Structure and expression of the glycine cleavage system in rat central nervous system.

The glycine cleavage system (GCS) is a mitochondrial multienzyme system consisting of four individual proteins, three specific components (P-, T-, and H-proteins) and one house-keeping enzyme, dihydrolipoamide dehydrogenase. Inherited deficiency of the GCS causes nonketotic hyperglycinemia (NKH), an inborn error of glycine metabolism. NKH is characterized by massive accumulation of glycine in serum and cerebrospinal fluids and severe neuronal dysfunction in neonates. To elucidate the neuropathogenesis of NKH, we cloned cDNAs encoding three specific components of the GCS and studied the gene expression in rat central nervous system. P-, T-, and H-protein cDNAs encoded 1024, 403, and 170 amino acids, respectively. In situ hybridization analysis revealed that P-protein mRNA was expressed mainly in glial-like cells, including Bergmann glias in the cerebellum, while T- and H-protein mRNAs were detected in both glial-like cells and neurons. T- and H-protein mRNAs, but not P-protein mRNA, were expressed in the spinal cord. Primary astrocyte cultures established from cerebral cortex had higher GCS activities than hepatocytes whereas those from spinal cord expressed only H-protein mRNA and had no enzymatic activity. An important role of glycine as inhibitory neurotransmitter has been established in the brainstem and spinal cord and another role of glycine as an excitation modulator of N-methyl-D-aspartate receptor is suggested in the hippocampus, cerebral cortex, olfactory bulbus, and cerebellum. Our results suggest that the GCS plays a major role in the forebrain and cerebellum rather than in the spinal cord, and that N-methyl-D-aspartate receptor may participate in neuropathogenesis of NKH.

c-fos mRNA, Fos, and Fos-related antigen induction by hypertonic saline and stress.

The induction of c-fos mRNA was assessed using Northern blots and in situ hybridization in adult rats administered hypertonic saline (HS) and isotonic saline (IS). HS induced c-fos mRNA in magnocellular paraventricular nucleus (PVNm), parvocellular paraventricular nucleus (PVNp), supraoptic nucleus (SON), and lamina terminalis (LMT). This occurred within 5 min, peaked at 30-60 min, and disappeared by 180 min. Fos protein, detected using a specific monoclonal antibody, was maximal at 1-2 hr and disappeared 4-8 hr after HS administration. This confirms observations showing that the c-fos gene response is transient even in the presence of a continuing stimulus. In contrast, Fos-like immunoreactivity (FLI), detected using two polyclonal antisera, was observed in PVNm, PVNp, SON, and LMT for 1-24 hr during continuous osmotic stimulation. Moreover, FLI was observable in these structures for 7 d in rats administered HS and allowed to drink water ad libitum beginning 24 hr later. At times greater than 8 hr, FLI presumably represents Fos-related antigens (FRA), proteins immunologically and functionally related to Fos, whose expression is much more prolonged than authentic Fos following the osmotic stimulus. In addition to induction of c-fos expression in regions specifically involved in osmotic regulation, HS injections also induced c-fos in many other forebrain regions. In order to assess the induction of c-fos mRNA due to the "stress" of the injections, rats injected with isotonic saline were compared to uninjected controls. Isotonic saline injections induced c-fos mRNA in the PVNp, anterior hypothalamus, suprachiasmatic nucleus, cingulate gyrus, neocortex, ventral lateral septal nucleus, piriform cortex, hippocampal pyramidal and dentate granule neurons, paraventricular and intralaminar thalamic nuclei, bed nuclei of stria terminalis, cortical and medial amygdaloid nuclei, and other structures. In accord with other work, we interpret this pattern of c-fos expression to result from the stress of handling and injections. Since Fos and FRA probably bind to the promoters of target genes and regulate their expression, they likely mediate biochemical changes in the cells activated by the osmotic and stressful stimuli. Whereas the Fos signal is transient, FRA may act on target genes for the duration of the stimulus or longer.

Marked neurotrophic effects of diffusible substances released from non-target cerebellar cells on thalamic neurons in culture.

A primary culture of thalamic cells from 6-day-old postnatal rats was co-cultured for 6 days with neocortical or cerebellar cells (neurons and astrocytes) from the same litter using a Transwell mesh system. The survival of thalamic neurons grown on the lower well, which were affected by substances released from cells grown on the upper wells, was remarkably promoted by both neocortical co-cultures (target for thalamic projection neurons) and cerebellar co-cultures (non-target). When the cells were seeded on mesh at lower density, the neurotrophic effects of neocortical co-cultures on thalamic neurons (204% of control) were significantly greater than those of cerebellar co-cultures (138%). When the cells were seeded on mesh at higher density, the effects of cerebellar co-cultures increased dramatically (517% of control), while the neurotrophic effects of neocortical co-cultures did not change. Morphologically, the survival of multipolar-shaped thalamic neurons was remarkably improved, as compared to the survival of monopolar, bipolar, and tripolar-shaped thalamic neurons. Basic fibroblast growth factor slightly promoted thalamic neuronal survival (136%), whereas nerve growth factor had no effect. These results suggest that neocortical and cerebellar cells release diffusible factor(s) that promote the survival of specific subpopulation of thalamic neurons, and that at least one of the non-target cerebellar cell-derived factor(s) might be more potent than those released from target neocortical cells.