Roger A. Sunde
Nutritional Sciences and Biochemistry
University of Missouri
Columbia, MO 65211
I would first like to express my high appreciation to Dr. Moxon and to his colleagues at Ohio State for the invitation to this celebration of Dr. Moxon's 90th birthday. The discoveries of the present generation are founded on those of the past, and this is certainly true for selenium. As you will see, the before-their-time observations of astute scientists can provide expert guidance for today's researchers, if we take but a little time to pay attention. I'm afraid, however, that today's web-based 500 MHz-paced students and scientists are increasingly overlooking this rich source of inspiration and data. Thus, I would like to pay homage to the key scientists who paved the way for the exciting unraveling of the molecular biology of selenium.
We've heard today about the story of Ft. Madison, in what is now the southeastern corner of South Dakota, and of the key studies conducted by Drs. Moxon (Moxon and Rhian, 1943) and Rosenfeld and Beath (1964). These careful studies illustrated to me the basic paradigm of nutritional science: a disease model that could be caused or prevented by dietary manipulation, and followed by logical reductive manipulation of the diet to identify the nature of this substance. To a young undergraduate, these painstaking reports on the development of the case that Se was the causative agent of a series of animal diseases were a lesson about the pattern of quality research. Clearly, there was something special about Se.
A parallel story emerged a generation later, in 1957, as Dr. Klaus Schwarz (Schwarz and Foltz, 1957) searched for and found three agents that prevented liver necrosis in rats: vitamin E, cysteine and Se. You should note that this was also the year that Mills (1957) discovered glutathione peroxidase (GPX), the enzyme that protects erythrocytes against hemolytic anemia. This was also at the end of the era when it seemed that demonstration of an additional essential nutrient was just around every bend, if one was clever enough to find the right experimental paradigm. As we know now, this stream has slowed to a trickle, but strides in our understanding of biochemistry shifted attention to finding molecular roles for the dietary essential nutrients.
John Rotruck, working under Prof. William Hoekstra's guidance in the Department of Biochemistry at the University of Wisconsin, was a careful reader of the literature. He zeroed in on the enzyme glutathione peroxidase, focused on the role of the pentose phosphate pathway and glutathione in the prevention of hemoglobin oxidation, and found that GPX was a Se-dependent enzyme (Rotuck et al., 1973). A little sidelight here is that one of the key literature reports that pointed toward GPX was the work of Gitler, Sunde and Bauman (1958) (where Sunde is my father, M.L. Sunde) that showed that media glucose had no effect on protection against hemolysis by vitamin E, but that Se's role required glucose in the media.
I was an undergraduate at Wisconsin who signed-up for Prof. Hoekstra's course on nutritional metabolism in Spring 1971. As part of the course, I had to write a paper postulating a biochemical role for a nutrient. I selected Se. I also signed up with my advisor, Professor Hoekstra, to do a senior thesis project the following Fall, and then went west to work in the mountains of New Mexico. Over that summer, John Rotruck put Se into GPX, and I came back to be asked to feed graded levels of Se to rats and to measure GPX activity (Hafeman et al., 1974). Selena, that goddess of the moon, had looked down and handed me my career.
So here I stand before you as a junior member on this August program. My task today is to review the molecular biology of Se, and in so doing, remind myself and you of the debt that I and the others in this generation of Se researchers owe to these pioneers -- Dr. Moxon, Dr. Schwarz, Dr. Hoekstra, Dr. Stadtman, Dr. Levander, my father, and others -- who still look over my shoulder and still whisper good ideas. Let us begin this journey. A more complete, but less historical, review of this material is available (Sunde, 1997).
When weanling rats are fed a Se-deficient torula yeast-based diet,
basically Schwarz's diet, liver GPX activity falls to virtually zero in about 24
days (Figure 1) (Hafeman et al., 1974). We now do this a bit more
sophisticatedly, but the experimental data remain virtually the same as we
obtained that Fall of 1971 and Winter of 1972. In fact, at that time suppliers
were obviously not supplementing their rat colonies with Se, we found that .05
ppm Se was required to prevent a 15% decrease in growth rate. Rats purchased
today and fed these same diets supplemented with vitamin E and with adequate
sulfur amino acids do not show any effect of Se on growth. We further found that
.1 ppm Se was required to maximally raise liver GPX activity to plateau levels.
What impressed me most was that additional Se beyond this point resulted in no
further increases in GPX activity, indicating that once the requirement was met,
something other that Se was rate-limiting in the accumulation of GPX.
| Figure 1. Time course of changes in liver GPX activity in rats fed graded levels of Se. |
|---|
 |
In 1972, the implications of Jacob and Monod's (1961) pioneering studies on the expression and regulation of ß-galactosidase in Escherichia coli had clear implications on mammalian biochemistry, and we knew the central dogma of molecular biology was DNA to mRNA to protein. For essential elements, the accepted view was that the metal was bound to the apoprotein after translation, following the motif that Bert Vallee had carefully shown with zinc removal from and addition back to metallothionein. It seemed logical that this could also occur with Se, in spite of the fact that this group VIa element is never cationic. Rotruck, Hafeman and Oh in Hoekstra's lab showed that injection of 75Se into rats or sheep resulted in radiolabeled GPX. I dialyzed Se-deficient erythrocyte lysates with Se as selenite or with the components of Ganther's selenite reduction pathway, but was unable to restore GPX activity (Hafeman et al., 1974; Rotruck et al., 1973). This indicated that something more than simple addition of Se was necessary to restore GPX activity in preparations from Se-deficient animals.
The nature of the Se moiety in GPX was slow to emerge. Following Rotruck's discovery, Flohé in Germany also found that purified GPX contained stoichiometric quantities of Se (Flohé et al., 1973). In 1976, Stadtman (Cone et al., 1976) reported that the Se in the bacterial enzyme, glycine reductase, was present as selenocysteine. In today's era of genome sequencing, extension to animal proteins would hardly deserve a nod, but Al Tappel and colleagues used the same approach to show in 1978 that the Se moiety in GPX was also selenocysteine (Forstrom et al., 1978). Equally importantly, they reported that this selenocysteine was incorporated into the peptide backbone of GPX (Zakowski et al., 1978). They further provided preliminary evidence that the Se as preformed selenocysteine was incorporated during translation in a tRNA-mediated mechanism.
Knowing that a substantial fraction of radioactive 75Se was incorporated into GPX in vivo, I searched for a model in vitro system that would incorporate 75Se into GPX at rates comparable to those in vivo. The erythrocyte-free isolated perfused rat liver met the requirement. I also searched the literature on Se and sulfur metabolism and found a wealth of knowledge; Drs. Moxon, Franke, Olson, Allaway, Shrift, Thomson, Robinson, Mathias, Ganther, McConnell and Burk all provided me with a foundation to formulate a diagram of Se metabolism, now modified many times, that still underlies my view of Se biochemistry (Figure 2) (Sunde, 1997). In the perfused rat liver experiment, I found that inorganic Se, as selenite or selenide, efficiently was metabolized to the form in GPX, but that 75Se as preformed selenocysteine had to pass through the inorganic pool before it was readily incorporated into GPX (Sunde and Hoekstra, 1980). This suggested a post- or co-translational mechanism for Se incorporation.
| Figure 2. Diagram of selenium metabolism. |
|---|
 |
The next key component arose from researchers focused on cancer. Paul Harrison and colleagues (Chambers et al., 1986) at the Beatson Institute in Scotland serendipitously focused on a protein that they found expressed at elevated levels in erythrogenesis, timed just as the nucleus is destroyed peroxidatively in reticulocytes. They cloned and sequenced the gene for this mouse protein but had to wait two years for identification until Flohé's amino acid sequence was posted on the protein sequence databanks. The protein was GPX1. Most importantly, the gene sequence revealed that the codon at the position corresponding to the selenocysteine moiety was the nonsense codon, UGA, which is universally one of three termination codons. Chambers et al. further speculated that this process might involve a unique tRNA, perhaps similar to a suppressor-tRNA described by Dolph Hatfield (1985). Se biochemistry had become molecular, as a Se-specific mechanism would be needed to differentiate between termination and Se incorporation.
During the same period, and after a postdoctoral fellowship at the Rowett Research Institute in Scotland, I took a faculty position at the University of Arizona and returned to the question of the mechanism of formation of the Se moiety. Using antibodies, we showed that GPX protein, as well as activity, disappeared during Se deficiency, thus demonstrating a coordinate regulation of protein level and activity by Se. Using the perfused rat liver model again, we showed that serine provided the carbon skeleton for the selenocysteine moiety of GPX (Sunde and Evenson, 1987). Furthermore, the specific activities of the selenocysteine and serine in GPX were quite similar, and we hypothesized that the same serine precursor pool that provides serine for esterification to tRNASer also provides serine for the pathway that synthesizes selenocysteine from serine and inorganic Se. The Hatfield suppressor tRNA, which recognizes UGA codons, which is acylated with serine by standard synthetases, and which is converted to phosphoserine-tRNA while attached to the tRNA, provided a molecular scheme for co-translational synthesis and insertion of selenocysteine into GPX as the peptide backbone was synthesized (Sunde and Evenson, 1987).
Confirmation of such a co-translational mechanism for Se incorporation was quick. August Bock and colleagues found that Se incorporation into bacterial formate dehydrogenase required a unique bacterial tRNA, similar to the mammalian serine suppressor tRNA, used serine, and resulted in co-translational incorporation of selenocysteine into this bacterial enzyme (Leinfelder et al., 1988). This further emphasized this co-translational mechanism of selenocysteine synthesis and insertion is an ancient solution to provide Se for crucial metabolic catalysts.
The coordinate down-regulation of GPX protein and activity during progressive Se deficiency thus was explained. I wondered, however, if Se deficiency would also result in down-regulation of the GPX mRNA. To answer this question, we used the probes from the cloned mouse GPX to determine if there was also a down-regulation of GPX mRNA in Se-deficient rats. We found that GPX mRNA levels in Se-deficient rat liver were approximately 10 percent of the levels found in Se adequate rat liver (Saedi et al., 1988). Furthermore, when weanling rats were fed a Se deficient diet, there was an exponential down-regulation of GPX activity, protein and mRNA level. The half-life of GPX activity and mRNA was approximately three days (Sunde et al., 1989). One could explain a lack of synthesis of GPX protein because Se deficiency depleted selenocysteine for insertion into GPX, but there was no known molecular role that required Se for the synthesis of mRNA. This raised a new question, which, as you'll see below, we have spent considerable time studying. More importantly, the modulation of level of GPX mRNA by Se status offers a potential, most sensitive parameter which could be used for the assessment of Se requirements.
We conducted a series of experiments that take advantage of the regulation of selenoprotein expression to determine minimal Se requirements. In these experiments, weanling rats were fed a Se-deficient torula yeast-based diet supplemented with graded levels of Se, just as we did in that first study 28 years ago, but we used molecular biology techniques to analyze for the changes in Se-dependent parameters (Weiss et al., 1996; 1997). Notably, today's weanling rats with adequate Se stores showed no dietary Se requirement for growth. Liver GPX1 is the major form of liver Se. Liver Se and liver GPX1 activity showed a sigmoidal response to increasing dietary supplementation: initial increments of dietary Se elicited little change as the limited Se was directed elsewhere. Then there was a rapid increase in Se/GPX1 as dietary Se increased. Then a plateau was reached where additional increments in dietary Se did not give rise to any further change (Figure 3). On these sigmoidal curves, we graphically determined a "plateau breakpoint" as the intersection of a line tangent to the point of maximum slope with the plateau line (Weiss et al., 1996). The plateau breakpoint thus is the minimal level of dietary Se, or minimum requirement, that elicits the maximal level of the measured parameter. Comparison of plateau breakpoints for various parameters thus allows direct comparison of the minimum dietary Se requirements necessary for maximum expression of Se-dependent biochemical parameters.
| Figure 3. Effect of dietary selenium on GPX1 and GPX4 (PHGPX) activity and mRNA levels. |
|---|
 |
When we looked at regulation of GPX1 mRNA (Figure 3), we again saw a sigmoidal curve, but the plateau breakpoint occurred at about .05 ppm Se. mRNA levels reached a maximum at about half of the level of dietary Se necessary for maximum GPX1 protein and activity (Weiss et al., 1996;1997). This clearly showed that the underlying molecular mechanism regulating GPX1 mRNA is distinct from the mechanism regulating synthesis of this selenoprotein (presumably simply due to availability of Se/selenocysteine). Interestingly, female rats have twice as much GPX1 mRNA and protein, and yet the requirements are the same as for male rats, indicating that this regulation is not just simple supply and demand.
So what about other selenoproteins? We cloned and sequenced the cDNA for GPX4, and found that this peroxidase with 40% nucleotide and amino acid sequence identity is differentially regulated as compared to GPX1 (Sunde et al., 1993). Activity only decreased to about 50% in Se deficiency, and GPX4 mRNA was not regulated substantially by dietary Se (Figure 3) (Lei et al., 1995). We further evaluated other selenoprotein mRNA expression (Table 1) and found that GPX1 mRNA is uniquely and highly regulated as compared to other selenoprotein mRNAs, that the minimum Se requirement for expression of the other selenoproteins occurs at the same level of dietary Se necessary for maximum level of GPX1 mRNA, and that the dietary level of Se necessary for plateau levels of liver and blood GPX1 activity is double that required for the other selenoproteins, as well as GPX1 mRNA. We used this approach to evaluate Se requirements under conditions less amenable to evaluating requirements, such as pregnancy, and found that Se requirements in pregnant and lactating rats and in older rats decreased rather than increased as was reported in the literature (Table 1) (Sunde et al., 1999).
| Table 1. Minimal dietary selenium requirements in female rats. | ||||||||
|---|---|---|---|---|---|---|---|---|
| Parameter | Weanling rats | Pregnant rats | Lactating rats | One-year old rats | ||||
| Def 1 | Req2 | Def 1 | Req2 | Def 1 | Req2 | Def 1 | Req2 | |
| (%) | (µg/g) | (%) | (µg/g) | (%) | (µg/g) | (%) | (µg/g) | |
| Growth | 100 | <.007 | 101 | <.007 | 92 | <.007 | 126 | <.007 |
| Erythrocyte GPX1 activity | 40 | .1 | 17 | .05 | 19 | .05 | 26 | <.05 |
| Plasma GPX3 activity | 8 | .07 | 14 | .05 | 4 | .1 | 18 | <.05 |
| Liver Se | 4 | .1 | 4 | .075 | 4 | .075 | 5 | <.05 |
| Liver GPX1 activity | 2 | .1 | 7 | .05 | 5 | .075 | 3 | <.05 |
| Liver GPX1 mRNA | 11 | .05 | 19 | .05 | 18 | .05 | ||
| Liver GPX4 activity | 29 | .05 | 35 | .05 | 23 | .05 | 39 | <.05 |
| Liver GPX4 mRNA | 58 | <.02 | 77 | <.02 | 87 | <.02 | ||
| Liver Sel P mRNA | 92 | <.02 | 86 | <.02 | ||||
| Liver 5'D1 mRNA | 87 | <.02 | 68 | <.02 | ||||
|
1 Level of parameter in Se-deficient rats relative to rats given .2 µg Se/g diet.
2 Minimal dietary requirement (plateau breakpoint) necessary to achieve plateau level for specified parameter (Weiss et al., 1996). | ||||||||
This unique regulation of GPX1 mRNA suggests that the function or at least a major function of GPX1 involves this regulation by Se status (Sunde 1994; 1997). Thus we proposed that the major role of GPX1 is as a biological Se buffer (Sunde 1997). GPX1 expression can be used effectively by Dr. Levander and others to establish Se requirements, because it is part of the homeostatic mechanism cells and organisms used to sense and regulate Se status.
So how does Se status regulate GPX1 expression? Marla Berry found the next critical piece of the molecular regulation when she discovered that a stem-loop structure was required in the 3'UTR (untranslated region) of deiodinase mRNA if Se was to be inserted at the UGA codon during translation (Berry et al., 1991a). This stem-loop, now called a SECIS (selenocysteine insertion) element, is present in all known mammalian selenoproteins. Prokaryotic selenoproteins mRNA also have a SECIS element, but it is located in the coding region of the mRNAs, immediately downstream of the UGA codon.
To incorporate Se into selenoproteins, serine esterification to the tRNA by the usual aminoacyl tRNA synthetases is the first step (Figure 4), reviewed in (Sunde, 1997). An activated Se intermediate, selenophosphate (HSePO4-2), is synthesized from selenide and ATP by a unique enzyme, selenophosphate synthetase (SelD). Then a second unique enzyme, selenocysteine synthetase (SelA), catalyzes the substitution of the serine hydroxyl with a selenol from selenophosphate with the tRNA acting like a biological blocking group to facilitate recognition of this unique seryl-tRNA. The resulting selenocysteyl-tRNA binds to a third necessary protein (SelB), only identified in prokaryotes, which is an elongation factor that also binds GTP and is recruited by the 3'UTR SECIS element, forming a tethered quaternary complex that orients the tRNA anticodon at the UGA in the A site of the ribosome and facilitates cotranslational insertion of selenocysteine at the position specified by UGA in the open reading frame of the selenoprotein mRNA. All mammalian selenoproteins use this mechanism for cotranslational synthesis of selenocysteine from inorganic Se and serine, followed by incorporation of this selenocysteine into the growing polypeptide chain during translation. This mechanism, clearly conserved for the most part in prokaryotes and eukaryotes, emphasizes the importance of retaining the ability to make and use selenoenzymes in biology.
| Figure 4. Diagram of selenocysteine (Sec) synthesis and incorporation into protein during translation. |
|---|
|
|
Careful studies by a number of researchers have shown that Se status does not affect transcription rates of any selenoproteins. So why is GPX1 mRNA uniquely regulated by Se status? We conducted a series of experiments with recombinant fusion genes in transfected cells to partially unravel this mechanism. Our initial hypothesis was that the SECIS of GPX1 was necessary for Se incorporation and uniquely necessary for Se regulation. We found that the fusion gene mRNAs were regulated in cultured cells by medium Se concentration, and that deletion of the 3'UTR resulted in loss of Se regulation, as well as ability to incorporate Se (Weiss and Sunde, 1997). We conducted a series of replacements of regions of the GPX1 mRNA with corresponding regions of the unregulated GPX4 mRNA. We found, however, that the GPX4 3'UTR is equally effective in conferring Se regulation onto GPX1 mRNA as is the mouse or rat GPX1 3'UTR, indicating that there is nothing unique about the GPX1 3'UTR or SECIS element (Weiss and Sunde, 1998). We next examined the coding region and found that mutation or substitution of sequences within the GPX1 coding region would result in loss of Se regulation. Finally, we found that a UGA codon located before the intron was sufficient to confer regulation to a GPX1 mRNA with a SECIS in the 3'UTR, and that we could make a recombinant -globin gene which had mRNA levels regulated by Se status if it contained a GPX1 3'UTR and a UGA 5' to an intron (Weiss and Sunde, 1998). This positioning of a nonsense codon in front of an intron is known to target mRNAs for "nonsense mediated decay," which is hypothesized to be a mechanism to remove mutant, untranslatable mRNAs from the cell. It may be that at this point the pathway for Se regulation of GPX1 mRNA levels thus rejoins and utilizes a general pathway for mRNA degradation.
Collectively, these studies indicate that both a UGA and a 3'UTR SECIS element are required for Se incorporation into a selenoprotein. In addition, the UGA must be positioned 5' to an intron in order to confer Se regulation of mRNA stability. This increasing-ly detailed understanding of the factors necessary for Se regulation of GPX1 mRNA stability, however, still does not reveal the actual selenostat, the Se thermostat, that senses the level of intracellular Se and regulates GPX1 mRNA level accordingly. Our current best hypothesis is that unique aspects of GPX1 mRNA structure, including placement of the UGA in front of the intron and a relatively low affinity of the SECIS to recruit selenocysteyl-tRNA for translation, make GPX1 mRNA the last selenoprotein mRNA to be fully occupied during translation under conditions of limited Se availability, thus making GPX1 mRNA most susceptible to nonsense mediated decay. Our recent studies on the importance of UGA position for translation efficiency (Wen et al., 1998) and new studies showing that the translational efficiency of GPX1 in rat liver is reduced compared to other selenoproteins (in preparation), support this hypothesis.
I want to finish by briefly discussing two additional series of exciting new experiments. Dr. Ho (Spector et al., 1996) at Wayne State University recently knocked-out the GPX1 gene in mice, and found that loss of ability to synthesize GPX1 under normal conditions is without phenotype! In those heady GPX1-is-a-selenoenzyme days of the 70's, this would have been heresy, but the discovery of other unregulated members of the glutathione peroxidase family and of other Se-dependent gene families, and our findings that Se regulation is a unique aspect of the GPX1 function, collectively make this no longer shocking. I should remind you that the selenocysteine tRNA knockout is fetally lethal (Bosl et al., 1997), demonstrating unequivocally that Se is essential.
We studied Se regulation in the GPX1 knockout mice, and found that the dietary Se regulation of other Se-dependent genes and proteins is unaffected by Se status (in preparation). Without GPX1, however, weanling mice do not have sufficient stores of Se to sustain growth past day 14 of weaning, and the plateau breakpoint for liver Se shifts from .1 to .05 ppm. These studies illustrate the role of GPX1 as an important biological Se buffer in maintaining Se status.
We also went back to restudy the Se deficiency conditions that Schwarz
reported in the 1960's, and that we observed in the early 1970's. When
Se-deficient weanling rat pups from Se-deficient dams were fed a Se-deficient
diet, they grew at half the rate of litter mates supplemented with Se (Thompson
et al., 1995). More interestingly, when we injected these rats at day 14
post-weaning with a single intraperitoneal injection of as little as 1 µg Se/100
g rat, the growth rate of the rats increased significantly within 24 hours.
Altered levels of GPX1 or circulating triiodothyronine (T3) were not responsible.
Rapid changes in level of liver Se and GPX4 in testes to date were the best
parameters that correlate with reversal of the growth defect (Thompson et al.,
1998). Fortunately for animal Se researchers, this limiting biochemical defect
is yet to be characterized, and thus illustrates but one of the many biochemical
functions of Se still looking for a protein. Alternatively, might Se have a
direct role in RNA structure and function?
My scope here was to review the emerging molecular biology of Se, including some of our research results, and to illustrate how this new knowledge is helping us better understand the role of Se in biology in general, and in animals and humans specifically. GPX1 has undoubtably been a boon for assessing Se status in humans and animals and is the basis for the setting of human and animal requirements. The recent difficulties of the National Research Council to set DRIs (daily recommended intakes) for nutrients long known to be required, such as calcium, illustrate how important it is to have meaningful biochemical indicators of nutrient status as a basis to set requirements. Our studies on the underlying molecular mechanism of Se regulation of GPX1 illustrate that this relationship is not coincidental, but instead is based on an evolutionarily-conserved regulatory mechanism. Se is becoming the gold-standard to which other nutrients are being compared in selecting meaningful markers of status.
This is but one chapter in the role of Se in biology. The discovery that pox viruses have captured mammalian GPX and apparently express it in a "star wars" defense against cellular antiviral mechanisms (Shisler et al., 1998), illustrates that novel roles for Se in health and disease are yet to emerge as well as be understood. The novel role that Melinda Beck and Orville Levander (Levander and Beck, 1997) are uncovering for Se in protection against virus infection may have important implications for future nutrition supplementation, as well as disease treatment, and may help to explain the signs of Se-dependent diseases so carefully chronicled in the literature but which we no longer can reproduce in today's clean rat colonies. I hear Klaus Schwarz whispering experimental ideas as I read between the lines of his papers on liver necrosis in rats and multi focal necrosis in mice. There is apparent protection against prostate cancer associated with daily ingestion of high levels of Se (Clark et al., 1996). This occurs, however, at levels way above those necessary to saturate GPX expression. Is this a biochemical effect, a pharmacological effect or some interaction with the detoxification mechanism of Se about which we have almost no information? Those careful studies conducted in sheep and cattle in South Dakota should be read carefully to help focus on where and how to look. Dr. Moxon and his colleagues (Moxon and Rhian, 1943; Rosenfeld and Beath, 1964) are whispering to us, if we only are smart enough today to listen. With the tools of molecular biology, biotechnology, human and animal genomics and microchip analysis, the answers should be only a few years away.
I would like to thank and dedicate this lecture to Dr. William G. Hoekstra, teacher, scientist and friend, who still figuratively speaks (not whispers) in my ear.
Akasaka, M., J. Mizoguchi and K. Takahashi. 1990. A human cDNA sequence for a novel glutathione peroxidase-related protein. Nucleic Acids Res. 18:4619.
Arthur, J.R., F. Nicol and G.J. Beckett. 1990. Hepatic iodothyronine 5'deiodinase: the role of selenium. Biochem. J. 272:537.
Behne, D., A. Kyriakopoulos, H. Meinhold and J. Kohrle. 1990. Identification of type I iodothyronine 5'-deiodinase as a selenoenzyme. Biochem. Biophys. Res. Commun. 173:1143.
Berry, M.J., L. Banu, Y. Chen, S.J. Mandel, J.D. Kieffer, J.W. Harney and P.R. Larsen. 1991a. Recognition of a UGA as a selenocysteine codon in Type I deiodinase requires sequences in the 3' untranslated region. Nature (London) 353:273.
Berry, M.J., L.Banu and P.R. Larsen. 1991b. Type I iodothyronine deiodinase is a selenocysteine-containing enzyme. Nature (London) 349:438.
Bosl, M.R., K. Takaku, M. Oshima, S. Nishimura and M.M. Taketo. 1997 Early embryonic lethality caused by targeted disruption of the mouse selenocysteine tRNA gene (Trsp). Proc. Natl. Acad. Sci. USA 94:5531.
Chambers, I., J. Frampton, P. Goldfarb, N. Affara, W. McBain and P.R. Harrison. 1986. The structure of the mouse glutathione peroxidase gene: the selenocysteine in the active site is encoded by the 'termination' codon, TGA. EMBO J. 5:1221.
Chu, F.F., J.H. Doroshaw and R.S. Esworthy. 1993. Expression, characterization, and tissue distribution of a new cellular selenium-dependent glutathione peroxidase, GSH-Px-GI. J. Biol. Chem. 268:2571.
Clark, L.C., G.F. Combs, B.W. Turnbull, E.H. Slate, D.K. Chalker, J. Chow, L.S. Davis, R.A. Glover, G.F. Graham, E.G. Gross, A. Krongrad, J.L. Lesher, H.M. Park, B.B. Sanders, C.L. Smith and J.R. Taylor. 1996. Effects of selenium supplementation for cancer prevention in patients with carcinoma of the skin. JAMA 276:1957.
Cone, J.E., R.M. del Rio, J.N. Davis and T.C. Stadtman. 1976. Chemical characterization of the selenoprotein component of clostridial glycine reductase: identification of selenocysteine as the organoselenium moiety. Proc. Natl. Acad. Sci. USA 73:2659.
Flohé, L., W.A. Günzler and H.H. Schock. 1973. Glutathione peroxidase: a selenoenzyme. FEBS Lett. 32:132.
Forstrom, J.W., J.J. Zakowski and A.L. Tappel. 1978. Identification of the catalytic site of rat liver glutathione peroxidase as selenocysteine. Biochemistry 17:2639.
Gitler, C., M.L. Sunde and C.A. Bauman. 1958. Effect of certain necrosis-preventing factors on hemolysis in vitamin E-deficient rats and chicks. J. Nutr. 65:397.
Gladyshev, V.N., K. Jeang and T.C. Stadtman. 1996. Selenocysteine, identified as the penultimate C-terminal residue in human T-cell thioredoxin reductase, corresponds to TGA in the human placental gene. Proc. Natl. Acad. Sci. USA 93:6146.
Hafeman, D.G., R.A. Sunde and W.G. Hoekstra. 1974. Effect of dietary selenium on erythrocyte and liver glutathione peroxidase in the rat. J. Nutr. 104:580.
Hatfield, D. 1985. Suppression of termination codons in higher eukaryotes. Trends Biochem. Sci. 10:201.
Hill, K.E., R.S. Lloyd, J.G. Yang, R. Read and R.F. Burk. 1991. The cDNA for rat selenoprotein P contains 10 TGA codons in the open reading frame. J. Biol. Chem. 266:10050.
Jocob, F. and J. Monod. 1961. Genetic regulatory mechanisms in the synthesis of proteins. J. Mol. Biol. 3:318.
Lei, X.G., J.K. Evenson, K.M. Thompson and R.A. Sunde. 1995. Glutathione peroxidase and phospholipid hydroperoxide glutathione peroxidase are differentially regulated in rats by dietary selenium. J. Nutr. 125:1438.
Leinfelder, W., E. Zehelein, M. Mandraand-Berthelot and A. Böck. 1988. Genes for a novel tRNA species that accepts L-serine and cotranslationally inserts selenocysteine. Nature (London) 331:723.
Levander, O.A. and M.A. Beck. 1997. Interacting nutritional and infectious etiologies of keshan disease: Insights from Coxsackie virus B-induced myocarditis in mice deficient in selenium or vitamin E. Biol. Tr. Elem. Res. 56:5.
May, J.M., S. Mendiratta, K.E. Hill and R.F. Burk. 1997. Reduction of dehydroascorbate to ascorbate by the selenoenzyme thioredoxin reductase. J. Biol. Chem. 272:22607.
Mills, G.C. 1957. Hemoglobin catabolism. I. Glutathione peroxidase, an erythrocyte enzyme which protects hemoglobin from oxidative breakdown. J. Biol. Chem. 229:189.
Moxon, A.L. and M.A. Rhian. 1943. Selenium poisoning. Physiol. Rev. 23:305.
Rosenfeld, I. and O.A. Beath. 1964. Selenium: Geobotany, Biochemistry, Toxicity and Nutrition. Academic Press, New York, NY, p. 1.
Rotruck, J.T., A.L. Pope, H.E. Ganther, A.B. Swanson, D.G. Hafeman and W.G. Hoekstra. 1973. Selenium: biochemical role as a component of glutathione peroxidase. Science 179:588.
Saedi, M.S., C.G. Smith, J. Frampton, I. Chambers, P.R. Harrison and R.A. Sunde. 1988. Effect of selenium status on mRNA levels for glutathione peroxidase in rat liver. Biochem. Biophys. Res. Commun. 153:855.
Schuckelt, R., R. Brigelius-Flohé, M. Maiorino, A. Roveri, J. Reumkens, W. Strassburger and L. Flohé. 1991. Phospholipid hydroperoxide glutathione peroxidase is a selenoenzyme distinct from the classical glutathione peroxidase as evident from cDNA and amino acid sequencing. Free Rad. Res. Commun. 14:343.
Schwarz, K. and C.M. Foltz. 1957. Selenium as an integral part of factor 3 against dietary necrotic liver degeneration. J. Am. Chem. Soc. 79:3292.
Shisler, J.L., T.G. Senkevich, M.J. Berry and B. Moss. 1998. Ultraviolet-induced cell death blocked by a selenoprotein from a human dematotropic poxvirus. Science 279:102.
Spector, A., Y. Yang, Y.-S. Ho, J.-L. Magnenat, R.-R. Wang, W. Ma and W.-C. Li. 1996. Variation in cellular glutathione peroxidase activity in lens epithelial cells, transgenics and knockouts does not significantly change the response to H2O2 stress. Exp. Eye Res. 62:521.
Sunde, R.A., M.S. Saedi, S.A.B. Knight, C.G. Smith and J.K. Evenson. 1989. Regulation of expression of glutathione peroxidase by selenium. In: Selenium in Biology and Medicine (Wendel, A., Ed.), Springer-Verlag, Heidelberg, Germany, p. 8.
Sunde, R.A., J.A. Dyer, T.V. Moran, J.K. Evenson and M. Sugimoto. 1993. Phospholipid hydroperoxide glutathione peroxidase: full-length pig blastocyst cDNA sequence and regulation by selenium sta-tus. Biochem. Biophys. Res. Commun. 193:905.
Sunde, R.A. 1994. Intracellular glutathione peroxidases - structure, regulation and function. In: Selenium in Biology and Human Health (Burk, R.F., Ed.), Springer-Verlag, New York, NY, p. 45.
Sunde, R.A. 1997. Selenium. In: Handbook of Nutritionally Essential Mineral Elements (O'Dell, B.L. and R.A. Sunde, Eds.), Marcel Dekker, New York, NY, p. 493.
Sunde, R.A., K.M.Thompson, J.K. Evenson and S.L. Weiss. 1999. Use of glutathione peroxidase-1 activity and mRNA levels and other selenium-dependent parameteres to assess selenium requirements in female rats thoughout the life-cycle. Proc. Nutr. Soc. (In press).
Sunde, R.A. and J.K. Evenson. 1987. Serine incorporation into the selenocysteine moiety of glutathione peroxidase. J. Biol. Chem. 262:933.
Sunde, R.A. and W.G. Hoekstra. 1980. Incorporation of selenium from selenite and selenocysteine into glutathione peroxidase in the isolated perfused rat liver. Biochem. Biophys. Res. Commun. 93:1181.
Thompson, K.M., H. Haibach and R.A. Sunde. 1995. Growth and plasma triiodothyronine concentrations are modified by selenium deficiency and repletion in second-generation selenium-deficient rats. J. Nutr. 125:864.
Thompson, K.M., H. Haibach and R.A. Sunde. 1998. Liver selenium and testes phospholipid hydro-peroxide glutathione peroxidase are associated with growth during selenium repletion of second-generation Se-deficient male rats. J. Nutr. 128:1289.
Vendeland, S.C., M.A. Beilstein, J.Y. Yeh, W. Ream and P.D. Whanger. 1995. Rat skeletal muscle selenoprotein W: cDNA clone and mRNA modulation by dietary selenium. Proc. Natl. Acad. Sci. USA 92:8749.
Weiss, S.L., J.K. Evenson, K.M. Thompson and R.A. Sunde. 1996. The selenium requirement for glutathione peroxidase mRNA level is half of the selenium requirement for glutathione peroxidase activity in female rats. J. Nutr. 126:2260.
Weiss, S.L., J.K. Evenson, K.M. Thompson and R.A. Sunde. 1997. Dietary selenium regulation of glutathione peroxidase mRNA and other selenium-dependent parameters in male rats. J. Nutr. Biochem. 8:85.
Weiss, S.L. and R.A. Sunde. 1997. Selenium regulation of classical glutathione peroxidase expression requires the 3' untranslated region in Chinese hamster ovary cells. J. Nutr. 127:1304.
Weiss, S.L. and R.A. Sunde. 1998. cis-Acting elements are required for selenium regulation of glutathione peroxidase-1 mRNA levels. RNA 4:816.
Wen, W., S.L. Weiss and R.A. Sunde. 1998. UGA codon position affects the efficiency of selenocysteine incorporation into glutathione peroxidase-1. J. Biol. Chem. 273:28533.
Zakowski, J.J., J.W. Forstrom, R.A. Condell and A.L. Tappel. 1978. Attachment of selenocysteine in the catalytic site of glutathione peroxidase. Biochem. Biophys. Res. Commun. 84:248.
