The differential expression of glutathione peroxidase ...

2 downloads 5 Views 1017KB Size Report
May 1, 2012 - expression of Gpx1 and Gpx4. We showed that, upon selenium fluctuation, the modulation of UGA recoding efficiency depends on the nature of ...

RNA Biology

ISSN: 1547-6286 (Print) 1555-8584 (Online) Journal homepage: http://www.tandfonline.com/loi/krnb20

The differential expression of glutathione peroxidase 1 and 4 depends on the nature of the SECIS element Lynda Latrèche, Stéphane Duhieu, Zahia Touat-Hamici, Olivier Jean-Jean & Laurent Chavatte To cite this article: Lynda Latrèche, Stéphane Duhieu, Zahia Touat-Hamici, Olivier Jean-Jean & Laurent Chavatte (2012) The differential expression of glutathione peroxidase 1 and 4 depends on the nature of the SECIS element, RNA Biology, 9:5, 681-690, DOI: 10.4161/rna.20147 To link to this article: https://doi.org/10.4161/rna.20147

View supplementary material

Published online: 01 May 2012.

Submit your article to this journal

Article views: 411

View related articles

Citing articles: 20 View citing articles

Full Terms & Conditions of access and use can be found at http://www.tandfonline.com/action/journalInformation?journalCode=krnb20

research paper

Research paper

RNA Biology 9:5, 681-690; May 2012; © 2012 Landes Bioscience

The differential expression of glutathione peroxidase 1 and 4 depends on the nature of the SECIS element Lynda Latrèche,1,2,† Stéphane Duhieu,1,† Zahia Touat-Hamici,1,† Olivier Jean-Jean2 and Laurent Chavatte1,* Centre de recherche de Gif-sur-Yvette; FRC 3115; Centre de Génétique Moléculaire; CNRS, UPR3404; Gif-sur-Yvette, France; 2UPMC Univ Paris 06; CNRS-FRE 3402; Biologie de l’ARN; Paris, France

1

These authors contributed equally to this work.



Key words: selenium, selenoprotein, UGA recoding, EFsec, SBP2, SECIS, glutathione peroxidase Abbreviations: DIO, iodothyronine deiodinases; EFsec, selenocysteine-specific elongation factor; Gpx1, glutathione peroxidase 1; Gpx4, glutathione peroxidase 4; NMD, nonsense mediated decay; SBP2, SECIS-binding protein 2; SECIS, selenocysteine insertion sequence; shRNA, small hairpin RNA; TR, thioredoxin reductase; UTR, untranslated region

© 2012 Landes Bioscience. Selenocysteine insertion into selenoproteins involves the translational recoding of UGA stop codons. In mammals, selenoprotein expression further depends on selenium availability, which has been particularly described for glutathione peroxidase 1 and 4 (Gpx1 and Gpx4). The SECIS element located in the 3’ UTR of the selenoprotein mRNAs is a modulator of UGA recoding efficiency in adequate selenium conditions. One of the current models for the UGA recoding mechanism proposes that the SECIS binds SECIS-binding protein 2 (SBP2), which then recruits a selenocysteine-specific elongation factor (EFsec) and tRNASec to the ribosome, where L30 acts as an anchor. The involvement of the SECIS in modulation of UGA recoding activity was investigated, together with SBP2 and EFsec, in Hek293 cells cultured with various selenium levels. Luciferase reporter constructs, in transiently or stably expressing cell lines, were used to analyze the differential expression of Gpx1 and Gpx4. We showed that, upon selenium fluctuation, the modulation of UGA recoding efficiency depends on the nature of the SECIS, with Gpx1 being more sensitive than Gpx4. Attenuation of SBP2 and EFsec levels by shRNAs confirmed that both factors are essential for efficient selenocysteine insertion. Strikingly, in a context of either EFsec or SBP2 attenuation, the decrease in UGA recoding efficiency is dependent on the nature of the SECIS, GPx1 being more sensitive. Finally, the profusion of selenium of the culture medium exacerbates the lack of factors involved in selenocysteine insertion.

Do not distribute. Introduction

Selenium is an essential trace element which is incorporated as selenocysteine, also defined as the 21st amino acid, into a small but essential family of proteins, the selenoproteins. Twentyfive selenoproteins have been identified in the human genome,1 some of them having essential functions in human health and disease.2-5 Many human selenoproteins are enzymes involved in oxidoreduction reactions where the selenocysteine residue plays a critical function. Among the selenoproteome, families of glutathione peroxidases (Gpx), thioredoxin reductases (TR) and iodothyronine deiodinases (DIO) have been thoroughly characterized. Selenoprotein synthesis follows an unusual mechanism which involves the translational recoding of a UGA codon, normally used as a stop signal, into a selenocysteine insertion signal.6-10 Cellular translation machinery uses UGA as a selenocysteine codon for selenoprotein mRNAs while

maintaining its use as a stop codon for all other cellular mRNAs. The recoding of UGA as selenocysteine is a critical regulatory stage of selenoprotein mRNA translation. A specific secondary structure present in the 3' UTR of selenoprotein mRNAs, named SECIS (SElenoCysteine Insertion Sequence) is necessary and sufficient to direct faithful UGA recoding. Recently, the SECIS elements were shown to have up to several thousand-fold range of recoding efficiencies in vivo and up to several hundredfold in an in vitro assay.11 Two categories of factors have been characterized as components of the selenocysteine insertion machinery. First, a class of tRNA Sec binding proteins is composed of EFsec (a dedicated translation elongation factor), SECp43, the phosphoserine tRNA kinase PSTK and a selenocysteine synthase.12-17 Then, a class of SECIS binding proteins have also been identified: SECIS binding protein 2 (SBP2), ribosomal protein L30 (rpL30), translation initiation factor 4A3 (eIF4A3) and nucleolin.18-21 However, despite many efforts, the precise

*Correspondence to: Laurent Chavatte; Email: [email protected] Submitted: 01/31/12; Revised: 03/20/12; Accepted: 03/26/12 http://dx.doi.org/10.4161/rna.20147 www.landesbioscience.com

RNA Biology

681

mechanism of selenoprotein synthesis and regulation remains elusive (reviewed in refs. 6–10). While the function of most proteins is attributed to selenocysteine insertion, several (such as L30 and eIF4A3) are involved in other cellular processes, such as ribosome biogenesis, mRNA splicing and translation regulation (L30), or are part of the exon junction complex (eIF4A3). Several reports from independent laboratories stipulate that SBP2, which is limiting in cells, is the central component among the recoding factors. Acting as a platform, SBP2 recruits EFsec complexed with the tRNA Sec and delivers it to the ribosome in which L30 is acting as an anchor.18,19,22-26 Most importantly, it has been proposed that SBP2 dictates selenocysteine insertion efficiency.22,26 The regulation of the selenoproteome is highly tuned according to selenium intake. Under conditions of selenium depletion, a regulatory mechanism that is still unclear, induces a prioritized response of the body, that maintains the production of essential seleno-enzymes at the expense of others.27-31 This differential expression mechanism has been particularly studied for Gpx1 and Gpx4, two members of the glutathione peroxidase (Gpx) selenoprotein family. It appeared from experiments performed in rodents that Gpx1 expression and activity were more affected than those of Gpx4 by dietary selenium intake.32 This regulation was mostly occurring at the translational level since Gpx1 and Gpx4 mRNA abundance was only weakly affected by moderate selenium deficiency. Recently characterized as a SECIS binding protein, eIF4A3 was found to be upregulated by selenium deficiency.20 Since the protein has been reported to selectively bind some SECIS elements to displace SBP2, the authors suggested that eIF4a3 is a selenium-regulated SECIS-binding protein that links selenium status to differential selenoprotein expression. This finding paves the way for new research directions to identify selenium-regulated factors, which can either potentiate or compete with selenocysteine insertion machinery. Among the trans-acting factors identified so far in the recoding process, SBP2 might be a central determinant together with the SECIS for UGA recoding efficiency. Here, we used Hek293 cell line and Gpx1 and Gpx4 as model selenoproteins to study the relative influence of SECIS element SBP2 and EFsec on the selective regulation of UGA recoding as a function of selenium availability. Using a set of validated luciferase based reporter constructs, we were able to quantitatively demonstrate that, in response to selenium level variations, selenocysteine insertion was driven by the nature of the SECIS element. Then, we showed that EFsec and SBP2 control the hierarchy of selenoprotein synthesis in a SECIS dependent manner.

found that the SECIS modulates UGA recoding efficiency,11 we elaborated different growth media with various selenium amounts mimicking adequate, deficient and supplemented dietary intake. These conditions are referred hereafter as unsupplemented (Unsup), depleted (Dpl) and supplemented (Sup), respectively. Growth medium (Dpl + Se) with a 30 nM selenium addition to the Dpl condition was used to validate the specificity toward selenium. As shown in Figure 1, in Hek293 cells, the protein levels of Gpx1 and Gpx4 detected by immunoblotting were highly dependent on selenium concentration in the growth medium. This was immediately observable after one passage of the cells. When comparing the protein levels between Unsup and Sup extracts, Gpx1 appeared approximately three times more sensitive to selenium supplementation than Gpx4 (Fig. 1). Our results obtained in Hek293 cells recapitulated what has been reported in other studies, either in rodents20,32 or in human.33 Interestingly, when comparing Dpl with Unsup extracts, the selenium deficiency led to a dramatic decrease of both Gpx1 and Gpx4 expression, although in the Dpl extract Gpx4 was still detectable while Gpx1 had almost disappeared. This effect uniquely depended on selenium concentration since cell growth in Dpl + Se media restored the expression of both Gpx1 and Gpx4 to the levels observed in Sup media. In summary, Gpx1 expression was regulated by selenium level with wider amplitude than Gpx4 (Fig. 1B), even though immunoblotting is considered a semi-quantitative technique. To study whether the expression of SBP2 and EFsec was modified by selenium level, we analyzed the different cellular extracts by immunoblotting (Fig. 1). When comparing the extracts from all growth conditions, we found that the levels of both SBP2 and EFsec were virtually insensitive to selenium concentration in the culture medium. Taken together, our data clearly show that the differential expression of Gpx1 vs. Gpx4 as a function of selenium concentration observed in Hek293 cells is comparable to that reported for animal models. Therefore, this cell line can be used to study selective regulation of selenoproteins at the molecular level. Selenoprotein expression is mostly regulated by selenium at the translational UGA recoding level. To precisely evaluate the influence of selenium supply at the translational level of UGA recoding, we used a previously validated reporter construct (Fig. 2A and reviewed in ref. 20). The luciferase open reading frame, which contains an in frame UGA codon (or a control UGU/Cys codon), was linked to the Gpx4 or Gpx1 SECIS element. The selected sequences were respectively 102 (Gpx4) and 91 (Gpx1) nucleotides in length, starting ~25 nucleotides before the highly conserved AUGA motif, which encompasses the minimum active domain for selenocysteine insertion.34-38 Secondary structures of Gpx1 and Gpx4 SECIS elements are illustrated in Figure 2B where the highly conserved and functional motifs are highlighted. Both sequences fold in a consensus stem-loop-stem-loop structure (Helix1-Internal loop-Helix2-Apical loop), an additional 3 base-pair helix being formed in the apical region of Gpx4. This difference is generally used to classify the SECIS as Type 1 (Gpx1) or Type 2 (Gpx4).36,38,39 As reported in previous studies, the Luc-SECIS

© 2012 Landes Bioscience. Do not distribute.

Results Gpx1 and Gpx4 are differentially regulated in Hek293 cells. Several reports have suggested that selenoprotein synthesis follows a hierarchy depending on selenium status. Most of these studies were performed on the Gpx members in rodent models upon different dietary selenium conditions.27-32 Strikingly Gpx1 expression was more sensitive to selenium variation than Gpx4, mostly via a translational regulation process that remains to be elucidated. For the culture of Hek293 cells, in which we

682

RNA Biology

Volume 9 Issue 5

© 2012 Landes Bioscience. Do not distribute. Figure 1. Differential regulation of selenoprotein expression as a function of selenium concentration. (A) Hek293 cells were grown in selenium depleted (Dpl), unsupplemented (Unsup) or supplemented (Sup and Dpl + Se) media. After 48 h growth, protein extracts were analyzed by immunoblotting using the indicated antibodies. (B) Relative expression of Gpx1 and Gpx4 over Tubulin were calculated from three independent experiments, and indicated as mean ± SD, the Unsup condition being set as 100%. The significant fold-change differences between Dpl and Dpl + Se or Unsup and Sup are indicated above the respective brackets.

reporter constructs we used allowed reliable quantification for Sec insertion in contrast with immunoblotting, since the enzymatic activity of luciferase can be measured linearly within a large dynamic range with a luminometer. Thus, these luciferase constructs (listed in Fig. 2A) were used to monitor the effect of selenium concentration on the recoding efficiency of Gpx4 and Gpx1 SECIS either by transient expression (Fig. 3A), or in stably expressing cell lines (Fig. 3B). For transient expression, luciferase activities obtained 48 h post-transfection of Luc UGA/gpx4 and Luc UGA/gpx1 constructs were found to be highly sensitive to the level of selenium in the growth media (black and gray bars in Fig. 3A, respectively). Interestingly, when comparing the luciferase activities obtained with Unsup and Sup extracts, recoding efficiency was stimulated 1.7- and 3.8-times for Luc UGA/gpx4 and Luc UGA/gpx1, respectively. Strikingly, this response to selenium was even higher when comparing the extracts from Dpl and Dpl + Se conditions. In the latter case, we observed a 9.5- and 24-fold increase in luciferase activities for Luc UGA/ gpx4 and Luc UGA/gpx1, respectively. The control experiment performed with Luc UGU/gpx4 construct in the different media

www.landesbioscience.com

showed virtually no variation (Fig. S1), indicating that the effect was directly linked to a modulation of the efficiency of UGA recoding as selenocysteine, and not to differences in mRNA expression or stability. As illustrated in Figure 3B, the results obtained with Hek293 cells stably expressing either the Luc UGA/gpx1 or Luc UGA/ gpx4 reporter were comparable to those obtained in Figure 3A in similar culture conditions. Indeed, in both cases, we observed a selective regulation of UGA recoding that is dependent on the nature of the SECIS element. Both experiments showed that Gpx1 SECIS recoding efficiency is 2- to 4-fold more sensitive to selenium concentration than that of Gpx4 SECIS. However, the response to selenium variation was markedly increased with stably expressing cell lines, in comparison with transient transfection experiments, indicating a higher sensitivity of our reporter construct in this experimental design. Since the copy number of the reporter gene is expected to be much lower in stably expressing than in transiently transfected cells, a possible saturation of the recoding machinery should be avoided in stably expressing cells. Levels of luciferase mRNAs were similar as a

RNA Biology

683

© 2012 Landes Bioscience. Do not distribute. Figure 2. Luciferase reporter assay for the measurement of UGA recoding efficiency. (A) Representation of the luciferase mRNAs expressed after transfection of the reporter constructs. The open reading frame of firefly luciferase (shown as light gray) with a UGA/selenocysteine codon at position 258 is linked to either Gpx1 (dark gray) or Gpx4 (black) SECIS elements, the secondary structures of which are represented in (B) (adapted from refs. 19 and 46). A control construct of the Luc UGA/gpx4 was made, where the UGA is replaced by a UGU/cysteine. Despite poor sequence conservation, the two SECIS fold in a consensus stem-loop-stem-loop structure. The SECIS Core and the AAR motif are shown in light gray.

function of selenium concentration (Fig. S2A). Taken together, our data from transient and stable expression of luciferase constructs demonstrated that the nature of the SECIS element modulates the response to selenium deficiency as well as that to selenium supplementation. Moreover our results indicated that this regulation occurs directly at the translational level when the elongating ribosome encounters an in frame UGA codon. By itself, the difference in UGA recoding efficiency of the two SECIS elements observed between either Unsup and Sup or Dpl and Dpl + Se media can account for the differential expression of selenoproteins Gpx1 and Gpx4 observed in Figure 1. Attenuation of EFsec and SBP2 leads to a downregulation of Gpx1 and Gpx4 selenoprotein expression. To evaluate the impact of attenuation of SBP2 and EFsec on selenoprotein synthesis in Hek293 cells, we designed specific shRNAs targeting either SBP2 or EFsec mRNAs. pSuper derivative plasmids expressing these shRNAs were transiently transfected

684

in Hek293 cells and the efficiency of EFsec and SBP2 attenuation was evaluated by immunoblotting (Fig. 4B and C). In both cases, shRNA treatment led to a dramatic decrease in the expression of the corresponding protein. Then, to measure the effects of SBP2 and EFsec downregulation on selenocysteine insertion, we performed co-transfection of the shRNA expressing plasmids with the Luc UGA/gpx4 and Luc UGU/gpx4 reporter constructs, and evaluated the cellular extracts for luciferase activities (Fig. 4A). In comparison to cells transfected with either pSuper or control shRNA expressing plasmid, the expression of shRNAs targeting either EFsec or SBP2 led to a strong decrease in luciferase activities (~50% and ~75%, respectively) when Luc UGA/gpx4 construct was co-transfected (Fig. 4A and black bars). In contrast, co-transfection experiments with Luc UGU/gpx4 construct showed no significant modification of the luciferase activities whatever the shRNA used (Fig. 4A and white bars). Ultimately,

RNA Biology

Volume 9 Issue 5

our data indicated that, when depleted, SBP2 and EFsec become both limiting for efficient UGA recoding in a luciferase construct containing a Gpx4 SECIS element. Our data also suggest that, although limiting, low levels of recoding factors are sufficient for significant UGA recoding. Specific shRNAs were also designed to knock-down the expression of ribosomal protein L30. This protein is an essential component of the ribosome with additional functions in several biological processes, including pre-rRNA processing, inhibition of pre-mRNA splicing and selenocysteine insertion.19,40,41 L30 is also involved in the regulation of its own mRNA translation.41 In yeast, L30 is an essential protein even though selenoproteins are not present in this organism.40 In our hands, expression of shRNAs targeting L30 mRNA had dramatic effect on cell viability soon after transfection (data not shown) preventing the analysis of the effect of L30 attenuation on UGA recoding efficiency. This effect of L30 attenuation was likely due to the impairment of L30 essential functions in mammalian cells. Attenuation of recoding factors has a SECIS-dependent influence on UGA recoding efficiency. To quantitatively evaluate the selective alteration of UGA recoding efficiency during attenuation of EFsec or SBP2, we used the cell lines that stably expressed Luc UGA/gpx1 or Luc UGA/gpx4. As illustrated in Figure 5, the effect of recoding factor attenuation was measured in Unsup (A), Sup (B) or Dpl (C) media. In all cases, luciferase activities were expressed relative to the activity of the pSuper transfection, which was set as 100%. No difference in activities was ever detected between pSuper and sh Ctrl extracts. Importantly, in all growth conditions, we observed a clear decrease in UGA recoding efficiency with either sh EFsec or sh SBP2 treatments. Interestingly, for unsupplemented and supplemented media, the decrease in UGA recoding efficiency was significantly (p value < 0.001) more pronounced for the Luc UGA/gpx1 (gray bars) construct than for Luc UGA/gpx4 (black bars). This result reinforces the ideas that (1) Gpx1 and Gpx4 expression are differentially regulated, Gpx1 being more sensitive to changes affecting the recoding mechanism than Gpx4 and (2) that the SECIS plays a major role in this regulation. Luciferase mRNA levels were not dramatically affected by shRNA treatments (Fig. S2B), suggesting the reporter constructs were not targeted by nonsense mediated decay (NMD). In addition, in a context of either EFsec or SBP2 attenuation, the decrease in UGA recoding efficiency is more marked in the supplemented medium than in the unsupplemented or depleted media (compare Fig. 5A and B with C), indicating that the profusion of selenium exacerbates the lack of the factors involved in the UGA recoding process. This suggests that, at high selenium concentration, when depleted, either SBP2 or EFsec availability are limiting for UGA recoding efficiency. Conversely, the results obtained with selenium-depleted medium suggest that, at low levels of selenium, the effects of EFsec and SBP2 attenuation were masked by the lack of selenium availability which became the limiting factor for UGA recoding. This could also explain the absence of differential regulation between Gpx1 and Gpx4 SECIS in the depleted medium.

© 2012 Landes Bioscience. Do not distribute.

www.landesbioscience.com

Figure 3. The UGA recoding efficiency as selenocysteine is differentially regulated by the selenium level of the culture medium. Luciferase based reporter constructs were used in transiently (A) or stably (B) expressing cells. After 48 h, enzymatic activities were measured on cellular protein extracts. Transfection efficiencies were normalized by calculating the ratio of luciferase and β-galactosidase activities in transient transfection (A). In Hek293 cells stably expressing the reporter constructs (B), luciferase enzymatic activity was normalized by protein concentration of the cell lysate. To compare the UGA recoding efficiency in (A and B), data from three independent experiments were arbitrarily expressed relative to the activity obtained in extracts from the unsupplemented media with the respective luciferase construct, which was set as 100%. The data are represented as mean ± SD. The significant fold-change differences between Dpl and Dpl + Se or Unsup and Sup are indicated above the respective brackets.

Discussion Hek293 cells provide a good model to study selective regulation of selenoproteins. Identifying cellular models that reproduce the selenium-dependent hierarchy of selenoprotein expression observed in animals is of crucial importance to delineate the molecular components and mechanisms regulating the mammalian selenoproteome. Most of the work on selective expression of selenoproteins has been made in rodents focusing on two members of the Gpx family, Gpx1 and Gpx4, at a time when the selenoproteome and the selenocysteine insertion

RNA Biology

685

components of the recoding factors, SBP2 and EFsec, led to a decrease of selenocysteine insertion that is dependent on the nature of the SECIS element, Gpx1 SECIS being more sensitive than Gpx4. Therefore, this experimental design will constitute an excellent model for the validation of factors involved in up or downregulation of the selenocysteine insertion provided these factors are not essential for cell growth. The SECIS elements are regulatory elements of UGA recoding efficiency. A comprehensive analysis of the SECIS elements from the human genome has found them highly variable in UGA recoding efficiency both in vivo and in vitro.11 Several thousand-fold difference in recoding efficiency was measured between the strongest and weakest SECIS elements. Structure-function analysis showed that the region surrounding the k-turn motif of the SECIS was mainly responsible for the UGA recoding efficiency. In addition to modulating the level of UGA recoding efficiency in adequate selenium conditions, we have herein determined that the nature of the SECIS element is also important for regulation of the hierarchy of selenium use by selenoproteins. Whether the binding constants of the different SECIS binding proteins for the large set of mRNA targets correlate with the hierarchy of UGA recoding is still an open issue. In vitro binding experiments with a limited set of SECIS elements (of variable size) showed that recombinant SBP2 had different RNA affinities.11,42-45 In these reports a 2-fold difference in SBP2-SECIS binding ability was observed between Gpx1 and Gpx4 SECIS.45 In vivo, it has been reported from pull down experiments that SBP2 exhibits widely differing levels of binding to selenoprotein mRNAs.26 Out of the 18 selenoprotein mRNAs detectable in MSTO-211H cells, the authors found efficient binding of SBP2 for the following partners: SelW, Sel15, Gpx4, SelH, TR1, Dio2, SelK, SelS and Sps2 mRNAs. When focusing on Gpx members, it appeared that SBP2 pull down ability was approximately 60-fold higher for Gpx4 than for Gpx1, even though the abundance of both mRNAs was similar.26 The variability in the strength of SBP2-SECIS interaction obtained in vivo and in vitro can be interpreted in different ways in light of recent findings. First the abundance of SBP2 relative to selenoprotein mRNAs in vivo might be far less than a 1:1 ratio. Therefore a slight difference in affinity might result in wide amplitude in the repartition of the various SBP2-mRNA complexes, particularly since 26 SBP2SECIS interactions are theoretically possible. Then, adding layers of regulation, other competing or auxiliary factors can selectively alter SBP2-SECIS binding. Thus, eIF4A3 selectively prevents SBP2-SECIS complex formation for a subset of selenoprotein mRNA including Gpx1, SelN and Dio1, but not for others such as Gpx4, SelT and SelW.20,46 Therefore, eIF4A3 selectively downregulates UGA recoding for some non-essential genes in selenium depletion conditions. Since eIF4A3 is downregulated in selenium-supplemented growth conditions, this protein is considered as a transcript-selective translational repressor of selenoprotein synthesis during selenium deficiency.20,46 Whether eIF4A3 has other selenoprotein mRNA targets and whether other competing or auxiliary factors playing a role in selenocysteine insertion exist remain to be investigated. Nevertheless, taking into account the reported literature and the results presented

© 2012 Landes Bioscience. Figure 4. Attenuation of SBP2 and EFsec leads to a specific decrease of UGA recoding efficiency. (A) Hek293 cells grown in Sup media were co-transfected with luciferase/SECIS constructs, β-galactosidase and shRNA containing plasmids targeting either EFsec, SBP2, a control sequence or an empty vector (pSuper). Three days post-transfection the cells were harvested in lysis buffer and the protein extracts were analyzed for luciferase and β-galactosidase activities. Normalized luciferase activities obtained with Luc UGA/gpx4 and Luc UGU/gpx4 were expressed relative to the activity obtained in extracts from the pSuper control transfection, which was set as 100%. Statistically significant differences with the pSuper condition are indicated by an asterisk. The efficiency of the shRNA treatments targeting SBP2 (B) or EFsec (C) was evaluated by immunoblotting as described in Materials and Methods.

Do not distribute.

machinery were only partially characterized.27-32 The SECIS element had been identified as a regulator of UGA recoding but not completely characterized and neither EFsec nor SBP2 were known. It was observed that Gpx1 protein expression was more sensitive to selenium fluctuation than Gpx4. Here, the use of the luciferase-SECIS reporter construct allowed a quantitative analysis of UGA recoding. Since our reporter constructs are designed without introns, a change in luciferase activity should only reflect alteration of UGA recoding efficiency. The reporter constructs allowed us to observe the SECIS dependent alteration of selenocysteine insertion efficiency upon selenium deficiency or supplementation with either transiently or stably expressing cells. Interestingly, we noticed that stable cell lines were more sensitive to modification of the selenium level. This was probably due to the low abundance of UGA recoding machinery which can be easily saturated or excess of luciferase in the transient transfection experiments. From this perspective, cell lines stably expressing the luciferase-SECIS constructs can recapitulate physiological regulation occurring in animal tissues and can be used to study the SECIS-dependent modulation of UGA recoding as selenocysteine. Indeed, we found that knock-down of two major

686

RNA Biology

Volume 9 Issue 5

here, it becomes clear that the SECIS elements together with their interacting partners are essential for the selective regulation of UGA recoding efficiency, which therefore modulates the hierarchy of selenoprotein expression upon selenium fluctuation. A careful characterization of the different complexes involving the SECIS elements should be performed to understand the dynamics of RNA-protein interactions leading to the alteration of UGA recoding efficiency. Downregulation of SBP2 and EFsec influence the hierarchy of selenoprotein synthesis. It has been shown that selenoprotein mRNP assembly is controlled by the Hsp90 chaperone machinery, using a similar set of factors as for snoRNP and snRNP assembly.47 Hence, Hsp90 chaperone in concert with human R2TP (hSpagh, hPih1, hRvb1 and hRvb2) and Nufip cochaperones are critical factors for efficient SBP2-mRNA complex formation. Indeed, the inhibition of Hsp90 by geldanamycin prevented the production of selenoprotein mRNPs and destabilized SBP2, most probably because of the inherent instability of the mRNP proteins. Presence of EFsec in the mRNPs has not been investigated here, but a previous work from Berry’s laboratory has reported a stable interaction between SBP2 and EFsec that was enhanced in vivo by SECp43, a tRNA Sec binding protein.16 In our present work, we found that reduced expression of either SBP2 or EFsec led to similar effect on UGA recoding efficiency. Most strikingly, the SECIS-selective decrease in luciferase activity is observed in both cases. This selective regulation is particularly visible in media containing adequate and supplemented levels of selenium. Our data clearly indicate that at high selenium levels, when depleted, SBP2 and EFsec become limiting factors in the UGA recoding process. Conversely, under conditions of selenium deficiency, other factors may become limiting and thus regulate the selective synthesis of selenoproteins. We anticipate that the SECIS element is responsible for the hierarchy of selenoprotein expression at the translational level by determining the efficiency of SBP2 and EFsec recruitment. Thus, the alteration of any partner of mRNP assembly would differentially affect selenoprotein expression. The selenium regulation hierarchy is at the crossroads of different translational mechanisms. Given the dual meaning of the UGA codon as a stop and as a selenocysteine codon in selenoprotein mRNAs, selenium availability regulates the hierarchy of selenoprotein expression via a complex network including (1) translation termination efficiency (2) decoding ability by the tRNA Sec and (3) recoding efficiency by the SECIS element. First, the action of release factors leads to translation termination, which for several selenoprotein mRNAs initiates NMD. This surveillance pathway is used by the cell to detect and degrade aberrant mRNAs with premature stop codons. In selenium deficient mice, Gpx1, SelH and SelW mRNA levels markedly decrease to < 40% of selenium adequate levels.48,49 Some other selenoprotein mRNAs, including TR1, TR2, SelP, SelK, SelM and Gpx3 were only moderately affected by low selenium conditions, while the majority of selenoprotein mRNA levels remained unaffected by selenium deficiency or supplementation.49 Another possible layer of regulation, that is not yet completely elucidated, is the mechanism of UGA decoding by the dedicated tRNA Sec. Indeed, the tRNA Sec is initially aminoacylated by a

© 2012 Landes Bioscience. Do not distribute.

www.landesbioscience.com

Figure 5. Effect of SBP2 and EFsec downregulation on UGA recoding efficiency as a function of selenium concentration. Hek293 cells stably expressing either Luc UGA/gpx4 or Luc UGA/gpx1 were grown in various selenium conditions 24 h before the transfection of plasmids expressing shRNAs (EFsec, SBP2 or Ctrl) or an empty vector (pSuper). After three days, cells were harvested and analyzed for luciferase activity. Normalization of enzymatic activity was made by calculating the ratio between luciferase activity relative to protein concentration. The UGA recoding efficiency was expressed relative to the activity obtained in extracts from pSuper transfection for each cell line, which was set as 100%. Data from three independent experiments are represented as mean ± SD. Statistically significant differences (p < 0.001) between sh SBP2 and sh EFsec treatments are indicated by an asterisk above the corresponding bracket.

serine which is subsequently converted into selenocysteine by the selenophosphate synthetase. Selenium levels regulate the ratio between the serine and selenocysteine tRNA Sec isoforms. Additionally, the 2'-O-methylation modification at the wobble

RNA Biology

687

position (U34) of the anticodon of tRNA Sec, yielding a unique mcm5Um, is stimulated by selenium in rodents and is related to specific selenoprotein gene expression.50 Evidence from transgenic mice, where the tRNA Sec gene was replaced with either wild type or mutant tRNA Sec transgenes, indicates that the expression of a specific subset of selenoproteins that includes Gpx1, Gpx3, SelR and SelT (but not Gpx4, TR1 and TR3) strictly require this highly specialized methyl group in tRNA Sec. The targets of this modification at the level of the selenoproteome remains to be delineated. In summary, selenium level not only regulates the absolute level of tRNA Sec but also its isoforms. And finally, we found that the nature of the SECIS element, which is a strong determinant for UGA recoding efficiency,11 conditions the differential response of selenocysteine insertion to selenium and recoding factor availability. An intricate combination of these different mechanisms probably leads to hierarchic expression of selenoproteins in different tissues with selective alteration by selenium availability. Materials and Methods

100 UI/ml penicillin, 1 mM sodium pyruvate and 2 mM L-glutamine. Media and supplements were purchased from Life Technologies. Cells were cultivated in 5% CO2 at 37°C and humidified atmosphere. Cells were grown 48 or 72 h (as indicated) in media with different concentrations of selenium, which mimic the normal, depleted and supplemented conditions. In the reference (Unsup) media, the selenium is provided by the addition of 10% (v/v) FCS. Selenium concentration in the unsupplemented media was comprised between 5 and 15 nM corresponding to adequate selenium supply in mammal diet. The selenium deficient media (Dpl), elaborated as described in references 29 and 54, was composed of D-MEM supplemented with 2% FCS, 100 μg/ ml streptomycin, 100 UI/ml penicillin, 1 mM sodium pyruvate, 2 mM L-glutamine, 5 mg/L transferrin, 10 mg/L insulin, 100 pM 3,5,3'-triiodothyronine (T3) and 50 nM hydrocortisone. In the Dpl media, the selenium concentration was 5-fold less than that of the Unsup media since 2% FCS was used instead of 10%. Supplemented media were elaborated by an addition of 30 nM sodium selenite to either the Unsup or the Dpl media, and were respectively referred to as Sup and Dpl + Se media. To establish stable cell lines expressing Luc UGA/gpx1 or Luc UGA/gpx4 reporter constructs, 2 μg of the pDNA3.1 corresponding plasmids were transfected in Hek293 cells using calcium phosphate precipitation method. After overnight incubation, the cells were replated by diluting them 1:10 in complete D-MEM. Clones were selected and amplified for 3 weeks in complete D-MEM containing 1 mg/ml G418 (Roche). Ten (Luc UGA/gpx4) and seven (Luc UGA/gpx1) positive clones were isolated, tested for luciferase activity and yielded a 4-fold difference between the lowest and highest activities in both cases. We verified for several clones that the response to selenium concentration was similar for high- and low-activity clones. Then we selected one clone either for Luc UGA/gpx1 or Luc UGA/gpx4 that was in the average of luciferase activity to perform subsequent experiments. Stable cell lines were propagated in the absence of G418 selection for about 30 passages without observable loss of luciferase activity. Transient transfection. Calcium phosphate precipitation method was performed to transiently transfect the Hek293 cells according to reference 11. Cells were split the day prior to transfection. Media were changed 24 h post-transfection and cells collected at 48 or 72 h (as indicated) with 300 μL of lysis buffer (25 mM Tris-phosphate pH 7.8, 2 mM DTT, 2 mM EDTA, 1% Triton x100, 10% glycerol). In 100 mm plates, 2 μg of luciferase constructs, 2 μg of β-galactosidase and 20 μg for shRNA containing plasmids were transfected as indicated in the respective figures. When necessary, empty pSUPER or pcDNA3.1 vectors (Oligoengine and Life Technologies, respectively) were added to obtain similar amount of transfected DNA. Cell extracts were assayed for luciferase and β-galactosidase activities by chemiluminescence (Promega Luciferase and Beta-Glo assay systems, respectively), in triplicate using a microplate reader FLUOstar OPTIMA (BMG Labtech). The luminescence detection was linear in a 4 to 5 log scale. In our experimental conditions, luciferase measurements were systematically at least a hundred fold higher than in the

© 2012 Landes Bioscience. DNA constructs. The UGA recoding efficiency was analyzed using previously developed and validated luciferase reporter constructs described in reference 20, and reference therein. In summary, sequences from rat Gpx1 and Gpx4 UTRs corresponding to the minimal active domain for selenocysteine insertion were cloned downstream of a luciferase coding sequence which has been modified to contain an in frame UGA codon at position 258 (constructs Luc UGA/gpx1 and Luc UGA/gpx4, see Fig. 2). As a control, we used an equivalent construct with a UGU codon instead, and therefore named Luc UGU/gpx4. With the Luc UGA/gpx1 and Luc UGA/gpx4 constructs that have been widely used11,19,20,43,51 and validated for selenocysteine incorporation in transfected cells,11,19,20,43,51 an active luciferase is made only when the UGA is recoded as a selenocysteine codon. For SBP2, EFsec and L30 downregulation, 19-nucleotide-long sequences were selected within their respective mRNAs to express shRNA after transient transfection of a pSuper expression system as described in reference 52 and 53. shRNAs targeting either SBP2 (5'-AAG GGT GAA ATA GTG GTG AAT T-3', position 799–821), EFsec (5'-AAG AGA CAG GCA GCA ATT GAT T-3', position 463–485) or L30 (5'-AGA TGA TCA GAC AAG GCA AAT C-3', position 106–128) were referred as sh SBP2, sh EFsec or sh L30, respectively. The control shRNA (5'-TCT GTG ATC GTA CCC TCA G-3') referred as sh Ctrl was previously shown to target a protein (eRF3b) which is not expressed in Hek293 cells and to have no effect on stop codon readthrough.52 Pairs of oligonucleotides containing shRNA target sequence in sense and reverse orientations with cohesive BglII and HindIII ends were annealed and inserted in the pSuper vector linearized by BglII and HindIII. Cell culture and transfection. Hek293 cells (Life Technologies) were grown and maintained in 100 mm plates in Dulbecco’s Modified Eagle Medium (D-MEM) supplemented with 10% fetal calf serum (FCS), 100 μg/ml streptomycin,

Do not distribute.

688

RNA Biology

Volume 9 Issue 5

untransfected extract. Protein concentrations were measured using the DC kit (Biorad) in microplate assays. Immunoblotting. Immunoblotting was performed using BisTris NuPAGE Novex Midi Gels (Life Technologies). Proteins were separated using MOPS SDS running buffer (Life Technologies). After SDS-PAGE, proteins were transferred to Nitrocellulose membrane by western blotting using iBlot® Dry blotting System (Life Technologies). The membranes were blocked with 2% of ECL Advance Blocking Reagent (GE Healthcare) and washed in TBS-0.1% Tween-20. Membranes were probed with primary antibodies directed against either Glutathione Peroxidase 1 (Enzo Life Sciences) 1/1,000 or Glutathione Peroxidase 4 (Ab Frontier) 1/1,000 or SBP2 (rabbit serum generously provided by Laura Papp) 1/1,000 or EFsec (rabbit serum produced by Covalab, see below) 1/2,000. Anti-EFsec antibodies were obtained by immunization of female rabbits with three different peptides corresponding to human protein (AFD KQP QSR ERG ITL, SFN FSQ EYL FQE QY, SLT FKR YVF DTH KR). Membranes were stripped and reprobed with anti-α tubulin antibody (Sigma-Aldrich) 1/10,000. HRP-conjugated Rabbit or Mouse antibodies (Sigma-Aldrich) were used to detect the primary antibodies. Chemiluminescence detection was performed with ECL Advance western Blotting Detection Kit (GE Healthcare) using a LAS 3000 CCD camera (GE Healthcare). RNA extraction and RT-qPCR. Total RNAs were cell extracted using Nucleospin RNA II kit (Macherey Nagel). Concentration and quality of extracted RNAs were determined using OD260/280 with a NanoVue Spectrophotometer (GE Healthcare). Synthesis of cDNA was performed using Transcriptor High Fidelity cDNA Synthesis kit (Roche Applied Science)

according to the manufacturer’s instructions. Quantitative PCR was performed in triplicate using LightCycler® 480 SYBR Green I Master (Roche Applied Science). PCR program was 95°C for 5 min, and 45 cycles of 95°C for 10 sec, 60°C for 20 sec, 72°C for 20 sec. Data were analyzed using LightCycler® 480 software and normalized relative to Gapdh. The primers used were as follows: Luc fw, GGA AAG ACG ATG ACG GAA A; Luc rev, TCG CGG TTG TTA CTT GAC TG; Gapdh fw, AGC CAC ATC GCT CAG ACA C; Gapdh rev, GCC CAA TAC GAC CAA ATC C. Disclosure of Potential Conflicts of Interest

No potential conflicts of interest were disclosed. Acknowledgements

This work was supported by the CNRS (ATIP program to L.C.), the Fondation pour la Recherche Médicale (L.C.), the Ligue Contre le Cancer (Comité de l’Essonne, L.C.), the Association pour la recherche sur le cancer [grants numbers 4.849 to L.C. and 4.891 to O.J.J.], the Institut Fédératif de Recherche IFR115 (L.C.), the program interdisciplinaire de recherche du CNRS longévité et vieillissement (L.C.), and the Agence Nationale de la Recherche [grant number ANR-09-BLAN-0048 to L.C.]. L.L. was awarded a fellowship from the Ministère de l’Enseignement Supérieur et de la Recherche. We thank Laura Papp and Kum Kum Khanna for the generous gift of SBP2 antisera.

© 2012 Landes Bioscience. Do not distribute.

References 1. Kryukov GV, Castellano S, Novoselov SV, Lobanov AV, Zehtab O, Guigó R, et al. Characterization of mammalian selenoproteomes. Science 2003; 300:143943; PMID:12775843; http://dx.doi.org/10.1126/ science.1083516. 2. Birringer M, Pilawa S, Flohé L. Trends in selenium biochemistry. Nat Prod Rep 2002; 19:693718; PMID:12521265; http://dx.doi.org/10.1039/ b205802m. 3. Driscoll DM, Chavatte L. Finding needles in a haystack. In silico identification of eukaryotic selenoprotein genes. EMBO Rep 2004; 5:140-1; PMID:14755306; http://dx.doi.org/10.1038/sj.embor.7400080. 4. Patrick L. Selenium biochemistry and cancer: a review of the literature. Altern Med Rev 2004; 9:239-58; PMID:15387717. 5. Rayman MP. Selenium in cancer prevention: a review of the evidence and mechanism of action. Proc Nutr Soc 2005; 64:527-42; PMID:16313696; http://dx.doi. org/10.1079/PNS2005467. 6. Berry MJ, Tujebajeva RM, Copeland PR, Xu XM, Carlson BA, Martin GW, 3rd, et al. Selenocysteine incorporation directed from the 3' UTR: characterization of eukaryotic EFsec and mechanistic implications. Biofactors 2001; 14:17-24; PMID:11568436; http:// dx.doi.org/10.1002/biof.5520140104. 7. Driscoll DM, Copeland PR. Mechanism and regulation of selenoprotein synthesis. Annu Rev Nutr 2003; 23:17-40; PMID:12524431; http://dx.doi. org/10.1146/annurev.nutr.23.011702.073318.

www.landesbioscience.com

8.

9.

10.

11.

12.

13.

Supplemental Materials

Supplemental materials may be found here: www.landesbioscience.com/journals/rnabiology/20147

Hatfield DL, Carlson BA, Xu XM, Mix H, Gladyshev VN. Selenocysteine incorporation machinery and the role of selenoproteins in development and health. Prog Nucleic Acid Res Mol Biol 2006; 81:97-142; PMID:16891170; http://dx.doi.org/10.1016/S00796603(06)81003-2. Papp LV, Lu J, Holmgren A, Khanna KK. From selenium to selenoproteins: synthesis, identity, and their role in human health. Antioxid Redox Signal 2007; 9:775-806; PMID:17508906; http://dx.doi. org/10.1089/ars.2007.1528. Allmang C, Wurth L, Krol A. The selenium to selenoprotein pathway in eukaryotes: more molecular partners than anticipated. Biochim Biophys Acta 2009; 1790:1415-23. Latrèche L, Jean-Jean O, Driscoll DM, Chavatte L. Novel structural determinants in human SECIS elements modulate the translational recoding of UGA as selenocysteine. Nucleic Acids Res 2009; 37:586880; PMID:19651878; http://dx.doi.org/10.1093/nar/ gkp635. Fagegaltier D, Hubert N, Yamada K, Mizutani T, Carbon P, Krol A. Characterization of mSelB, a novel mammalian elongation factor for selenoprotein translation. EMBO J 2000; 19:4796-805; PMID:10970870; http://dx.doi.org/10.1093/ emboj/19.17.4796. Tujebajeva RM, Copeland PR, Xu XM, Carlson BA, Harney JW, Driscoll DM, et al. Decoding apparatus for eukaryotic selenocysteine insertion. EMBO Rep 2000; 1:158-63; PMID:11265756; http://dx.doi. org/10.1093/embo-reports/kvd033.

RNA Biology

14. Ding F, Grabowski PJ. Identification of a protein component of a mammalian tRNA(Sec) complex implicated in the decoding of UGA as selenocysteine. RNA 1999; 5:1561-9; PMID:10606267; http://dx.doi. org/10.1017/S1355838299991598. 15. Xu XM, Mix H, Carlson BA, Grabowski PJ, Gladyshev VN, Berry MJ, et al. Evidence for direct roles of two additional factors, SECp43 and soluble liver antigen, in the selenoprotein synthesis machinery. J Biol Chem 2005; 280:41568-75; PMID:16230358; http://dx.doi. org/10.1074/jbc.M506696200. 16. Small-Howard A, Morozova N, Stoytcheva Z, Forry EP, Mansell JB, Harney JW, et al. Supramolecular complexes mediate selenocysteine incorporation in vivo. Mol Cell Biol 2006; 26:2337-46; PMID:16508009; http://dx.doi.org/10.1128/MCB.26.6.2337-46.2006. 17. Carlson BA, Xu XM, Kryukov GV, Rao M, Berry MJ, Gladyshev VN, et al. Identification and characterization of phosphoseryl-tRNA[Ser]Sec kinase. Proc Natl Acad Sci USA 2004; 101:12848-53; PMID:15317934; http://dx.doi.org/10.1073/pnas.0402636101. 18. Copeland PR, Fletcher JE, Carlson BA, Hatfield DL, Driscoll DM. A novel RNA binding protein, SBP2, is required for the translation of mammalian selenoprotein mRNAs. EMBO J 2000; 19:30614; PMID:10637234; http://dx.doi.org/10.1093/ emboj/19.2.306. 19. Chavatte L, Brown BA, Driscoll DM. Ribosomal protein L30 is a component of the UGA-selenocysteine recoding machinery in eukaryotes. Nat Struct Mol Biol 2005; 12:408-16; PMID:15821744; http://dx.doi. org/10.1038/nsmb922.

689

20. Budiman ME, Bubenik JL, Miniard AC, Middleton LM, Gerber CA, Cash A, et al. Eukaryotic initiation factor 4a3 is a selenium-regulated RNA-binding protein that selectively inhibits selenocysteine incorporation. Mol Cell 2009; 35:479-89; PMID:19716792; http:// dx.doi.org/10.1016/j.molcel.2009.06.026. 21. Miniard AC, Middleton LM, Budiman ME, Gerber CA, Driscoll DM. Nucleolin binds to a subset of selenoprotein mRNAs and regulates their expression. Nucleic Acids Res 2010; 38:4807-20; PMID:20385601; http://dx.doi.org/10.1093/nar/gkq247. 22. Low SC, Grundner-Culemann E, Harney JW, Berry MJ. SECIS-SBP2 interactions dictate selenocysteine incorporation efficiency and selenoprotein hierarchy. EMBO J 2000; 19:6882-90; PMID:11118223; http:// dx.doi.org/10.1093/emboj/19.24.6882. 23. de Jesus LA, Hoffmann PR, Michaud T, Forry EP, Small-Howard A, Stillwell RJ, et al. Nuclear assembly of UGA decoding complexes on selenoprotein mRNAs: a mechanism for eluding nonsense-mediated decay? Mol Cell Biol 2006; 26:1795-805; PMID:16478999; http://dx.doi.org/10.1128/MCB.26.5.1795-805.2006. 24. Papp LV, Lu J, Striebel F, Kennedy D, Holmgren A, Khanna KK. The redox state of SECIS binding protein 2 controls its localization and selenocysteine incorporation function. Mol Cell Biol 2006; 26:4895910; PMID:16782878; http://dx.doi.org/10.1128/ MCB.02284-05. 25. Caban K, Kinzy SA, Copeland PR. The L7Ae RNA binding motif is a multifunctional domain required for the ribosome-dependent Sec incorporation activity of Sec insertion sequence binding protein 2. Mol Cell Biol 2007; 27:6350-60; PMID:17636016; http://dx.doi. org/10.1128/MCB.00632-07. 26. Squires JE, Stoytchev I, Forry EP, Berry MJ. SBP2 binding affinity is a major determinant in differential selenoprotein mRNA translation and sensitivity to nonsense-mediated decay. Mol Cell Biol 2007; 27:7848-55; PMID:17846120; http://dx.doi. org/10.1128/MCB.00793-07. 27. Bermano G, Arthur JR, Hesketh JE. Role of the 3' untranslated region in the regulation of cytosolic glutathione peroxidase and phospholipid-hydroperoxide glutathione peroxidase gene expression by selenium supply. Biochem J 1996; 320:891-5; PMID:9003377. 28. Bermano G, Nicol F, Dyer JA, Sunde RA, Beckett GJ, Arthur JR, et al. Tissue-specific regulation of selenoenzyme gene expression during selenium deficiency in rats. Biochem J 1995; 311:425-30; PMID:7487877. 29. Bermano G, Arthur JR, Hesketh JE. Selective control of cytosolic glutathione peroxidase and phospholipid hydroperoxide glutathione peroxidase mRNA stability by selenium supply. FEBS Lett 1996; 387:157-60; PMID:8674540; http://dx.doi.org/10.1016/00145793(96)00493-0. 30. Wingler K, Böcher M, Flohé L, Kollmus H, BrigeliusFlohé R. mRNA stability and selenocysteine insertion sequence efficiency rank gastrointestinal glutathione peroxidase high in the hierarchy of selenoproteins. Eur J Biochem 1999; 259:149-57; PMID:9914487; http:// dx.doi.org/10.1046/j.1432-327.1999.00012.x. 31. Weiss Sachdev S, Sunde RA. Selenium regulation of transcript abundance and translational efficiency of glutathione peroxidase-1 and -4 in rat liver. Biochem J 2001; 357:851-8; PMID:11463357; http://dx.doi. org/10.1042/0264-6021:3570851.

32. Lei XG, Evenson JK, Thompson KM, Sunde RA. Glutathione peroxidase and phospholipid hydroperoxide glutathione peroxidase are differentially regulated in rats by dietary selenium. J Nutr 1995; 125:1438-46; PMID:7782896. 33. Rebsch CM, Penna FJ, 3rd, Copeland PR. Selenoprotein expression is regulated at multiple levels in prostate cells. Cell Res 2006; 16:940-8; PMID:17160069; http://dx.doi.org/10.1038/sj.cr.7310117. 34. Berry MJ, Banu L, Chen YY, Mandel SJ, Kieffer JD, Harney JW, et al. Recognition of UGA as a selenocysteine codon in type I deiodinase requires sequences in the 3' untranslated region. Nature 1991; 353:273-6; PMID:1832744; http://dx.doi. org/10.1038/353273a0. 35. Martin GW, 3rd, Harney JW, Berry MJ. Selenocysteine incorporation in eukaryotes: insights into mechanism and efficiency from sequence, structure and spacing proximity studies of the type 1 deiodinase SECIS element. RNA 1996; 2:171-82; PMID:8601283. 36. Walczak R, Westhof E, Carbon P, Krol A. A novel RNA structural motif in the selenocysteine insertion element of eukaryotic selenoprotein mRNAs. RNA 1996; 2:367-79; PMID:8634917. 37. Martin GW, 3rd, Harney JW, Berry MJ. Functionality of mutations at conserved nucleotides in eukaryotic SECIS elements is determined by the identity of a single nonconserved nucleotide. RNA 1998; 4:65-73; PMID:9436909. 38. Fagegaltier D, Lescure A, Walczak R, Carbon P, Krol A. Structural analysis of new local features in SECIS RNA hairpins. Nucleic Acids Res 2000; 28:267989; PMID:10908323; http://dx.doi.org/10.1093/ nar/28.14.2679. 39. Grundner-Culemann E, Martin GW, 3rd, Harney JW, Berry MJ. Two distinct SECIS structures capable of directing selenocysteine incorporation in eukaryotes. RNA 1999; 5:625-35; PMID:10334333; http://dx.doi. org/10.1017/S1355838299981542. 40. Dabeva MD, Warner JR. The yeast ribosomal protein L32 and its gene. J Biol Chem 1987; 262:16055-9; PMID:3316213. 41. Vilardell J, Yu SJ, Warner JR. Multiple functions of an evolutionarily conserved RNA binding domain. Mol Cell 2000; 5:761-6; PMID:10882112; http://dx.doi. org/10.1016/S1097-2765(00)80255-5. 42. Allmang C, Carbon P, Krol A. The SBP2 and 15.5 kD/Snu13p proteins share the same RNA binding domain: identification of SBP2 amino acids important to SECIS RNA binding. RNA 2002; 8:1308-18; PMID:12403468; http://dx.doi.org/10.1017/ S1355838202020034. 43. Bubenik JL, Driscoll DM. Altered RNA binding activity underlies abnormal thyroid hormone metabolism linked to a mutation in selenocysteine insertion sequence-binding protein 2. J Biol Chem 2007; 282:34653-62; PMID:17901054; http://dx.doi. org/10.1074/jbc.M707059200.

44. Takeuchi A, Schmitt D, Chapple C, Babaylova E, Karpova G, Guigo R, et al. A short motif in Drosophila SECIS Binding Protein 2 provides differential binding affinity to SECIS RNA hairpins. Nucleic Acids Res 2009; 37:2126-41; PMID:19223320; http://dx.doi. org/10.1093/nar/gkp078. 45. Fletcher JE, Copeland PR, Driscoll DM, Krol A. The selenocysteine incorporation machinery: interactions between the SECIS RNA and the SECIS-binding protein SBP2. RNA 2001; 7:1442-53; PMID:11680849. 46. Budiman ME, Bubenik JL, Driscoll DM. Identification of a signature motif for the eIF4a3-SECIS interaction. Nucleic Acids Res 2011; 39:7730-9; PMID:21685449; http://dx.doi.org/10.1093/nar/gkr446. 47. Boulon S, Marmier-Gourrier N, Pradet-Balade B, Wurth L, Verheggen C, Jády BE, et al. The Hsp90 chaperone controls the biogenesis of L7Ae RNPs through conserved machinery. J Cell Biol 2008; 180:579-95; PMID:18268104; http://dx.doi. org/10.1083/jcb.200708110. 48. Sun X, Li X, Moriarty PM, Henics T, LaDuca JP, Maquat LE. 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 2001; 12:1009-17; PMID:11294903. 49. Sunde RA, Raines AM, Barnes KM, Evenson JK. Selenium status highly regulates selenoprotein mRNA levels for only a subset of the selenoproteins in the selenoproteome. Biosci Rep 2009; 29:329-38; PMID:19076066; http://dx.doi.org/10.1042/ BSR20080146. 50. Diamond AM, Choi IS, Crain PF, Hashizume T, Pomerantz SC, Cruz R, et al. Dietary selenium affects methylation of the wobble nucleoside in the anticodon of selenocysteine tRNA([Ser]Sec). J Biol Chem 1993; 268:14215-23; PMID:8314785. 51. Mehta A, Rebsch CM, Kinzy SA, Fletcher JE, Copeland PR. Efficiency of mammalian selenocysteine incorporation. J Biol Chem 2004; 279:37852-9; PMID:15229221; http://dx.doi.org/10.1074/jbc. M404639200. 52. Chauvin C, Salhi S, Le Goff C, Viranaicken W, Diop D, Jean-Jean O. Involvement of human release factors eRF3a and eRF3b in translation termination and regulation of the termination complex formation. Mol Cell Biol 2005; 25:5801-11; PMID:15987998; http:// dx.doi.org/10.1128/MCB.25.14.5801-11.2005. 53. Brummelkamp TR, Bernards R, Agami R. A system for stable expression of short interfering RNAs in mammalian cells. Science 2002; 296:550-3; PMID:11910072; http://dx.doi.org/10.1126/science.1068999. 54. Berry MJ, Harney JW, Ohama T, Hatfield DL. Selenocysteine insertion or termination: factors affecting UGA codon fate and complementary anticodon:codon mutations. Nucleic Acids Res 1994; 22:3753-9; PMID:7937088; http://dx.doi. org/10.1093/nar/22.18.3753.

© 2012 Landes Bioscience. Do not distribute.

690

RNA Biology

Volume 9 Issue 5

Suggest Documents