By the 1980s, it had been demonstrated that the known selenoproteins, animal glutathione peroxidase and bacterial selenoprotein A of glycine reductase, contained selenium as selenocysteine in their primary structures [1,2]. Whether this unusual amino acid was incorporated at the time of translation or was made as a posttranslational modification of cysteine or serine was not known. The post-translational modification theory was supported by the observation that serine provided the carbon skeleton of the selenocysteine in glutathione peroxidase [3]. Other workers proposed that co-translational incorporation occurred.
While these seemingly incompatible views were being debated, the cDNA of mouse glutathione peroxidase was cloned and sequenced [4]. It contained a TGA codon at the position of the selenocysteine in the polypeptide chain. Thus, the preponderance of evidence suggested that incorporation of selenocysteine was cotranslational. However, this conclusion required that TGA be able to code for selenocysteine. Because TGA was already known to code for termination of translation, it was necessary to postulate that this codon have the alternative function of coding for selenocysteine insertion. A means to designate which of the alternative functions each TGA codon exerts must therefore exist.
Characterization of the molecular mechanism by which TGA (UGA in mRNA) directs selenocysteine incorporation into the primary structure of bacterial selenoproteins was achieved by the research group of August Bo ยจck. The Bo ยจck group published the Landmark Paper reprinted here [5]. This seminal paper demonstrated that four genes were necessary to effect selenocysteine incorporation into E. coli selenoproteins. It thereby set the stage for the detailed characterization of the whole process of selenoprotein synthesis which provided the explanation of how UGA serves as a selenocysteine codon.
The Landmark Paper employed the strategy of identifying E. coli mutants that could not metabolize formic acid. This defect implied that those mutants could not synthesize functional forms of either of the two E. coli formate dehydrogenases. Both of these enzymes have selenoprotein subunits. The mutant genes were identified by complementation experiments and named selA, selB, selC, and selD. Subsequent work by Bo ยจck's group characterized the genes and their functions, providing a detailed picture of selenoprotein synthesis in prokaryotes (see Figure 1).
The importance of this work for the selenium field cannot be overemphasized. This series of studies reconciled the early observations on selenoprotein synthesis and