Treatment of refractory ITP and Evans syndrome by haematopoietic cell transplantation: is it indicated, and for whom?

Several lines of therapy have been established for patients with immune thrombocytopenia (ITP) and Evans syndrome. However, these therapies generally require prolonged administration, lead to profound immunosuppression and increased infectious risk, and are often poorly tolerated. While most patients with these disorders will respond to first-line steroid therapy, others will prove refractory or intolerant to multiple treatments. In these patients (and possibly even selected patients who are not considered refractory), autologous or allogeneic haematopoietic stem cell transplantation (HCT) may provide definitive therapy. We review the literature on the treatment of ITP and Evans syndrome with HCT and discuss its use in the management of these disorders. We also pose, for the purpose of discussion, research questions that will be important to address if HCT is to be considered a viable option for more patients with these diseases.

Microtubules and filaments in the axons and astrocytes of early postnatal rat optic nerves.

Changes in the population of microtubules and filaments within the cytoplasm of maturing axons and astrocytes have been studied during the early postnatal development of rat optic nerves. At birth, all of the axons are unmyelinated; most have a diameter of 0.2-0.3 micro and contain 4-10 microtubules. Neurofilaments do not occur with any frequency until about 5 days postnatal when they appear as individual groups, each containing 4-12. Subsequently, the neurofilaments of each group disperse so that they become more evenly distributed in mature axons. Developing astrocytes show similar but rather more dramatic changes. Most astrocytic processes contain only microtubules at birth, but during maturation filaments begin to appear in increasing numbers while microtubules become less common. This process continues until, in the mature fibrous astrocytes, filaments pack the cytoplasm and microtubules are rare. These observations suggest that the filaments within axons and astrocytes may be formed by the breakdown of microtubules.

A quantitative study of synapses on motor neuron dendritic growth cones in developing mouse spinal cord.

The proportion of synaptic contacts occurring on dendrites as well as on dendritic growth cones and filopodia was determined from electron micrographs of developing mouse (C57BL/6J) spinal cord. Comparable areas of the marginal zone adjacent to the lateral motor nucleus were sampled from specimens on the 13th-16th days of embryonic development (E13-E16). At the beginning of this period, synapses upon growth cones and filopodia comprise about 80% of the observed synaptic junctions, but this proportion decreases with developmental time so that in E16 specimens growth cone synapses account for slightly less than 30% of the synaptic population. Conversely, at E13, synapses upon dendrites comprise less than 20% of the total number of synapses, but increase with developmental time so that they account for about 65% of the synaptic population of E16 specimens. From these data, we suggest the following temporal sequence for the formation of synaptic junctions on motor neuron dendrites growing into the marginal zone. New synapses are initially made upon the filopodia of dendritic growth cones. A synaptically contacted filopodium expands to become a growth cone while the original growth cone begins to differentiate into a dendrite. This process is repeated as the dendrite grows farther into the marginal zone so that synapses originally made with filopodia come to be located upon dendrites. This speculation is briefly discussed in relation to the work and ideas of others concerning synaptogenesis and dendritic development.

Inhibitory, GABAergic nerve terminals decrease at sites of focal epilepsy.

Using an immunocytochemical method for the localization of the gamma-aminobutyric acid (GABA) synthesizing enzyme, glutamic acid decarboxylase (GAD), we have observed GABAergic nerve terminals distributed throughout all layers of normal monkey sensorimotor cortex. These terminals displayed ultrastructural characteristics that suggested that they arose from aspinous and sparsely spinous stellate neurons. In monkeys (Macaca mulatta and M. fascicularis) made epileptic by cortical application of alumina gel, a highly significant numerical decrease of GAD-positive nerve terminals occurred at sites of seizure foci indicating a functional loss of GABAergic inhibitory synapses. A loss of such inhibition at seizure foci could lead to epileptic activity of cortical pyramidal neurons.

Commissural fibers may guide cholinergic neuronal migration in developing rat cervical spinal cord.

The present investigation examines the role of intercellular relationships in the guidance of neuronal migration in embryonic rat cervical spinal cord. A "U-shaped" group of cholinergic neurons, was first detected on embryonic days (E) 15.5-16 surrounding the ventral proliferative zone. At these stages, no cholinergic cells were observed in the dorsal spinal cord, but by E17, many of the "U-shaped" group of cholinergic cells appeared to have translocated dorsally, to become the cholinergic dorsal horn cells seen in older animals. Between E16 and E17, these choline acetyltransferase (ChAT)-immunoreactive cells displayed primitive processes oriented dorsoventrally, suggesting migration along that axis. Two early forming substrates present in embryonic spinal cord have been implicated in the guidance of other populations of migrating neurons: glial cells organized in radial arrays and commissural axons aligned along the dorsoventral axis. Involvement of the commissural fibers with cholinergic cell migration seems more likely because the fibers and the translocation pathway have similar orientations. In double-labeling immunocytochemical studies of E15.5-17 spinal cord, some immature ChAT-containing neurons were directly adjacent to commissural fibers, as identified by SNAP/TAG-1 immunoreactivity. The temporal and spatial coincidence of developing cholinergic neurons and commissural axons is consistent with the hypothesis that these neurons could use commissural fibers as migratory guides. In addition, conventional electron micrographs were examined to determine if immature neuronal profiles were physically apposed to commissural axons. Immature neurons with leading and trailing processes oriented dorsally and ventrally, respectively, were embedded within and aligned along bundles of commissural fibers or along other similarly oriented neurons. This direct apposition of immature cells to the surfaces of commissural axons and other bipolar neurons is consistent with the hypothesis that the "U-shaped" group of cholinergic neurons may use commissural axons and other cohort neurons for guidance during their dorsal migration.

Development of cholinergic terminals around rat spinal motor neurons and their potential relationship to developmental cell death.

Neuron death seems to be regulated mainly by postsynaptic target cells. In chicks, nicotinic antagonists such as alpha-bungarotoxin (alphaBT) can prevent normal cell death of somatic motor neurons (SMNs). For this effect, however, alphaBT could be acting at peripheral neuromuscular junctions and/or central cholinergic synapses. To investigate this issue, we first studied the development of cholinergic terminals in the rat spinal cord by using vesicular acetylcholine transporter immunocytochemistry. Labeled terminals were seen in the ventral horn as early as embryonic day 15 (E15), the beginning of the cell death period. Thus, central cholinergic synapses form at the correct time and place to be able to influence SMN death. We next added alphaBT to organotypic, spinal slice cultures made at E15. After 5 days in vitro, the number of SMNs in treated cultures was substantially greater than in control cultures, indicating that alphaBT can reduce SMN cell death in rats as it does in chicks. Moreover, peripheral target removal led to extensive loss of SMNs, and such a loss occurred even in the presence of alphaBT, indicating the necessity of peripheral target for the alphaBT effect. Finally, to determine whether central cholinergic terminals also may be involved in SMN death, we delayed the alphaBT treatment until after central cholinergic terminals had disappeared from the slice cultures. The increased number of surviving SMNs, even in the absence of central terminals, argued that alphaBT acts at peripheral, not central, cholinergic synapses to rescue SMNs from developmental cell death.

Golgi-impregnated amacrine cells and GABAergic retinal neurons: a comparison of dendritic, immunocytochemical and histochemical stratification in the inner plexiform layer of rat retina.

This paper concerns the banding pattern produced in the inner plexiform layer of rat retina by glutamic acid decarboxylase (GAD) immunocytochemistry. It presents a comparison of this pattern with the dendritic stratification of neurons that are reasonable candidates for GABAergic amacrine cells in Golgi preparations, and also with the banding patterns produced by other histochemical techniques. First, the spacing of five dense GAD-positive bands and four intervening less dense bands in central retina is quantitatively described. Second, examples of a particular, morphologically homogenous group of Golgi-impregnated amacrine cells are examined in the details of their structure, especially with regard to their dendritic stratification. Computer reconstructions of the dendritic trees of some of these narrow-field, multistratified amacrines are compared with the GAD-positive banding pattern. This group of amacrines is judged to represent many of the GABAergic neurons in rat retina, accounting for the form and distribution of GAD-positive synaptic terminals by their dendritic morphology and stratification. Third, a general schema for the laminar subdivision (stratification) of the inner plexiform layer in rat retina is derived from a comparison of the results of several histochemical procedures. Finally, similarities and differences in the distribution of GAD-positive amacrine cell dendrites are noted among mammals and the functional implications of their broad distribution are discussed. A conspicuous difference is cited between mammals and certain nonmammalian vertebrates in which GAD-positive dendrites are restricted to sublamina beta (ON-center cells) of the inner plexiform layer.

Differences in developmental cell death between somatic and autonomic motor neurons of rat spinal cord.

Considerable knowledge concerning developmental cell death has come from the study of somatic motor neurons (SMNs), but a related set of spinal neurons, the autonomic motor neurons (AMNs), have been studied less extensively in this respect. In the present study, we used three different approaches to determine the amount of AMN cell death during normal development in the rat. First, target dependency was studied in organotypic slice cultures, and it was found that AMNs survived for at least 12 days after removal of their postsynaptic targets. No factors were added to the serum-free medium to substitute for the ablated targets, indicating that AMNs were able to survive without target-derived trophic factors. Such target-independent survival is not characteristic of neurons that undergo typical developmental cell death. Second, AMNs were counted in double-stained choline acetyltransferase immunocytochemical and NADPH diaphorase histochemical preparations at ages (postnatal days 4-22) encompassing the period when AMN postsynaptic target cells undergo developmental death. Neuron numbers were essentially identical at all ages examined, indicating that no AMN cell death occurred postnatally. Finally, from embryonic day 13 to postnatal day 22, animals were analyzed by using terminal transferase-mediated nick-end labeling to identify dying cells. Many fewer labeled cells were observed among AMNs than among SMNs. Thus, all three approaches indicated that there is a significant SMN/AMN difference in developmental cell death. The phenotypic trait(s) that underlies this difference may also be important in the relative resistance of AMNs to pathological conditions that induce death of SMNs, e.g., those involved in amyotrophic lateral sclerosis and excitotoxicity.

The generation of neurons involved in an early reflex pathway of embryonic mouse spinal cord.

The generation of lateral motor neurons (LMNs), interneurons and dorsal root ganglion (DRG) neurons of the cervical mouse spinal cord has been investigated by [3H]thymidine autoradiographic techniques. This investigation has two main objectives: (a) to determine on which embryonic days these three neuronal populations are born, and (b) to investigate the possibility that the neurons comprising early reflex circuits might be formed by a retrograde temporal sequencing of generation. LMNs are the first neurons generated in the cervical spinal cord. They arise between E8.8 and E11.5, and approximately 90% of these cells are born within a 36-hour period between E9 and E10.5. The earliest time of origin for interneurons is on E9.5, and those cells which are generated between E9.5 and E10.5 cluster in two distinct regions of the adult spinal cord. One of these regions is the lateral portions of laminae IV through VI; this appears to be the location of many ipsilateral association neurons. DRG neurons begin to arise on E9.5 and their generation is completed by E14. There is a trend within the DRG population for large neurons to be born before small neurons. Those cells with diameters of 40 micron or greater reach their generation peak on E10.5, while those smaller than 40 micron arise in the greatest numbers on E12. The findings of other investigations have provided evidence for a retrograde sequence of synaptic closure in the formation of the early disynaptic forelimb reflex pathway. The temporal difference in synapse formation in the terminal fields of DRG and association neurons is discussed in terms of our observation that both of these populations appear to have similar generation times. We suggest that factors responsible for the delayed synaptic closure of DRG afferents include the greater distances and the degree of collateralization which these afferents must undergo in order to establish their terminal fields. Finally, we discuss the possibility that the temporal sequence of neuronal generation and factors involved with the growth of neurites combine to produce a retrograde sequence of synaptic closure in the early disynaptic forelimb reflex pathway of mouse spinal cord.

The GABA neurons and their axon terminals in rat corpus striatum as demonstrated by GAD immunocytochemistry.

Glutamic acid decarboxylase (GAD, EC 4.1.1.15), the enzyme which catalyzes the alpha-decarboxylation of L-glutamate to form the neurotransmitter gamma-aminobutyric acid (GABA), was localized immunocytochemically in rat neostriatum, pallidum and entopeduncular nucleus. A large amount of GAD-positive reaction product was observed in both the pallidum and entopeduncular nucleus in light microscopic preparations and was localized ultrastructurally to axon terminalis that surrounded dendrites and large somata. In the neostriatum the relative numbers of GAD-positive axons terminals per unit area were substantially less than in the pallidum. GAD-positive terminals predominantly formed symmetric synapses with somata, dendrites and spines, but a small number of them formed asymmetric synapses with either dendrites or spines. The presence of GAD within these terminals is consistent with results of other investigations which have indicated that the striatopallidal and striatoentopeduncular pathways as well as neostriatal local circuit neurons and/or collaterals from neostriatal projection neurons, use GABA as a neurotransmitter. GAD-positive reaction product was also localized within the somata and dendrites of neostriatal and pallidal neurons in colchicine-injected preparations. The GAD-positive somata in the pallidum were medium-sized neurons and since such cells project to the substantia nigra, our results are in agreement with those from other studies which demonstrate a GABAergic, pallidonigral pathway. In the neostriatum, GAD-positive somata were identified light microscopically as medium-sized neurons with either round or fusiform shapes. Electron microscopic examinations also showed GAD-positive reaction product within the perikaryal and dendritic cytoplasm of these neurons, as well as in dendritic spines. These findings are in accord with the results of studies which have indicated that medium-sized, spinous neurons of the neostriatum give rise to a GABAergic, striatonigral pathway. The significance of GAD localization within these neostriatal neurons is discussed in relation to recent findings which show that substance P is contained within this same class of striatonigral projection neuron.

Autonomic motor neuron migration and expression of nicotinamide adenine dinucleotide phosphate reduced diaphorase are dependent upon peripheral target.

Interactions of developing neurons with their postsynaptic targets play a significant role in neuronal differentiation. The goal of the present study was to determine if target contact affected the migration or differentiation of autonomic motor neurons (AMNs) in developing rat spinal cord. The peripheral targets of AMNs were excised microsurgically from histotypic spinal slices before the arrival of AMN axons. The migration of AMNs was assessed in DiI retrogradely labeled preparations, and the differentiation of these cells was evaluated by beta-nicotinamide adenine dinucleotide phosphate reduced diaphorase (NADPH-d) histochemistry. In target-deprived specimens, NADPH-d expression in AMNs was virtually eliminated. In addition, DiI-labeled AMNs were scattered throughout the intermediate spinal gray matter instead of being aggregated in the intermediolateral nucleus as in control slices. This observation indicated that migration of AMNs had occurred, but that it had been disorganized significantly by target removal on embryonic day 13 (E13). In sham, "incision-only" specimens from which peripheral target tissue was not removed, AMNs expressed NADPH-d and migrated normally, indicating that axotomy alone was not sufficient to disrupt AMN development. Previous studies have shown that target removal after the arrival of AMN axons at their postsynaptic targets on E14 has no affect on the organized migration of AMNs (Barber et al. [1993] J. Neurosci. 13:4898-4907). This observation together with the present results indicate that initial target contact is necessary for both the differentiation and directed migration of AMNs, and that this contact does not need to be sustained for these developmental events to progress normally.

Embryonic development of choline acetyltransferase in thoracic spinal motor neurons: somatic and autonomic neurons may be derived from a common cellular group.

This investigation focused on the relationship between neurotransmitter phenotype expression and rat motor neuron development, as studied with choline acetyltransferase (ChAT) immunocytochemical techniques. The development of two subclasses of motor neurons, somatic and autonomic efferents, was examined in the upper thoracic spinal cord. ChAT was first detected in a few neurons on embryonic day 12 1/2 (E12 1/2), and in numerous cells located in a single, ventrolaterally located column in the intermediate zone on E13. By E14, this group of ChAT-positive neurons was more intensely immunoreactive, and their axons could be traced to appropriate targets in developing somatic muscle and paravertebral sympathetic ganglia. During the E15-16 period, somatic and autonomic motor neurons separated into two distinct subgroups, with the latter cells being observed to translocate dorsally. By E17, these autonomic motor neurons reached their final positions in the midportion of the intermediate zone. The autonomic motor neurons were observed to extend transverse dendritic bundles across the spinal cord between E15-16, but evidence of the longitudinal bundles of sympathetic preganglionic dendrites was not observed until after birth. A recent study of cholinergic thoracic motor neurons found that both somatic and autonomic cells were generated synchronously during the E11-12 period (Barber et al., Soc Neurosci Abstr 15:588, 1989). In combination with the present results, these data indicate that no more than 1 1/2 days are necessary after motor neuron genesis before a few cells begin to express detectable levels of ChAT, and that no more than 2 days are required before large numbers express this marker of the cholinergic phenotype. Further comparisons of the present findings with those of previous investigations of the development of both somatic and autonomic motor neurons (Dennis et al., Dev Biol 81:266, 1981; Rubin, J Neurosci 5:685, 697, 1985) indicate that these cells contain ChAT at the time their axons are growing toward their respective peripheral targets 1 day before the time when physiological evidence of function is manifest. Furthermore, the present results suggest that both subclasses of motor neurons initially migrate together from the ventricular zone into a single motor column within the ventral intermediate zone, and that the autonomic neurons subsequently translocate dorsally. Thus, autonomic motor neurons appear to be an exception to the generalization that postmitotic neurons migrate directly from the germinal zone to their final positions within the central nervous system.

Generation patterns of immunocytochemically identified cholinergic neurons at autonomic levels of the rat spinal cord.

The time at which a neuron is "born" appears to have significant consequences for the cell's subsequent differentiation. As part of a continuing investigation of cholinergic neuronal development, we have combined ChAT immunocytochemistry and [3H]thymidine autoradiography to determine the generation patterns of somatic and autonomic motor neurons at upper thoracic (T1-3), upper lumbar (L1-3), and lumbosacral (L6-S1) levels of the rat spinal cord. Additionally, the generation patterns of two subsets of cholinergic interneurons (partition cells and central canal cluster cells) were compared with those of somatic and autonomic motor neurons. Embryonic day 11 (E11) was the first day of cholinergic neuronal generation at each of the three spinal levels studied, and it also was the peak generation day for somatic and autonomic neurons in the upper thoracic spinal cord. The peak generation of homologous neurons at upper lumbar and lumbosacral spinal levels occurred at E12 and E13, respectively. Somatic and autonomic motor neurons were generated synchronously, and their production at each rostrocaudal level was virtually completed within a 2-day period. Cholinergic interneurons were generated 1 or 2 days later than motor neurons at the same rostrocaudal level. In summary, the birthdays of all spinal cholinergic neurons studied followed the general rostrocaudal spatiotemporal gradient of spinal neurogenesis. In addition, the generation of cholinergic interneurons also followed the general ventrodorsal gradient. In contrast, however, autonomic motor neurons disobeyed the rule of a ventral-to-dorsal progression of spinal neuronal generation, thus adding another example in which autonomic motor neurons display unusual developmental patterns.

Quantitative analyses of neuronal development in the lateral motor column of mouse spinal cord. I. Genetically associated variations in somal growth patterns.

The relative somal and nuclear sizes of neurons in the lateral motor column (LMC) of adult and embryonic mouse brachial spinal cord were determined by light microscopic morphometry. Three genetically varying mouse strains, previously shown to differ in the development of a forelimb reflex pathway, were studied. In adults, the size distribution of somata, nuclei, and nucleoli were bimodal for each strain, indicating that there are two distinct size classes of LMC neurons. The size division between large and small LMC neurons differed among strains with more large LMC neurons occurring in strain CBA/CaJ than in either LP/J or C57BL/6J. In embryos, the growth of LMC cells was studied by determining the average area of nuclear profiles for specimens ranging in age from embryonic day 11 (E11) to 16. The average nuclear profile area increased significantly during this period in all three strains, and differences were found in the initial size and apparent rate of growth among strains. Early in development (E11-12), strain differences in apparent cell size were: C57BL/6J greater than CBA/CaJ greater than LP/J, and this strain order corresponds to observed strain differences in the onset of reflexogenesis and synaptogenesis (Vaughn et al., '75). Later in development (E16), strain differences in apparent cell size were: CBA/CaJ greater than LP/J greater than or equal to C57BL/6J, and this relationship corresponds to a more rapid increase of presumptive afferent synapses in CBA/CaJ than in the other two strains between E15 and E16. Possible causal relationships among neuronal size, growth, and synaptogenesis are suggested by these strain differences.

Quantitative analyses of neuronal development in the lateral motor column of mouse spinal cord. II. Development of motor neuronal organelles.

The development of organelles within presumptive alpha motor neuronal somata was studied by electron microscopic morphometric analysis. Cells with large nuclei were selected for sampling from the lateral motor column of the brachial spinal cord of mouse embryos ranging in age from embryonic day 11 (E11) through E16. The first objective was to compare the cytodifferentiation of alpha motor neuronal somata among three genetically different strains of mice that differ in the development of forelimb reflex behavior and associated pathway synaptogenesis (Vaughn et al., '75). On the basis of multiple linear regression analyses, no significant differences were found among strains for either the initial levels or rates of cytodifferentiation. As a result, the data were combined for all three strains to analyze organelle changes during early neuronal development. The average areas of perikaryal cytoplasm and nuclei increased significantly. In addition, the relative areas of nucleoli, mitochondria, Golgi complexes, and rough endoplasmic reticulum increased, while the relative areas of heterochromatin and "free" ribosomes decreased. There were significant increases in the number of mitochondria and Golgi complexes per unit area of perikaryal cytoplasm. The average size of mitochondria appeared to increase during development, but was significantly smaller in adult alpha motor neurons than in embryonic specimens. In contrast, the average size of individual Golgi complexes was relatively constant throughout embryonic development, as well as in the adult. In general, the cytodifferentiation of alpha motor neurons appeared to progress in a relatively constant, linear fashion between E11 and E16.

Quantitative analyses of neuronal development in the lateral motor column of mouse spinal cord. III. Generation and settling patterns of large and small neurons.

The generation and settling patterns of large and small lateral motor column (LMC) neurons were compared in the spinal cords of three inbred strains of mice by means of tritiated thymidine autoradiography. No significant strain differences were observed for the number of large LMC cells (presumptive alpha motor neurons) that were heavily labeled on each injection day, although there were significant strain variations for this measure with regard to small LMC neurons (presumed gamma motor neurons and interneurons). The generation of both large and small LMC neurons began at the same time, but peak production of large cells preceded that of the small neurons. There were no strain differences observed for this relationship between the large and small cells. These findings indicate that the LMC, from the time of its initial formation, contains cells destined to become large and small neurons. The positions of large and small neurons within the adult LMC relative to their times of origin (settling patterns) were analyzed statistically. A significant ventrodorsal sequence for early-to-late generated cells was observed for both large and small LMC neurons. No significant strain differences were found in the analysis of settling patterns. A ventrodorsal settling pattern also has been described for amphibia (Prestige, '73) and, in conjunction with the proximodistal sequence of limb development described by other investigators, the ventrodorsal sequence could play a key role in the development of motor neuronal somatotopic organization.

Embryonic development of rat sympathetic preganglionic neurons: possible migratory substrates.

Spinal somatic and autonomic (sympathetic preganglionic) motor neurons are generated synchronously and, subsequently, migrate from the ventricular zone together to form a common primitive motor column. However, these two subsets of motor neurons ultimately express several phenotypic differences, including somal size, peripheral targets, and spinal cord locations. While somatic motor neurons remain ventrally, autonomic motor neurons (AMNs) move both dorsally and medially between embryonic days 14 and 18, when they approximate their final locations in spinal cord. The goal of the present investigation was to determine the potential guidance substrates available to AMNs during these movements. The dorsal translocation was studied in developing upper thoracic spinal cord, because, at this level, the majority of AMNs are located dorsolaterally. Sections were double-labeled by ChAT (choline acetyltransferase) and SNAP/TAG-1 (stage-specific neurite associated protein/transiently expressed axonal surface glycoprotein) immunocytochemistry to visualize motor neurons and the axons of early forming circumferential interneurons, respectively. Results showed that during the developmental stage when AMNs translocated dorsally, SNAP/TAG-1 immunoreactive lateral circumferential axons were physically located along the borders of the AMN region, as well as among its constituent cells. These findings indicate that lateral circumferential axons, as well as the SNAP/TAG-1 molecules contained upon their surfaces, are in the correct spatial and temporal position to serve as guidance substrates for AMNs. The medial translocation was studied in developing lower thoracic-upper lumbar spinal cord, because, at this level, more than half of the AMNs are medially located. Sections were double-labeled by ChAT and vimentin immunocytochemistry to visualize motor neurons and radial glial fibers, respectively. Observations on consecutive developmental days of the medial translocation revealed that AMNs were aligned with parallel arrays of radial glial fibers. Thus, the glial processes could serve as guides for the AMN medial movement. Future experimental analyses will examine whether circumferential axons and radial glial fibers are in fact functioning as migratory guides during AMN development, and, if so, whether specific surface molecules on these guides trigger the subsequent differentiation of AMNs.

Immunocytochemical localization of choline acetyltransferase within the rat neostriatum: a correlated light and electron microscopic study of cholinergic neurons and synapses.

Monoclonal antibodies to choline acetyltransferase (ChAT) were used in an immunocytochemical study to characterize putative cholinergic neurons and synaptic junctions in rat caudate-putamen. Light microscopy (LM) revealed that ChAT-positive neurons are distributed throughout the striatum. These cells have large oval or multipolar somata, and exhibit three to four primary dendrites that branch and extend long distances. Quantitative analysis of counterstained preparations indicated that ChAT-positive neurons constitute 1.7% of the total neuronal population. Electron microscopy (EM) of immunoreactive neurons initially studied by LM revealed somata characterized by deeply invaginated nuclei and by abundant amounts of organelle-rich cytoplasm. Surfaces of ChAT-positive neurons are frequently smooth, but occasional somatic protrusions and dendritic spines occur. Although infrequently observed, axons of ChAT-positive neurons branch, receive synapses, and become myelinated. Unlabeled boutons make both symmetrical and asymmetrical synapses with ChAT-positive somata and proximal dendrites, but are more numerous on distal dendrites. In addition, some unlabeled terminals form asymmetrical synapses with ChAT-positive somata and dendrites that are distinguished by prominent subsynaptic dense bodies. Light microscopy demonstrated a dense distribution of ChAT-positive fibers and punctate structures in the striatum, and these structures appear to correlate, respectively, with labeled preterminal axons and presynaptic boutons identified by EM. ChAT-positive boutons contain pleomorphic vesicles, and make symmetrical synapses primarily with unlabeled dendritic shafts. Furthermore, they establish synaptic contacts with somata, dendrites and axon initial segments of unlabeled neurons that ultrastructurally resemble medium spiny neurons. These observations, together with the results of other investigations, suggest that medium spiny GABAergic projection neurons receive a cholinergic innervation that is probably derived from ChAT-positive striatal cells. The results of this study also indicate that cholinergic neurons within caudate-putamen belong to a single population of cells that have large somata and extensive sparsely spined dendrites. Such neurons, in combination with dense concentrations of ChAT-positive fibers and terminals, are the likely basis for the large amounts of ChAT and acetylcholine detected biochemically within the neostriatum.

Generation patterns of four groups of cholinergic neurons in rat cervical spinal cord: a combined tritiated thymidine autoradiographic and choline acetyltransferase immunocytochemical study.

This report examines the generation of cholinergic neurons in the spinal cord in order to determine whether the transmitter phenotype of neurons is associated with specific patterns of neurogenesis. Previous immunocytochemical studies identified four groups of choline acetyltransferase (ChAT)-positive neurons in the cervical enlargement of the rat spinal cord. These cell groups vary in both somatic size and location along the previously described ventrodorsal neurogenic gradient of the spinal cord. Thus, large (and small) motoneurons are located in the ventral horn, medium-sized partition cells are found in the intermediate gray matter, small central canal cluster cells are situated within lamina X, and small dorsal horn neurons are scattered predominantly through laminae III-V. The relationships among the birthdays of these four subsets of cholinergic neurons have been examined by combining 3H-thymidine autoradiography and ChAT immunocytochemistry. Embryonic day 11 was the earliest time that neurons were generated within the cervical enlargement. Large and small ChAT-positive motoneurons were produced on E11 and 12, with 70% of both groups being born on E11. ChAT-positive partition cells were produced between E11 and 13, with their peak generation occurring on E12. Approximately 70% of the cholinergic central canal cluster and dorsal horn cells were born on E13, and the remainder of each of these groups was generated on E14. Other investigators have shown that all neurons within the rat cervical spinal cord are produced in a ventrodorsal sequence between E11 and E16. In contrast, ChAT-positive neurons are born only from E11 to E14 and are among the earliest cells generated in the ventral, intermediate, and dorsal subdivisions of the spinal cord. However, all cholinergic neurons are not generated simultaneously; rather their birthdays are correlated with their positions along the ventrodorsal gradient of neurogenesis. The fact that large motoneurons and medium-sized partition cells are born before small central canal cluster and dorsal horn cells would appear to support the generalization that large neurons are generated before small ones. However, the location of spinal cholinergic neurons within the neurogenic gradient seems to be more importantly associated with the time of cell generation than somal size. For example, when large and small motoneurons located at the same dorsoventral spinal level are compared, both sizes of cells are generated at the same time and in similar proportions.(ABSTRACT TRUNCATED AT 400 WORDS)

Localization of Drosophila neurons that contain choline acetyltransferase messenger RNA: an in situ hybridization study.

In situ hybridization with radiolabeled complementary RNA (cRNA) probes was used to determine the location of the messenger RNA (mRNA) encoding choline acetyltransferase (ChAT) in Drosophila nervous system. Areas in the cell-rich cortical regions of the cerebrum and optic lobes hybridized with substantial concentrations of the probe. This contrasted with the cell-sparse neuropil areas where no significant concentrations of probe were observed. Although most of the cortical regions were substantially labeled, there were regions within all of the areas where labeling was sparse or nonexistent. For example in the lamina, even though the monopolar cell layer appeared to be heavily labeled, there were some neuronal profiles that were not associated with the probe. Moreover, the epithelial glia that form an arch of cell profiles subjacent to the monopolar cells were not labeled, nor were amacrine neurons in the apex of the lamina near the external optic chiasma. The highest concentration of probe (approximately 140 grains/400 microns2) was observed in the laminar monopolar cell region and the cerebral cortical rind. The next most heavily labeled region (approximately 90 grains/400 microns2) occurred over cortical cells of the medulla-lobula. In the peripheral nervous system, label over the antennal sensory neurons amounted to about 75 grains/400 microns2, and the retinular cell layer of the compound eye exhibited about 60 grains/400 microns2. The control probe did not hybridize in significant quantities in either cellular or noncellular regions. This study presents evidence that large numbers of Drosophila cortical and primary sensory neurons contain the messenger RNA necessary for the production of ChAT, the acetylcholine-synthesizing enzyme. Further, our findings provide baseline information for use in ontogenetic studies of cholinergic neurons in Drosophila, and they also provide normative data for studying the effects of mutant alleles at the Cha or Ace loci upon the transcription of ChAT messenger RNA.