Autoantibodies against a serine tRNA-protein complex implicated in cotranslational selenocysteine insertion

Selenocysteine insertion or termination: factors affecting UGA codon fate and complementary anticodon:codon mutations

Translation of UGA as selenocysteine instead of termination occurs in numerous proteins, and the process of recording UGA requires specific signals in the corresponding mRNAs. In eukaryotes, stem-loops in the 3' untranslated region of the mRNAs confer this function. Despite the presence of these signals, selenocysteine incorporation is inefficient. To investigate the reason for this, we examined the effects of the amount of deiodinase cDNA on UGA readthrough in transfected cells, quantitating the full-length and UGA terminated products by Western blotting. The gene for the selenocysteine-specific tRNA was also cotransfected to determine if it was limiting. We find that the concentrations of both the selenoprotein DNA and the tRNA affect the ratio of selenocysteine incorporation to termination. Selenium depletion was also found to decrease readthrough. The fact that the truncated peptide is synthesized intracellularly demonstrates unequivocally that UGA can serve as both a stop and a selenocysteine codon in a single mRNA. Mutation of UGA to UAA (stop) or UUA (leucine) in the deiodinase mRNA abolishes deiodinase activity; but activity is partially restored when selenocysteine tRNAs containing complementary mutations are contransfected. Thus, UGA is not essential for selenocysteine incorporation in mammalian cells, provided that codon:anticodon complementarity is maintained.

A factor protecting mammalian [75Se]SeCys-tRNA is different from EF-1 alpha

In Escherichia coli, an elongation factor (EF-Tu-like) specific to SeCys-tRNA, SELB, has been identified; however, a mammalian counterpart of SELB has not been reported to date. We searched for and found this factor in bovine liver extracts using the assay of [75Se]SeCys-tRNA protecting activity against alkaline hydrolysis (SePF activity). We found SePF activity in the protein extracts of the precipitate (microsomal fraction) collected at 150,000 x g from bovine liver. The proteins were separated by Sephacryl S-300 chromatography, and the SePF and EF-1 alpha activities were found in the same fraction, indicating that SePF and EF-1 alpha have the same molecular mass (approximately 50 kDa). We then chromatographed this active fraction using CM-Sephadex C-25 columns. The SePF activity was eluted after the peak of EF-1 alpha activity. This result indicated that this SePF activity was not dependent on EF-1 alpha. In addition to performing these two chromatographies, we investigated pure EF-1 alpha from Bombyx mori but could not detect any SePF activity in B. mori EF-1 alpha. Thus we showed that the SePF activity in bovine liver differs from that of EF-1 alpha in eukaryotes. Therefore the factor protecting [75Se]SeCys-tRNA in bovine liver is not EF-1 alpha and must be a SELB-like factor.