of any known calcium buffering protein. Similarly, calbindin does not prevent striatal projection neuron vulnerability to QA excitotoxicity. Nonetheless, our data do suggest that calbindin may offer striatal neurons some protection against moderate excitotoxic insults, and this may explain the reportedly slightly greater vulnerability of striatal neurons that are poor in calbindin to ischemia and Huntington's disease.

Transient global ischemia in rats yields striatal projection neuron and interneuron loss resembling that in Huntington's disease.

The various types of striatal projection neurons and interneurons show a differential pattern of loss in Huntington's disease (HD). Since striatal injury has been suggested to involve similar mechanisms in transient global brain ischemia and HD, we examined the possibility that the patterns of survival for striatal neurons after transient global ischemic damage to the striatum in rats resemble that in HD. The perikarya of specific types of striatal interneurons were identified by histochemical or immunohistochemical labeling while projection neuron abundance was assessed by cresyl violet staining. Projectionneuron survival was assessed by neurotransmitter immunolabeling of their efferent fibers in striatal target areas. The relative survival of neuron types was determined quantitatively within the region of ischemic damage, and the degree of fiber loss in striatal target areas was quantified by computer-assisted image analysis. We found that NADPHd(+) and cholinergic interneurons were largely unaffected, even in the striatal area of maximal damage. Parvalbumin interneurons, however, were as vulnerable as projection neurons. Among immunolabeled striatal projection systems, striatoentopeduncular fibers survived global ischemia better than did striatopallidal or striatonigral fibers. The order of vulnerability observed in this study among the striatal projection systems, and the resistance to damage shown by NADPHd(+) and cholinergic interneurons, is similar to that reported in HD. The high vulnerability of projection neurons and parvalbumin interneurons to global ischemia also resembles that seen in HD. Our results thus indicate that global ischemic damage to striatum in rat closely mimics HD in its neuronal selectivity, which supports the notion that the mechanisms of injury may be similar in both.

Cellular localization of huntingtin in striatal and cortical neurons in rats: lack of correlation with neuronal vulnerability in Huntington's disease.

Immunohistochemistry and single-cell RT-PCR were used to characterize the localization of huntingtin and/or its mRNA in the major types of striatal neurons and in corticostriatal projection neurons in rats. Single-label immunohistochemical studies revealed that striatum contains scattered large neurons rich in huntingtin and more numerous medium-sized neurons moderate in huntingtin. Double-label immunohistochemical studies showed that the large huntingtin-rich striatal neurons include nearly all cholinergic interneurons and some parvalbuminergic interneurons. Somatostatinergic striatal interneurons, which are medium in size, rarely contained huntingtin. Calbindin immunolabeling showed that the vast majority of the medium-sized striatal neurons that contain huntingtin are projection neurons, but only approximately 65% of calbindin-labeled projection neurons (localized to the matrix compartment of striatum) were labeled for huntingtin. Calbindin-containing projection neurons of the matrix compartment and calbindin-negative projection neurons of the striatal patch compartment contained huntingtin with comparable frequency. Single-cell RT-PCR confirmed that striatal cholinergic interneurons contain huntingtin, but only approximately 65% of projection neurons contained detectable huntingtin message. The finding that huntingtin is not consistently found in striatal projection neurons [which die in Huntington's disease (HD)] but is abundant in striatal cholinergic interneurons (which survive in Huntington's disease) suggests that the mutation in huntingtin that causes HD may not directly kill neurons. In contrast to the heterogeneous expression of huntingtin in the different striatal neuron types, we found all corticostriatal neurons to be rich in huntingtin protein and mRNA. One possibility raised by our findings is that the HD mutation may render corticostriatal neurons destructive rather than render striatal neurons vulnerable.