Vicky R. Holets

Variable Morphological Differentiation of a Raphé-Derived Neuronal Cell Line Following Transplantation into the Adult Rat CNS

A clonal, neuronally-differentiating cell line, RN33B, was previously developed by retroviral infection of neural tissue derived from embryonic Sprague-Dawley raphé nuclei with a retrovirus encoding the temperature-sensitive allele of SV40 large T-antigen. In the present study, RN33B cells were transplanted into two target areas of the raphé nuclei, the spinal cord and hippocampal formation, of adult allogeneic hosts. Prior to transplantation, RN33B cells were infected in vitro with a retroviral vector carrying the Escherichia coli lacZ reporter gene and were visualized in vivo using a beta-galactosidase immunohistochemical technique. RN33B cells were seen throughout the spinal cord and hippocampal formation of the adult hosts at 15 days post-transplantation. T-antigen-immunoreactive nuclei were detected where RN33B cells were observed, but in much greater numbers than beta-galactosidase-immunoreactive cells. Bipolar RN33B cells were found in the spinal cord grey matter. RN33B cells with multipolar morphologies were visualized in the hippocampal and subicular pyramidal cell layers, and also in the dentate gyrus granule cell and polymorph layers, while bipolar RN33B cells were seen in the remainder of the hippocampal formation. The results suggest that immortalized neural cell lines of CNS origin can differentiate in the adult CNS with their ultimate morphology being determined by local tissue signals. We speculate that endogenous neutrophins may significantly influence RN33B cell differentiation in vivo.

Selective hippocampal lesions differentially affect the phenotypic fate of transplanted neuronal precursor cells.

RN33B cells, a CNS-derived neuronal precursor cell line, transplanted into normal adult rat hippocampus can survive and morphologically differentiate with their ultimate morphology dependent on the integration site. This study examined the differentiation and structural integration of RN33B cells transplanted into the lesioned adult hippocampus. Pyramidal neurons of the CA1-3 regions or granular neurons in the dentate gyrus were preferentially destroyed by unilateral intraventricular kainic-acid or intradentate colchicine injections, respectively. One week after the lesion, a suspension of undifferentiated beta-galactosidase (beta-gal)-labeled RN33B cells was stereotaxically transplanted into the lesioned or the contralateral hippocampus. After 5-7 weeks, sections of the recipient brains were analyzed by toluidine blue staining and immunohistochemistry for beta-gal, GFAP, and OX-42. A reactive gliosis was observed on the lesioned side which persisted up to 7 weeks postlesion (the latest time point examined). RN33b cells survived in the lesioned hippocampus and assumed variable morphologies depending on the hippocampal layer into which they integrated. Only RN33B cells located in intact or partially damaged cell layers or in the unlesioned contralateral hippocampus differentiated with morphologies similar to those of endogenous neurons characteristic of those layers. Cells located in layers completely depleted of endogenous neurons assumed bipolar morphologies or sent out multiple processes with no structural polarity, unlike the neuronal morphologies characteristically seen in intact hippocampal cell layers. These data suggest that the presence of some endogenous neurons and partially conserved cytoarchitectural organization are essential for immortalized neuroepithelial precursor cells to differentiate into region-specific neuronal cell types.

Induction of mature neuronal properties in immortalized neuronal precursor cells following grafting into the neonatal CNS

SummaryRN33B, a conditionally-immortalized neuronal cell line, survives and differentiates following grafting into the neocortex and hippocampus of adult and neonatal rat hosts. We have previously shown that these cells assume shapes characteristic of endogenous neurons at the integration site and persist up to 24 weeks post-grafting. In the present study we use electron microscopy and immunohistochemistry to characterize such cells. Differentiated RN33B cells were identical in size to endogenous neurons and their sizes depended on the specific location of integration. RN33B cells in the granule cell layer of the dentate gyrus and CA3 and CA1 pyramidal layers were 9.0, 15.3, and 12.6 μm in diameter, respectively. Grafted RN33B cells received synapses from fibres of host origin. Differentiated cells expressed neuronal markers, but not glial markers. Some differentiated cells expressed glutamate bothin vitro andin vivo whereas undifferentiated cells did not. Grafted RN33B cells that differentiated with morphologies similar to CA3 pyramidal neurons and pyramidal cortical neurons expressed Py antigen, a neuronal marker that is differentially expressed in endogenous large pyramidal neurons of the cerebral cortex and large pyramids of hippocampal field CA3. This Py immunoreactivity was region-specific and corresponded to the endogenous pattern of Py immunostaining. Collectively, these data indicate that RN33B cells are capable of region-specific differentiation and have the potential to integrate functionally into the host CNS.

Photochemically induced cystic lesion in the rat spinal cord: I. Behavioral and morphological analysis

The present study describes the production of a spinal cord lesion which is initiated by vascular occlusion resulting from the interaction between the photosensitizing dye erythrosin B and an argon laser beam. The lesion has characteristics similar to those of the central cavity thought to lead to the production of post-traumatic syringomyelia (PTS) in humans. The present study examines the behavioral and morphological characteristics of this injury over a 28-day period. Histological analysis revealed a cavity extending from the dorsal horns to lamina VIII, with some lateral and ventral pathways being spared. The cavity volume reached a maximum 7 days after lesion induction. Behavioral changes were assessed using six different tests of motor and reflex function (motor function, climbing, waterbath, inclined plane, withdrawal to pain, and withdrawal to extension). Lesioned animals exhibited flaccid paralysis for 3-5 days, which resolved afterward. The photochemically induced cavity should provide a reproducible model for examining the effects of cystic spinal cord injury on locomotor and reflex function.

Multiple putative neuromessenger inputs to the phrenic nucleus in rat

Immunohistochemical reactions for 12 putative neuromessengers combined with retrograde labeling of phrenic motoneurons identified seven neuromessengers (5-hydroxytryptamine, substance P, thyrotropin-releasing hormone, methionine enkephalin, cholecystokinin, galanin, neuropeptide Y) located within terminal varicosities in the phrenic nucleus. The degree of terminal labeling in the phrenic nucleus varied depending on the peptide. Substance P, thyrotropin-releasing hormone and methionine enkephalin were each tested for colocalization with 5-hydroxytryptamine within terminal varicosities in the phrenic nucleus, and the coincidence of double-labeling varied for each peptide. These results indicate that phrenic motoneurons are subject to modulation by many peptide neuromessengers that may alter their responsiveness to primary excitatory and inhibitory inputs.

A computerized grid walking system for evaluating the accuracy of locomotion in rats

A microcomputer-based system employing photoelectric devices to record rat movements and footfalls in the grid walking test paradigm was developed and evaluated. Behavioral data obtained from the system were: distance traversed on the grid, time taken to traverse the distance, number of footfalls, times of footfalls, positions of footfalls, durations of footfalls, and whether the footfall was due to a hind or a fore limb. Validation of the system was performed by comparing the data obtained from the videotape analysis with that obtained from the computerized system. Correlation coefficients between the data obtained from the two methods were found to be 0.92 for one observer, 0.84 for a second observer, and 0.88 with the mean of the two observers. An experimental study in which a group of rats was administered dorsal hemisection lesions of the spinal cord was also conducted. Animals in the lesion group took the same amount of time to cross the runway as the control animals, but made more footfalls per crossing and had longer durations per footfall. The studies validate the capacity of the computerized system to efficiently detect fine locomotory deficits, suggesting that it is a viable tool for the evaluation of neurological dysfunctions in experimental rats.