Selenium is an essential micronutrient in the diet of humans and other mammals. Many health benefits have been attributed to selenium that include preventing various forms of cancer (e.g., colon, prostate, lung and liver cancers), heart disease and other cardiovascular and muscle disorders. Numerous human clinical trials have been undertaken in recent years to assess the role of this element in cancer prevention, delaying the progression of AIDS, etc., at a cost of hundreds of millions of dollars, but little was known about the mechanism of how selenium acts at the metabolic level in mammals to incorporate selenium into protein. We proposed several years ago that the health benefits of selenium are due largely to the presence of this element in selenoproteins as the selenium-containing amino acid, Sec. In the ensuing years, we established the biosynthetic pathway of Sec in eukaryotes and Archaea and focused on the two Sec tRNA isoforms that we demonstrated were responsible for the synthesis of the two subclasses of selenoproteins, housekeeping and stress-related selenoproteins; and pursued studying the methylase, designated Um34 methylase, that synthesizes the methyl group at the 2'-O-postion on the ribosyl moiety at nucleotide 34 of Sec tRNA. We provided strong evidence that addition of Um34 to the isoform, 5-methylcarobxymethyl-uridine (mcmU), to form 5-methylcarboxymethyl-2'-O-methyluridine (mcmUm) requires that mcmU is aminoacylated with Sec, i.e., that the substrate for the methylase (designated Um34 methylase) which carries out this reaction is selenocysteyl-tRNA. In addition, our program focused on developing mouse models to assess the role of all selenium-containing proteins within the two subclasses, housekeeping and stress-related selenoproteins, and on individual selenoproteins in preventing and promoting cancer and in mammalian development. In the past year, we completed and published the following collaborative studies: 1) inhibition of S-adenosylhomocysteine (SAH) in endothelial cells resulted in reduced Um34 synthesis on the Sec tRNA isoform, mcmU, and in turn reduced expression of glutathione peroxidase 1, which supported our earlier findings that the mcmUm isoform supports stress-related selenoprotein synthesis; 2) the targeted removal of SECp43, a protein of unknown function, in mouse liver, which we initially determined was embryonic lethal, had no effect on selenoprotein expression in liver or neural tissue, resulted in no liver damage, but impaired motor function, suggesting that SECp43 is not essential for selenoprotein synthesis in either liver or neural tissue; 3) the knockdown of selenoprotein 15 (Sep15) or thioredoxin reductase 1 (TR1) in a mouse colon cancer cell line, CT26, reversed several of the malignant properties, while surprisingly, the double knockdown of both selenoproteins reversed the anti-cancer effects of the loss of either selenoprotein individually; 4) targeted knockdown of Sep15 in Chang liver cells impeded cell proliferation at the G1 phase by upregulating p21 and p27, relocation of focal adhesions to the periphery of the cell basement, inhibition of migratory and invasive ability and caused cell blebbing; these changes were reversed by reintroduction of Sep15 expression suggesting that this selenoprotein has roles in regulation of the G1 phase and in motility; 5) targeted removal of the SECIS-binding protein (Secisbp2) that forms complexes with the Sec insertion sequence, which is a stem loop structure present in the 3'-UTR of all mammalian selenoprotein mRNAs and is responsible for directing Sec attached to its tRNA into protein, was embryonic lethal, wherein the embryos implanted but died before gastrulation, and conditional Secisbp2 knockout in hepatocytes dramatically reduced selenoprotein expression and reduced the expression of some, but not all, selenoprotein mRNAs; and interestingly, the overall effects of Secisbp2 loss in hepatocytes was not as detrimental as SectRNA gene (designated Trsp) loss; 6) targeted removal of Secisbp2 in brain reduced cerebral selenoprotein expression, but to a lesser extent than removal of Trsp which permitted us to examine several parameters of selenoprotein synthesis that was not possible by Trsp loss in brain tissue, e.g., PV+ interneuron density was reduced in the somatosensory cortex, hippocampus and striatum, and in situ hybridization for Gad67 (glutamic acid decarboxylase 67) confirmed the reduction of GABAergic (where GABA is ?-aminobutyric acid) interneurons; and 7) examined proteins that interact with selenoprotein S (SelS) and its mutant forms and identified all previously known SelS targets and an additional almost 200 proteins, wherein the findings suggested that SelS functions in intracellular membrane transport and maintenance of protein complexes through anchoring them to the endoplasmic membrane. We continued our ongoing collaboration with Dr. Krishna Chatterjee on assessing the effect of a C-G mutation at position 65 in Sec tRNA on selenoprotein synthesis in an eight year old male. The patient suffers from fatigue, muscle weakness and abdominal discomfort. We completed our commitments to this project in the last year by analyzing the quantity and isoform distribution of Sec tRNA produced in patient cells, and by analyzing the modified bases produced by position 65 mutant sec tRNA injected into Xenopus oocytes. In addition, we previously observed that the knockout of Trsp resulted in the total loss of selenoprotein synthesis in mouse liver, but the animals survived for several months without any apparent phenotype, whereas loss of the single selenoprotein, glutathione peroxidase 4, in liver resulted in death within the first 24-48 hours after birth. In the past year, we have shown that GPx4 is critical for hepatocyte survival and proper liver function, and that vitamin E administered in the diet may compensate for loss of GPx4 by protecting cells against membrane lipid peroxidation. Also in the last year, we completed and are preparing for publication our project on examining the antioxidant proteins in liver and lung tumors and comparing them to the corresponding surrounding normal tissues and to each other. We had observed major differences in the thioredoxin and glutathione systems between liver and lung cancer tissues, differences in the antioxidant enzymes, superoxide dismutase 1 and catalase, and found that TR1 levels were elevated in both malignant tissues compared to surrounding normal tissues as assessed by western blotting, but their enzymatic activities were similar suggesting amino acid replacement of the Sec moiety.
Showing the most recent 10 out of 15 publications
