The turnover of a CNS‐specific cell adhesion glycoprotein, ependymin, has earlier been found to increase during periods of neuronal plasticity. Here, ependymin mRNA expression was analyzed by semiquantitative in situ hybridization in goldfish. Learning of an active avoidance response resulted in a significant increase in ependymin mRNA expression 20 min to 4 h after acquisition of the task. In contrast, yoked control animals that were exposed to the same numbers of conditioned and unconditioned stimuli in a random, unpaired manner exhibited a strong down‐regulation of ependymin mRNA. Hybridization signals were also increased by injection of anti‐ependymin antiserum into brain ventricles. Ependymin mRNA was exclusively localized to reticular‐shaped fibroblasts of the inner endomeningeal cell layer. Immunoelectron microscopic investigation, however, revealed ependymin also in distinct neuronal and glial cell populations in which no ependymin mRNA had been detected. Uptake of meningeal protein factors into glial and neuronal cells may therefore be of functional importance for plastic adaptations of the CNS.

The Escherichia coli arcA gene product regulates chromosomal gene expression in response to deprivation of oxygen (Arc function; Arc stands for aerobic respiration control) and is required for expression of the F plasmid DNA transfer (tra) genes (Sfr function; Sfr stands for sex factor regulation). Using appropriate lacZ fusions, we have examined the relationship between these two genetic regulatory functions. Arc function in vivo was measured by anaerobic repression of a chromosomal sdh-lacZ operon fusion (sdh stands for succinate dehydrogenase). Sfr function was measured by activation of a plasmid traY-lacZ gene fusion. An eight-codon insertion near the 5' terminus of arcA, designated arcA1, abolished Arc function, as previously reported by S. Iuchi and E.C.C. Lin (Proc. Natl. Acad. Sci. USA 85:1888-1892, 1988), but left Sfr function largely (greater than or equal to 60%) intact. Similarly, the arcB1 mutation, which depressed sdh expression and is thought to act by abolishing the signal input that elicits ArcA function, had little effect (less than or equal to 20%) on the Sfr function of the arcA+ gene product. Conversely, a valine-to-methionine mutation at codon 203 (the sfrA5 allele) essentially abolished Sfr activity without detectably altering Arc activity. These data indicate that Sfr and Arc functions are separately expressed and regulated properties of the same protein.

A rapid increase in ependymin mRNA expression demonstrated by semiquantitative in situ hybridization after avoidance conditioning on goldfish suggested a molecular demand for newly synthesized ependymin translation product. To inhibit de novo synthesis of ependymin molecules without interference with preexisting ones, 18 mer anti‐ependymin mRNA‐phosphorothioate oligodeoxynucleotides (S‐ODNs) were injected into the perimeningeal brain fluid before active avoidance training. S‐ODN‐injected animals learned the avoidance response; however, they were amnesic in the test. When injected into overtrained animals, S‐ODNs did not interfere with retrieval or performance of the avoidance response. Fish treated with randomized S‐ODN sequences served as further controls. Incorporation of S‐ODNs was analyzed by injection of fluorescein isothiocyanate (FITC)‐conjugated oligodeoxynucleotide probes. Microscopic observation revealed strong FITC‐S‐ODN fluorescence in reticular‐shaped fibroblasts, the only known site of ependymin synthesis. Results demonstrate that selective inhibition of ependymin gene expression in vivo can specifically prevent memory formation. We conclude that in particular the newly synthesized ependymin molecules are involved in memory consolidation, possibly because they have not yet undergone irreversible molecular changes, which have been reported of this glycoprotein in a low‐calcium microenvironment.

No abstract

No abstract