![[Feedback]](/icons/banner/feedbackACT.gif){kind=icon}
![[For Subscribers]](/icons/banner/subscriberACT.gif){kind=icon}
![[Archive]](/icons/banner/archiveACT.gif){kind=icon}
![[Search]](/icons/banner/searchACT.gif){kind=icon}
![[Contents]](/icons/banner/tocACT.gif){kind=icon}
|
|
|
|
|
||
Untranslated Region in Chinese Hamster Ovary Cells1,2,3
Departments of Biochemistry and Nutritional Sciences, University of Missouri, Columbia, MO 65211
ABSTRACT
INTRODUCTION
MATERIALS AND METHODS
RESULTS
DISCUSSION
ACKNOWLEDGMENTS
FOOTNOTES
LITERATURE CITED
ABSTRACT 
Classical glutathione peroxidase (GPX) mRNA levels fall dramatically in selenium (Se)-deficient animals, but it is not known whether this mechanism is related to the mRNA 3
untranslated region (3UTR) sequences that have been shown to direct Se incorporation. In this study, we used recombinant GPX constructs to investigate the role of the GPX 3UTR in Se regulation of GPX mRNA levels in Chinese hamster ovary (CHO) cells. The CHO cells were transfected with GPX (pRc/GPX), GPX lacking the 3UTR (pRc/
3UTR) or the pRc/CMV vector alone, and GPX activity and GPX mRNA levels were determined in stable transfectants grown in low Se basal medium with a range of added Se concentrations. We identified two pRc/GPX transfectants with significantly elevated GPX activity levels compared with pRc/CMV transfectants. The elevated GPX expression did not dramatically shift the amount of Se that was sufficient for GPX activity to reach the Se-adequate plateau level (100 nmol/L added Se). As expected, GPX activity was not significantly different when pRc/3UTR transfectants were compared with pRc/CMV control transfectants. Among the wild type and transfected CHO cells, Se-deficient GPX activity levels averaged 35 ± 5% of Se-adequate levels. Selenium-deficient levels of endogenous GPX mRNA as well as recombinant pRc/GPX mRNA averaged 54-58% of Se-adequate levels; 3-4 nmol/L added Se was sufficient for maximal GPX mRNA levels. In contrast, pRc/3UTR mRNA levels in the unsupplemented cells remained at Se-adequate levels and showed no distinct Se regulation. These studies demonstrate that the GPX 3UTR is necessary for Se regulation of GPX mRNA levels in addition to its role in Se incorporation.
INTRODUCTION
Classical selenium (Se)-dependent glutathione peroxidase (GPX, glutathione:H2O2 oxidoreductase, EC 1.11.1.9)5 activity can fall to undetectable levels in Se-deficient rats (Hafeman et al. 1974
). Glutathione peroxidase protein levels (Knight and Sunde 1987 and 1988, Takahashi et al. 1986) also decline dramatically in Se deficiency, and Se-deficient GPX mRNA levels fall to 10% of Se-adequate levels (Lei et al. 1995, Saedi et al. 1988, Sunde et al. 1991, Toyoda et al. 1990, Weiss et al. 1996 and 1997, Yoshimura et al. 1988). The mechanism for Se regulation of GPX mRNA levels is unclear, but transcription of GPX mRNA and transport of GPX mRNA out of the nucleus are not affected by Se status, indicating that Se regulation occurs post-transcriptionally (Christensen and Burgener 1992, Moskow et al. 1992, Sugimoto and Sunde 1992).
Glutathione peroxidase activity has been reported to drop to 4% of Se-adequate levels in Se-deficient HL-60 cells (Chada et al. 1989, Takahashi et al. 1986) and 7% of Se-adequate levels in Se-deficient Hep3B cells (Baker et al. 1993), suggesting that cultured cells would be a good model for further studies on Se regulation of GPX. Glutathione peroxidase mRNA levels can also decrease in Se-deficient cultured cells but have not been reported to drop below 30% of Se-adequate levels for unknown reasons (Baker et al. 1993, Chada et al. 1989). In an experiment using a pRSV-GPX expression vector and MCF-7 cells, which have undetectable levels of endogenous GPX mRNA, transfected GPX mRNA levels in Se-deficient cells were 32% of the levels found in Se-adequate cells (Chu et al. 1990).
3UTR8 and pRc/3UTR12 cells grown in Se-adequate medium for 3 d. Total RNA (30 µg/lane) was subjected to formaldehyde denaturing electrophoresis, transferred to a nylon membrane and hybridized with the glutathione peroxidase (GPX) probe (A), the CMV probe (B) and the 18S rRNA probe (C). Gel positions of 28S and 18S rRNA bands are shown. A: The GPX probe hybridized to the 13S endogenous GPX mRNA, the 16S GPX mRNA in pRc/GPX transfectants, and the 14S GPX mRNA in pRc/3UTR transfectants. B: The CMV probe hybridized only to the 16S and 14S GPX mRNAs found in GPX-transfected cells. C: The signal for 18S rRNA was used to normalize GPX mRNA levels in each transfectant.
To study Se regulation of GPX mRNA levels, we chose Chinese hamster ovary (CHO) cells, which have moderate levels of endogenous GPX expression, to compare regulation of recombinant and endogenous GPX mRNAs in the same cell population. Furthermore, we investigated the role of the 3 untranslated region (3UTR) in Se regulation of GPX because Se incorporation into selenoproteins has been shown to require a selenocysteine insertion sequence (SECIS) motif in the mRNA 3UTR (Berry et al. 1991 and 1993, Shen et al. 1993). Here we report that transfected GPX mRNA levels decrease in Se deficiency and respond to graded levels of supplemental Se in a manner similar to endogenous GPX mRNA levels. In addition, we demonstrate that the GPX 3UTR is necessary for this Se regulation, suggesting that the GPX 3UTR contains a Se-responsive element with a role in the reduction of GPX mRNA levels during Se deficiency.
MATERIALS AND METHODS
), was isolated from a cDNA library (catalogue no. 936513, Stratagene, La Jolla, CA) using the 0.7-kb EcoRI fragment of mouse genomic GPX that we used previously for Northern blot analysis (Saedi et al. 1988). The isolated GPX clone contained 34 bp of 5UTR, the full 606-bp GPX coding region and 204 bp of 3UTR. To construct the full-length GPX expression vector (pRc/GPX), the 951-bp Xbal/Apal fragment from the pBluescript II SK
multiple cloning site (Stratagene) containing the GPX cDNA was ligated into the Xbal/Apal restriction sites of the eukaryotic expression vector pRc/CMV (Invitrogen, Carlsbad, CA) (Fig. 1). This placed the GPX cDNA in the sense orientation relative to the CMV promoter. To construct the 3UTR GPX expression vector (pRc/3UTR), pRc/GPX was digested with Bsu36I and Apal to excise the GPX 3UTR, and the remaining cohesive ends were blunted by T4 DNA polymerase and then religated to form pRc/3UTR; the GPX stop codon was retained in this pRc/3UTR construct. Sequences were confirmed using the TaqDyeDeoxy terminator (Applied Biosystems, Foster City, CA) at the University of Missouri DNA core facility.
UTR (Bsu36I/Apal) was removed from pRc/GPX to construct pRc/3UTR. The HindIII/Xbal fragment of the pRc/CMV multiple cloning site was used as a CMV probe for transfected GPX transcripts. (X, Xbal; B, Bsu36l; A, Apal; H, Hindlll; [Se]Cys, selenocysteine UGA codon; SECIS, selenocysteine insertion sequence).
Culture and transfection of Chinese hamster ovary cells. The CHO cells (CHO-K1, American Type Culture Collection, Rockville, MD) were maintained in Ham's F-12 medium with 2 mmol/L L-glutamine under standard incubation conditions (37°C, 5% carbon dioxide, 95% air, 95% humidity). Unless otherwise stated, fetal bovine serum (FBS) was maintained at 10% of the medium and the same lot of "low Se" serum was used for all culturing and experiments with these cells. Low selenium serum. A low Se lot of FBS (no. 4001635, Biocell, Rancho Dominguez, CA, containing 224 nmol/L Se) was identified by screening various lots of FBS by neutron activation analysis (McKown and Morris 1978
). Medium with 10% FBS thus contained 22 nmol/L Se, and this basal medium is referred to as "low Se." In a limited, second series of experiments, identified as "1% FBS" experiments, we used a second lot of low Se FBS (no. 3001L, Atlanta Biologicals, Norcross, GA, containing 170 nmol/L Se), reduced the medium FBS content to 5% (9 nmol/L Se) for cell maintenance and transfection, and further reduced the medium FBS to 1% (2 nmol/L Se) for Se regulation experiments.
Transfection and selection.
Calcium phosphate-precipitated DNA was used to transfect CHO cells (Ausubel et al. 1989). Briefly, CHO cells were plated at a density of 9 × 105 cells per 16 mL in 10-cm tissue culture plates and grown for 24 h. Calcium phosphate-precipitated pRc/GPX, pRc/3UTR or pRc/CMV DNA (20 µg/plate) was then added to the medium, and the cells were incubated for 16 h. The medium was removed, and cells were washed twice with PBS and then fed with fresh medium. After an additional 48 h, fresh, selective medium containing 600 mg G418/L (Gibco BRL, Gaithersburg, MD) was used to isolate stable cell lines that had incorporated the expression vector containing the neomycin-resistance gene. G418-resistant colonies were visible, without the use of a microscope, 2 wk after transfection. We continued to use G418-supplemented selective medium for all experiments with these transfectants. For the experiments with isolated transfectants, individual G418-resistant colonies were transferred to separate flasks and the cells were grown to confluency. An initial GPX activity screen was conducted by splitting newly confluent cells 1:10 into low Se basal medium with or without supplemental Se (1 µmol/L Se) (1 plate per treatment). Glutathione peroxidase activity was determined after 3 d of growth. For each transfectant, the remaining cells were divided into 10 aliquots and frozen in liquid nitrogen so that further experiments could be conducted with cells having an equal passage number post-transfection. To conduct experiments with individual transfectants, one aliquot was removed from liquid nitrogen and grown for 4 d in low Se basal medium. Cells were split 1:15 into 10-cm tissue culture dishes (6 × 105 cells per plate) containing the low Se basal medium supplemented with graded levels of Se (0, 25, 50, 100, 200 and 400 nmol/L Se as Na2SeO3 for GPX activity experiments, and 0, 1.25, 2.5, 3.8, 6.3 and 12.7 nmol/L Se for GPX mRNA experiments), grown for 3 d, and then harvested by mechanical scraping for further analysis. Three replicate plates for each transfectant at each Se level were grown for GPX activity experiments. Because two plates were needed to isolate a useful amount of total RNA, two replicate plates of each transfectant at each Se level were pooled for RNA isolation.
Experiments with 1% fetal bovine serum medium.
A limited series of experiments was conducted using cells transfected and grown in basal medium containing reduced levels of FBS and thus Se. For these experiments, CHO cells were maintained and transfected in Ham's F-12 medium containing 5% of a second lot of FBS, which provided a medium Se content of 9 nmol/L Se. After transfection and selection, cells were split 1:15 into 10-cm tissue culture dishes containing medium with 1% FBS, providing a basal Se level of 2 nmol/L Se. In these experiments, 1:15-split cells were grown for 4 d in basal medium or medium supplemented with 20 nmol/L Se as Na2SeO3 . All other aspects were conducted the same as for experiments in 10% FBS.
) with hydrogen peroxide, so that only Se-dependent GPX activity was measured. Protein was determined by the method of Lowry et al. (1951).
Northern blot analysis.
In an individual experiment, two replicate 10-cm plates of cells for each Se level were pooled, harvested and homogenized in the presence of guanidine isothiocyanate and sarkosyl, and total RNA was isolated by centrifugation (20 h) on 5.7 mol/L CsCl as described previously (Saedi et al. 1988). Total RNA (30 µg/sample) was then separated by electrophoresis in 1.5% agarose gels containing 6% formaldehyde. The positions of the 18S and 28S ribosomal RNA were noted after staining with acridine orange. RNA was capillary transferred to modified nylon 66 membranes (Pall Biosupport, East Hills, NY) and hybridized with 25 ng of 32P-labeled GPX probe (700-bp EcoRI fragment from murine GPX) (Saedi et al. 1988) or CMV probe (94-bp HindIII/Xbal fragment of the pRc/CMV multiple cloning site) (Fig. 1). Hybridization probes were labeled (Oligolabeling kit, Pharmacia LKB, Piscataway, NJ) with [
-32P]dCTP (37 kBq/ng, Du Pont NEN, Wilmington, DE). Incorporation of 32P was 50-60% for the GPX probe and 30-40% for the CMV probe. Membranes were exposed to X-ray film for 6-72 h, and autoradiograms were scanned using a LKB 2222 UltraScan XL Laser Densitometer (Pharmacia LKB Biotechnology, Uppsala, Sweden). Peak areas were quantified using the GelScan XL program. For reprobing, blots were stripped with 0.1× SSPE, 0.1% SDS (18 mmol/L NaCl, 1 mmol/L sodium phosphate, 0.1 mmol/L EDTA, 1 g SDS/L) for 15 min at 100°C and 10 min at 25°C and baked at 80°C for 1-2 h. The level of 18S rRNA was determined using a 1.4-kb BamHI genomic DNA fragment for human 18S rRNA (Erickson et al. 1981) as described previously (Hesketh et al. 1994). The GPX mRNA levels were normalized to the signal for 18S rRNA within each sample.
Statistical analysis.
For each transfectant examined, GPX activity values for three replicate plates at each Se level were subjected to one-way ANOVA, and Duncan's multiple range analysis was used to determine the significance of treatment differences using a probability level of P < 0.05 for significance (Steel and Torrie 1960). There were no significant differences in GPX activity levels among the 100, 200 and 400 nmol/L Se treatments for wild-type cells or for any of the transfectants. To determine the significance of the effect of the low Se basal medium on GPX activity, GPX activities in replicates at these three points on plateau were used as Se-adequate values (n = 9) and compared with GPX activities of three replicates grown in basal medium, using the unpaired t test at a probability level P < 0.02. For mRNA analysis, two replicate plates of cells were pooled for each sample, and RNA was isolated from that pool. To determine the significance of the decrease in mRNA levels for cells transfected with different constructs, paired t test analysis was first used to show that Se status had no effect on 18S rRNA levels. Levels of Se-deficient GPX mRNA relative to Se-adequate GPX mRNA for each transfectant were then subjected to one-way ANOVA, and Duncan's multiple range analysis (Steel and Torrie 1960) was used to identify significantly different means for the experimental treatments. Values in the text are means ± SEM.
|
Table 1. Relative glutathione peroxidase (GPX) activity levels in wild-type and transfected Chinese hamster ovary (CHO) cells, and relative effect of low Se basal medium1 |
|
Table 2. Relative effect of Se deficiency on 18S rRNA and glutathione peroxidase (GPX) mRNA levels in transfected Chinese hamster ovary (CHO) cells1 |
RESULTS
3UTR expression vector produced over 50 large G418-resistant colonies on each 10-cm plate, whereas CHO cells transfected with the calcium phosphate precipitate alone did not survive in selective media. Five single pRc/CMV colonies, 12 pRc/GPX colonies and 12 pRc/3UTR colonies were transferred to individual flasks and grown to confluency.
3UTR and pRc/CMV transfectants were selected for further study, and these cell lines were retrieved from liquid nitrogen storage (Table 1). Cells were plated in low Se basal medium (22 nmol/L Se) with a range of supplemental Se concentrations. We observed that thawed pRc/GPX transfectants continued to express elevated GPX activity, whereas GPX activity in pRc/CMV and pRc/3UTR transfectants was lower after freezing but not as low as wild-type levels. The pRc/GPX transfectant (pRc/GPX5) with the highest initial level of GPX activity continued to express the highest GPX activity after freezing.
3UTR and pRc/CMV transfectants (Fig. 2). In each case, GPX activity reached the plateau level by 100 nmol/L Se. Similar Se response curves were observed for other transfectants (data not shown) and for wild-type CHO cells. The plateau GPX activity was thus calculated as the average activity for cells grown in 100, 200 and 400 nmol/L Se. The plateau GPX activity for pRc/GPX5 was 410% of wild-type CHO levels. Other pRc/GPX transfectants had 159-255% of wild-type activity (Table 1). Cells transfected with the pRc/CMV vector also had moderately increased GPX activity (128-197%) compared with wild-type cells, but this activity was generally not as high as for pRc/GPX transfectants.
3UTR3. Each point represents the mean ± SEM for cell supernatants derived from three replicate plates of cells. Values are expressed relative to the Se-adequate plateau (average of 100, 200 and 400 nmol/L Se points) for wild-type cells, which was set at 100%. Plots with different letters have statistically different GPX activity plateau levels (P < 0.05). Glutathione peroxidase activity was not significantly different in cells without supplemental Se.
3UTR transfectants studied, GPX activity levels were similar to those for pRc/CMV transfectants, and Se regulation of GPX activity was similar to that of pRc/CMV transfectants (Fig. 2). Transfection with pRc/3UTR was not expected to increase GPX activity because the SECIS motif located in the GPX 3UTR is necessary for Se insertion and thus enzyme activity (Berry et al. 1991); only the endogenous GPX protein should contribute to the GPX activity detected in pRc/3UTR and pRc/CMV transfectants.
, Saedi et al. 1988). This 13S species was the only signal detected with the GPX probe in wild-type CHO cells and in CHO cells transfected with the pRc/CMV vector alone (Fig. 3A). The pRc/GPX transfectants had a second, 16S GPX mRNA species in addition to the endogenous 13S GPX mRNA (Fig. 3A). The 16S transcript was expected because approximately 100 nucleotides of vector sequences should precede the GPX insert on the 5 end of the transcript and 231 nucleotides encoding termination sequences from the bovine growth hormone (BGH) gene should be added to the 3 end of mRNAs transcribed from the CMV promoter (Fig. 1). Densitometry showed that endogenous GPX mRNA levels were twice as high in pRc/CMV2 transfectants as in wild-type cells, in agreement with the increased GPX activity expression of this cell line (Table 1). Thus, transcription of the endogenous GPX gene might be stimulated by the transfection process, illustrating the importance of control experiments that establish nonspecific changes in gene expression caused by the vector alone. Transfected 16S GPX mRNA levels in pRc/GPX transfectants were five- to 10-fold higher than the endogenous GPX mRNA levels in wild-type cells. As expected, the 14S GPX transcript in pRc/3UTR transfectants was slightly larger than the 13S endogenous GPX mRNA, again due to the 5 and 3 vector sequences. We took advantage of the 5 vector sequences to further establish that the 14S and 16S GPX transcripts were products of the pRc/CMV vector constructs. When Northern blots were stripped and reprobed with the 94-bp HindIII/Xbal probe specific for CMV transcripts (Fig. 1), this CMV probe detected a single 16S mRNA species in pRc/GPX transfectants and a single 14S mRNA species in pRc/3UTR transfectants (Fig. 3B). As expected, these species were not detected in wild-type CHO cells or in pRc/CMV transfectants. Figure 3B and Figure 4 show that steady-state levels of pRc/3UTR mRNA were 45-82% higher than full-length pRc/GPX mRNA, suggesting that GPX mRNA may be more stable when the 3UTR is removed.
3UTR transfectants (pRc/3UTR8 and pRc/3UTR12) grown in low Se basal medium (22 nmol/L Se) with a range of supplemental Se concentrations (0-13 nmol/L Se). Endogenous GPX mRNA levels, transfected 16S GPX mRNA levels, and transfected 14S GPX mRNA levels were determined as described in the legend to Figure 3. GPX mRNA levels were normalized to 18S rRNA levels in each sample, and the Se-adequate plateau for endogenous GPX mRNA (average of the three highest Se points in each pRc/GPX transfectant) was set at 100%.
Effect of selenium on endogenous and transfected glutathione peroxidase mRNA levels. We next investigated the effect of medium Se on endogenous and transfected GPX mRNA levels. In low Se basal medium containing 10% FBS, 16S GPX mRNA levels in three different pRc/GPX transfectants, pRc/GPX5, pRc/GPX8 and pRc/GPX9, averaged 66 ± 10% of Se-adequate levels (data not shown), whereas low Se basal medium had no effect on 18S rRNA levels (Table 2). The Se-deficient GPX mRNA levels, normalized to 18S rRNA signals, averaged 58 ± 8% of Se-adequate levels (Table 2). Simultaneous detection of both transfected and endogenous mRNAs using the GPX probe showed that Se regulation of 16S pRc/GPX mRNA paralleled the regulation of endogenous GPX mRNA (Fig. 4). With graded levels of supplemental Se (0-13 nmol/L Se), both pRc/GPX and endogenous GPX mRNA reached maximal plateau levels with 3-4 nmol/L added Se.
3UTR transfectants grown in low Se basal medium, pRc/3UTR3, pRc/3UTR8, and pRc/3UTR12, were not significantly different from the levels in transfectants grown in Se-adequate medium (Table 2). The 14S GPX mRNA levels also showed no distinct regulation by medium Se concentration (Fig. 4).
UTR again eliminated Se regulation, further showing that the GPX 3UTR is necessary for Se regulation.
DISCUSSION
), protein (Knight and Sunde 1987 and 1988) and mRNA levels (Lei et al. 1995, Saedi et al. 1988, Sunde 1994, Sunde et al. 1989, Weiss et al. 1996 and 1997). These experiments demonstrated that GPX activity and GPX mRNA levels reach a plateau with increasing levels of Se, suggesting that the Se regulatory mechanism becomes saturated at a plateau breakpoint (Weiss et al. 1996), similar to a titration endpoint. Thus, any level of Se at or beyond the plateau breakpoint for a given parameter would be an adequate level of Se for that parameter. The present experiments extend this model to cultured cells in which we can monitor regulation of both endogenous and transfected GPX expression.
, Weiss et al. 1996 and 1997). Thus, the CHO cell model affirms that maximal GPX mRNA levels occur at distinctly lower Se status than for maximal GPX activity, suggesting that these plateaus represent saturation of two distinct processes.
UTR is necessary for Se regulation, the GPX 3UTR was deleted from pRc/GPX. Transfection with pRc/3UTR failed to increase GPX activity above that of control transfections, supporting the findings by Berry et al. (1991 and 1993) that the GPX 3UTR is necessary for synthesis of an active selenoenzyme. Importantly, deletion of the GPX 3UTR also eliminated Se regulation of GPX mRNA levels. The pRc/3UTR mRNA was also expressed at levels up to threefold higher when compared with full-length pRc/GPX mRNA levels, suggesting that the GPX 3UTR contains sequences that might function as a Se-responsive element.
3UTR mRNA levels were unresponsive to Se, additional experiments were conducted with 1% FBS medium and thus lower total medium Se. With this approach, endogenous GPX mRNA levels in Se-deficient transfectants dropped to 27-38% of Se-adequate levels and yet pRc/3UTR mRNA levels continued to be unaffected by Se deficiency. It must be noted, however, that pRc/GPX mRNA levels fell to 63-76% of Se-adequate levels in the 1% FBS model, similar to Se regulation in 10% FBS. The BGH termination and polyadenylation sequences added to mRNA by the pRc/CMV vector are intended to stabilize the CMV transcripts. This added stabilization may partially override the destabilization of GPX mRNA caused by Se deficiency. Alternatively, the strong CMV promoter may rapidly replace pRc/GPX mRNA lost by degradation during Se deficiency when compared with the wild-type GPX promoter. Either way, pRc/GPX mRNA levels are consistently lower in Se-deficient CHO cells than in Se-adequate CHO cells, showing that the Se-specific regulation mechanism is effective on both endogenous and pRc/GPX mRNAs.
UTR to insert selenocysteine at the position specified by a UGA codon (Berry et al. 1991 and 1993, Shen et al. 1993). It is unlikely, however, that the SECIS and the mechanism for selenocysteine insertion during translation can fully explain Se regulation of GPX mRNA levels, because mRNA levels for selenoproteins other than GPX are resistant to Se deficiency or decrease only gradually in Se deficiency (Hill et al. 1992, Sunde 1994, Weiss et al. 1997). For example, in Se-deficient rats, mRNA levels for phospholipid hydroperoxide glutathione peroxidase (PHGPX) (GPX4), a second cytosolic Se-dependent glutathione peroxidase (Ursini et al. 1985), do not decrease significantly in Se deficiency, whereas classical GPX mRNA levels decrease to <10% of Se-adequate levels within the same samples (Lei et al. 1995, Sunde et al. 1993). This dramatic regulation of classical GPX by Se could hypothetically be explained by the presence of a unique site, separate from the conserved SECIS motif, which targets GPX mRNA for degradation when Se is scarce.
UTR contains several iron-responsive elements (IRE) that are necessary for iron regulation of transferrin receptor mRNA levels (Casey et al. 1989). This regulation is mediated by an IRE-binding protein, now called iron regulatory protein (IRP), which has been shown to be identical to cytosolic aconitase (Kaptain et al. 1991). The IRP binding to the 3UTR stabilizes the transferrin receptor mRNA, possibly by blocking access to a specific endonuclease site (Binder et al. 1994).
), a selenocysteinyl-tRNA (Hatfield et al. 1991) and the SECIS mRNA stem-loop structure (Berry et al. 1993), is formed to mediate selenocysteine incorporation during translation. When Se is adequate, the formation of this Se-insertion complex could mask a unique endonucleolytic cleavage site present only in GPX mRNA, resulting in increased GPX mRNA levels.
UTR is necessary for Se regulation of GPX mRNA levels as well as for Se insertion. Dissection of the GPX 3UTR will be necessary to further understand the specific sequence requirements that enable GPX mRNA to respond to Se status. We hypothesize that the SECIS stem-loop within the GPX 3UTR might provide the recognition site for a Se-responsive regulatory factor that stabilizes the GPX mRNA. This regulatory factor might "sense" intracellular Se status, by having specific affinity for a Se species, which then causes a conformational change to enable GPX mRNA binding. The affinity of this regulatory factor for Se would result in a common plateau breakpoint for both transfected and endogenous GPX mRNAs. The identification of a unique sequence that targets GPX mRNA for degradation in Se deficiency would explain the distinctive Se regulation of GPX mRNA levels and would lead to a better understanding of GPX expression in relation to the expression of other selenoproteins.
ACKNOWLEDGMENTS
FOOTNOTES
3UTR, deleted 3UTR GPX expression vector; pRc/GPX, full-length GPX expression vector; SECIS, selenocysteine insertion sequence; 3UTR, 3 untranslated region.
Manuscript received 26 November 1996. Initial reviews completed 10 January 1997. Revision accepted 11 March 1997.
LITERATURE CITED
untranslated region.
Nature (Lond.)
1991;
353:273-276
[Medline]
UTR and does not involve poly(A) tail shortening.
EMBO J.
1994;
13:1969-1980
[Medline]
untranslated region of the mRNA.
EMBO J.
1989;
8:3693-3699
[Medline]
untranslated region.
Biochem J.
1994;
298:143-148-flanking sequence of the human cytosolic selenium-dependent glutathione peroxidase gene (hgpx1).
J. Biol. Chem.
1992;
267:5949-5958-untranslated region of the human cellular glutathione peroxidase gene are necessary and sufficient for selenocysteine incorporation at the UGA codon.
J. Biol. Chem.
1993;
268:11463-11469
structure, regulation and function. In: Selenium in Biology and Human Health (Burk, R. F., ed.), pp. 45-77. Springer-Verlag, New York, NY.This article has been cited by other articles:
|
|
 |
|
  P. R. Hoffmann, S. C. Hoge, P.-A. Li, F. W. Hoffmann, A. C. Hashimoto, and M. J. Berry The selenoproteome exhibits widely varying, tissue-specific dependence on selenoprotein P for selenium supply Nucleic Acids Res., June 9, 2007; 35(12): 3963 - 3973. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 R. R. Jameson, B. A. Carlson, M. Butz, K. Esser, D. L. Hatfield, and A. M. Diamond Selenium Influences the Turnover of Selenocysteine tRNA[Ser]Sec in Chinese Hamster Ovary Cells J. Nutr., July 1, 2002; 132(7): 1830 - 1835. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 X. Sun, X. Li, P. M. Moriarty, T. Henics, J. P. LaDuca, and L. E. Maquat Nonsense-mediated Decay of mRNA for the Selenoprotein Phospholipid Hydroperoxide Glutathione Peroxidase Is Detectable in Cultured Cells but Masked or Inhibited in Rat Tissues Mol. Biol. Cell, April 1, 2001; 12(4): 1009 - 1017. [Abstract] [Full Text]
|
|
|
|
|
|
S. D. Clarke and S. K. Kim Molecular Methodologies in Nutrition Research J. Nutr., November 1, 1998; 128(11): 2036 - 2037. [Full Text]
|
|
|
|
 |
|
 W. Wen, S. L. Weiss, and R. A. Sunde UGA Codon Position Affects the Efficiency of Selenocysteine Incorporation into Glutathione Peroxidase-1 J. Biol. Chem., October 23, 1998; 273(43): 28533 - 28541. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 P. M. Moriarty, C. C. Reddy, and L. E. Maquat Selenium Deficiency Reduces the Abundance of mRNA for Se-Dependent Glutathione Peroxidase 1 by a UGA-Dependent Mechanism Likely To Be Nonsense Codon-Mediated Decay of Cytoplasmic mRNA Mol. Cell. Biol., May 1, 1998; 18(5): 2932 - 2939. [Abstract] [Full Text]
|
|
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
