Characterization Of The Transcriptional Activity Of Selenoprotein P Biology Essay

Published: Last Edited:

This essay has been submitted by a student. This is not an example of the work written by our professional essay writers.

Selenoprotein P is usually found abundantly in plasma. This protein is selenium-rich. Generally, it has ten in-frame UGA codons and a special co-translational pathway is required for selenoprotein synthesis. In this study, I have constructed the coding sequence of SEPP1 gene into pCDNA3.1 vector then further investigated its activity in western blot. Eventhough it was found that the SEPP1 mRNAs were present abundantly in the transient transfected cell but I could not detect the selenoproteinon on the blot. To express the selenoprotein, I have carried out a series of optimization works. The selenocysteine insertion sequence (SECIS) at 3'UTR is thought to be important for selenocysteine synthesis and mediate the selenoprotein translation in mammalian cell. Thus I tried to clone the full-length SEPP1 gene in CMV2 vector as well as pCDNA3.1 expression vector. I also supplemented the sodium selenite in media. With belief, other than UGA triplet codon, selenocysteine encoding may occur with adequate selenium uptake in cell. Still, there was no significant signal detected neither with the insertion of 3'UTR nor supplementation of sodium selenite in media.



Selenoprotein P (SEPP1) is encoded by SEPP1 gene. It is defined as secreted glycoprotein, interacting with the extracellular matrix through its cytoplasmic projection. This selenoprotein P is extremely unique. There are ten selenocysteine (Sec) residues found flanking within the coding region of SEPP1 (see fig. 1). In general, UGA is characteristically recognized as stop codon in the genetic code. In contrast with it, this UGA is required to encode selenoprotein at the presence of a special elongation factor (SelB) in E.coli (Forchhammer et al., 1989), or else, it can be incorporated with the unique stem-loop-formed cis-acting element then further code the selenoprotein in mammalian cell (Nasim et al., 2000). Ten years earlier, the purified SEPP1 fractions has been derived from rat strain (Tujebajeva et. al., 2000) as well as human plasma (Åkesson et. al., 1994) using Se75- labeling method. Further works on this selenoprotein is not only restricted to purification and expression profile analysis, but many scientists have also investigated its function and cellular regulation pathway, with belief its downregulation may cause disease in human. Evidence for selenoprotein studies has reported that the deficiency of selenium might give rise to cancer (Li et al., 2007). These significant studies have aroused worldwide interest. Recently, our laboratory has preliminarily discovered that SEPP1 was significantly downregulated in the isogenic pairs of esophageal cancer (ESCC) cell lines, with which those daughter cell lines had a higher invasion or migration ability. In this study, I attempted to generate SEPP1 recombinant protein and further examine its immunochemical activity using western blot. This project is anticipated to provide a fundamental study to investigate the functions of SEPPl as well as its critical role in cancer.



Figure 1 A and B: Distribution of selenocystein on the SEPP1 domain (Raymond & Kristina, 2009).


To generate a SEPP1 recombinant protein

To determine the immunogenicity of influenza A NS1 recombinant protein

To investigate the expression of selenoprotein P mRNA in mammalian cell


My attempt for this project was trying to express the SEPP1 in mammalian cell and further investigated its function. First, the SEPP1 gene was extracted from reference cell, A503371. Preliminarily, I have cloned the coding sequence of SEPP1 gene into both pcDNA3.1/V5-HIS vector (Invitrogen) (see appendix 1) followed by a series of investigations to determine the activity of SEPP1. Two main examinations were conducted: (1) determination of SEPP1 immunogenicity, and (2) determination whether SEPPl established strong expression signal at both RNA and protein level. In this study, I have refined the cloning part and the cell transfection works that have been done by research team of Stoytcheva (Stoytcheva et. al., 2006) and Tujebajeva (Tujebajeva et. al., 2000). In their study, they have generated a radiolabeled SEPP1 protein then successfully expressed the purified Se75- labeled selenoprotein in mammalian cell. In addition to it, my following optimization works included cloning of full-length SEPP1 gene in N-terminal pFLAG-CMV2 (see appendix 2) and pcDNA3.1/V5-HIS vector (Invitrogen) respectively.

PCR amplification and DNA cloning

SEPP1 CDS cloning and transformation

The human selenoprotein P cDNA was amplified by PCR. Two specific primers complementary to the selenoprotein P coding sequence region (CDS), 5'-GTTGTGACAACCCCAGCAATG-3' and 5'-GTTTGAAGGTCATTCTCACTTTTTTG-3', were purposely designed. Subsequently, the PCR product was cloned into pcDNA™3.1/V5-His-TOPO® (see appendix 1). Transformation into E. coli TOP10 host was initiated by heatshock.

Full length SEPP1 cloning and transformation

Amplification of full-length human selenoprotein P cDNA was performed using specifically designed upstream primer 5'-TACGAATTCAATGTGGAGAAGCCTGGGGCTT-3' and downstream primer 5'-TTACTCGAGCAGCTTTAAGGTTTTTATTGAATTTATTTG-3'. The upstream primers was designed to create sequence starting from nucleotides 101-115 in the SEPP1 (NM_005410.2) coding region by introduced an EcoRI cutting site, whereas the downstream primer containing XhoI site amplified the SEPP1 3'UTR sequence. The ~2-kilobase pair PCR products were subcloned into EcoRI/SalI-cleaved N-terminal pFLAG-CM2 vector according to standard procedure (see appendix 2 and 3). Next, the plasmids DNA were transformed into E. coli DH5α host by heatshock. Plasmid S7 and S9 were generated after inserting the 3'UTR nucleotide of SEPP1 into pCDNA3.1-SEPP1-CDS plasmid. While, the full-length SEPP1 gene was constructed into pFLAG-CMV expression vector then generated the L6 and L7 plasmids.

Clone identification

The colonies were quick-screened using either colony PCR or phenol-chloroform extraction. Next, those false or true positive clones were confirmed by plasmid isolation, restriction enzymes digestion, then further sent to dideoxy sequencing for sequence verification.

Cell transfection and western blotting

Transfecting pcDNA3.1-SEPP1-CDS into HEK293T

Human embryonic kidney (HEK) 293T cells were grown in Dulbecco's modified Eagle's medium supplemented with 10% fetal bovine serum (Invitrogen) containing appropriate antibiotics at 37°C. Transfection was designed using LipofectamineTM 2000 (Invitrogen) and the cells seeded in 6-well dish were transfected with 3ug of pcDNA3.1-SEPP1-CDS. After 6 hour-incubation, media were changed to 5ml DMEM containing 10% FBS then incubated at 37°C in a CO2 incubator for 48 hours.

After 48-hour-incubation, both the supernatant and cell lysate were harvested. The western blot was performed to determine the antigenicity of SEPP1 recombinant protein. After the blotting transfer was stopped, the membrane was removed from the cassette and subsequently blocked with 1% bovine serum albumin (BSA) in 1% TBST (1X TBS with 0.1% Tween-20) for 1 hour at room temperature. Then the membrane was washed 3 times with TBST for 15 minute. After being washed with TBST, the membrane was allowed for overnight incubation at room temperature with mouse anti-v5 tag monoclonal antibody at 1:2500 dilution in 1% BSA then further probed by HRP-conjugated goat anti-mouse IgG with 1:5000 dilution in 5% skim milk for 1 hour. 3-times-washing with 1% TBST was applied on the membrane after each process of incubation. After that, signal was detected by Pierce ECL Western Blotting Substrate.

To verify activity of full-length SEPP1

Alternatively, four constructs containing full-length of SEPP1, which were CMV2-SEPP1 (L6 and L7) and pcDNA3.1-SEPP1 (S7 and S9 were produced by Shen-Yang) were subjected to western blot analysis to verify activity of SEPP1. The transfection conditions were described as below (see table 1). Same as previously mentioned, the cells were seeded in DMEM containing 10% fetal bovine serum three days prior to transfection. Cells in 60-mm dish were transfected with 8ug of pcDNA3.1-SEPP. After 6 hour-incubation, media were changed to 5ml DMEM containing 10% FBS then incubated at 37°C in a CO2 incubator for 48 hours. Western blot performed here with condition described in table 2 was used to verify activity of full-length SEPP1.





CMV2 vector




pCDNA3.1 vector




Plasmid L6




Plasmid L7




Plasmid S7




Plasmid S9




Table 1: Transfection using LipofectamineTM 2000 (Invitrogen) in HEK293T cells.

Table 2: Transfection using LipofectamineTM 2000 (Invitrogen) in HEK293T cells.

A trial of selenium supplementation

HEK293 cells were grown in Dulbecco's modified Eagle's medium supplemented with 10% fetal bovine serum (Invitrogen). Cells were seeded in 60-mm plate then incubated at 37°C for overnight. The day before transfection, cells were supplemented with 100mM sodium selenite then allowed to grow until 60-80% confluency. Cells were transfected with 8ug of plasmid DNA as described as above (see table 1). After 6 hour-incubation, media were changed to 5ml 1% FBS DMEM then incubated at 37°C. 2ul of 100mM sodium selenite were added to media after 24-hour-incubation. Transfection lasts for 48 hours and both the supernatant and cells were harvested for analysis.


Preliminary data had implied that human selenoprotein P might express in the colorectal cancer patients (unpublished data), therefore, I re-examined this possibility and investigated whether the SEPP1 recombinant protein also established the significant activity. To produce SEPP1 recombinant protein, the SEPP1 gene was initially constructed into a eukaryotic expression vector. The successful clones were subjected to NCBI-BLAST sequence alignment and the result revealed that significant protein homology to Homo sapiens selenoprotein P, plasma, 1 (SEPP1), transcript variant 1, mRNA.

Expression of Recombinant Selenoprotein P in Mammalian Cell Lines

In my first batch experiment, I performed western blot to further verify the SEPP1 protein expression, with pcDNA™3.1/V5-His-TOPO® vector-infected cells was served as negative control. SEPP1 is defined as secreted protein, thus, other than media, I also analyzed the cell lysate from those transciently transfected cells and intended to detect any difference between both. Somewhat unexpected that there was no significant activity detected on blot (see fig. 2), SEPP1 gene-containing plasmids showed similar protein pattern with pcDNA3.1 control vector. It was indicated that pcDNA3.1-SEPP1-CDS failed to enhance the selenoprotein production in mammalian cells. Conceivably, a presumption can be made that the depletion of mRNA transcription was not sufficient for reducing the protein expression. To qualitatively verify the SEPP1 RNA, I extracted the RNA from trancfected cell then subjected them to qualitative PCR analysis.

According to gene expression profile, mRNA extracted from cell transfected with SEPP1-containing plasmid (pcDNA3.1-SEPP1-CDS) was observed showing an elevated fold change if compared to both negative controls, naïve and pCDNA3.1 (see chart 1). Therefore, I concluded that SEPP1 gene was expressed and mRNA transcripts were present in cells eventhough it was no protein accumulated detected either in the supernatant or cell lysate, One of the possible explanations is that the protein translational event was brought to premature termination by a cluster of UGA codons which appear abundantly within the SEPP1 coding region. Accordingly, the selenoprotein failed to complete its production in the transfected cell. In fact, seven years earlier, Stoytcheva and his co-workers have unraveled that the selenoprotein syntheses more or less depend on the incorporation of cis-acting mRNA element at UGA site to stimulate the protein synthesis (Stoytcheva et. al., 2006). Thus, I hypothesize that the unique tRNA secondary structure in the 3'-untranslated region may play a critical role in UGA decoding.

Other than that, comparison has been done to verify the efficiency of secondary antibody. The antibodies seemed to show non-specific binding on the protein because there was no much difference shown between membrane probed with or without primary antibodies (fig.2). Thus, some problems regarding high background and non-specific binding on blot I have to tackle.

Chart 1: QPCR result: RNA expression level of pcDNA3.1-SEPP1-CDS plasmid, pCDNA3.1 vector only, and naïve after 48-hour transfection.

Figure 2: Western blot analysis on expression of SEPP1 fusion protein and efficiency test on mouse anti-v5 tag monoclonal antibody. Lane M: MW marker (PageRuler Prestained Protein Ladder, product #: 26616); lane 1: naive, lane 2: pcDNA3.1-SEPP1-CDS plasmid, lane 3: pCDNA3.1 vector only. Samples were probed to mouse anti-v5 tag monoclonal antibody at 1:2500 dilution in 1% BSA. α-actin acted as the loading control in western blot.

Optimizing Expression in Transiently Transfected Cells

The presence of two SECIS elements were unsatisfactory for protein encoding.

With belief, the presence of two selenocysteine incorporation sequence (SECIS) elements in the 3'-untranslated region may be incorporated with selenocystein at UGA site to generate selenoprotein (Burk & Hill, 2009). Then I further tested whether SEPP1 is able to restore its translational activity in the transfected cell with the presence of 3'-untranslated regions (UTR) nucleotide. The S7 and S9 were produced by adding 3'UTR nucleotide into previously cloned pcDNA3.1-SEPP1-CDS constructs, whereas the PCR generated full-length SEPP1 gene were cloned directly into EcoRI/SalI-cleaved N-terminal pFLAG-CM2 vector then generated both the L6 and L7 plasmids. After sequenced the S7 and S9, there has been found that S7 contains single-nucleotide polymorphism (SNP) in the 3'UTR sequence. I was not sure whether the protein sequence will be affected. Moreover, mutations were detected on L7 with the new shift frame was created. To test their effects on protein synthesis. I transfected those plasmid to HEK293T then analyzed their activity in western blot. First, by referring to the qPCR result, SEPP1 gene-containing CM2-SEPP1 constructs (L6 and L7) were found to be expressed in the HEK293T if compared to its control vector. Indeed, we observed the similar scenario in HEK293 cell transfected with the pcDNA3.1-SEPP1 (see chart 2), which this result was consistent with the preliminary expectation. Still, it was observed that neither pcDNA3.1-SEPP1 nor CMV2-SEPP1 successfully produced selenoprotein in the transfected cells (refer to fig. 3). To my mind, the insertion of 3'-UTR is not thought to be sufficient for protein production.

According to the result, I suspected that the human selenoprotein P may perhaps possess a high degradation rate, indicating the SEPP1 mRNA transcripts degrade rapidly when they are exported to cytomplasm (Burk & Hill, 1994). Next, the proteasome inhibitor was proposed to be added into the media, which was thought to be ideal solution to inhibit or reduce the degradation of selenoprotein. However, this solution can be rule out because previous work done by research team of Tujebajeva has demonstrated that there was no apparent change found. They still did not detect accumulation of selenoprotein in the transfected cell (Tujebajeva et. al., 2000) after the cells were treated by proteasome inhibitor. The similar failure to detect significant protein expression on the blot have prompted me to look for whether SEPP1 requires other supplements to give aid to the UGA decoding mechanism in mRNA then enhance selenoprotein synthesis. Perhaps the premature termination of protein synthesis resulted by selenium starvation in the cell (Stoytcheva et. al., 2006). After reviewing relevant literature, I get to know that the supplementation of selenium in cell culture may perhaps enhance the expression of selenoprotein, indicating that the protein expression was significantly elevated after relatively treated with sodium selenite. It has been long reported that inadequate selenium amount in cell might deficit the SEPP1 mRNA transcription in cell. Strong evidences have been demonstrated on glutathione peroxidase (Bermano et. al., 1999) and hepatic selenoprotein (Li et. al., 2007). Strikingly, elevated selenoprotein expression was recorded in previous studies on effect of 100nM sodium selenite (Tujebajeva et. al., 2000). Based on this hypothesis, I accessed this possibility and concluded that selenium supplementation could augment the protein expression other than insertion only the cis-acting element at 3'UTR.

Chart 2: Quantitative PCR (qPCR) analysis of SEPP1 expression in HEK293T cells. Fold changes of CMV2-SEPP1 (L6 and L7) were normalized by CMV2 vector whereas activities of pcDNA3.1-SEPP1 (S7 and S9) were compared to pCDNA3.1 vector only.

Figure 3: Western blots of total cellular protein from HEK293T cells transfected with indicated expression plasmids (8μg). Control cells were transfected with 8μg of pCDNA3.1 and CMV2. M: MW marker (PageRuler Prestained Protein Ladder, product #: 26616); V: vector only (pCDNA3.1 or CMV2); L6 and L7: CMV2-SEPP1 and S7 and S9: pcDNA3.1-SEPP1. α-actin acted as the loading control in western blot showed representative bands at ~43kDa.

No detectible selenoprotein accumulation after supplemented with sodium selenite.

Since SEPP1 gene was found to express in HEK293T cells efficiently, however, the insertion of SECIS at UGA site for selenoprotein encoding did not establish any apparent changes on SEPP1 expression pattern in my previous experiment. Thus, it can be concluded that the introduction of 3'UTR in SEPP1 gene is unsatisfactory for selenoprotein production in mammalian cell. Possibly, the selenium amount is not sufficient for protein synthesis in cell particularly after the transcient transfection. Moreover, strong evidence has been reported that supplementation of sodium selenite in media typically could enhance the selenoprotein expression (Tujebajeva et. al., 2000).

In order to verify our suspicion, I further investigated the effect of sodium selenite on selenoprotein expression. In my following batch experiment, I attempted to test the hypothesis by treating the cells with 100nM sodium selenite. All cells were treated with sodium selenite before transfected by SEPP1-gene-flanked-plasmids. I also prepared a control plate in this trial with which the S7* was not additionally treated with sodium selenite after transfection. On the other hand, plasmid Clo and JR was taken as positive control vector for primary antibodies α-V5 and α-FLAG respectively.

No doubt, in this latest qPCR result, mRNA derived from all transfected cells were found to be highly expressed in HEK293T, significant increase was recorded up-to 10-fold (see chart 3). Interestingly, in Chart 3, S7 and S7* demonstrated similar expression level, indicating that there is no obvious effect established on gene expression for the trial adding sodium selenite after transfection. Next, the total cellular proteins were subjected to western blotting (see fig.4). However, the selenoprotein expression pattern was not much improved after addition of sodium selenite in culture media. Besides, both control vectors, Clo and JR, were significantly detected in western blot, demonstrating that the specific detection of flag as well as v5 fusion proteins by antibodies was satisfactory.

In my previous experiment, a cluster of unrecognized proteins (suspected those are albumin or other cellular proteins) with an approximate molecular weight of 70 kDa has been detected on blot (see fig. 3 and 2). Therefore, to avoid highly accumulation of albumin in media, the media was changed to low serum DMEM after 6-hour transfection in my latest trial. Surprisingly, significant improvement was shown (see fig. 4), no strong protein signal detected at ~70kDa molecular weight.

Chart 3: Quantitative PCR (qPCR) analysis of SEPP1 expression in HEK293T cells. Fold changes of CMV2-SEPP1 (L6 and L7) were normalized by CMV2 vector whereas activities of pcDNA3.1-SEPP1 (S7, S7* and S9) were compared to pCDNA3.1 vector only. S7* acted as control in this experiment without additionally treated with sodium selenite after transfection.

Figure 4: Western blots of total cellular protein from HEK293T cells transfected with indicated expression plasmids (8μg). Control cells were transfected with 8μg of pCDNA3.1 and CMV2. M: MW marker (PageRuler Prestained Protein Ladder, product #: 26616); V: vector only (pCDNA3.1 or CMV2); L6 and L7: CMV2-SEPP1 and S7, S7* and S9: pcDNA3.1-SEPP1. Clo and JR was positive control for α-V5 and α-FLAG respectively whereas α-actin acted as the loading control in western blot showed representative bands at ~43kDa.


With an adequate selenium level, the two SECIS element is essential for co-translation of selenocystein at UGA site to syntheses selenoprotein. Nonetheless, my attempts to express the selenoprotein in mammalian cell could not obtain a significant result. Despite of SEPP1 mRNA expressed abundantly in cell, the production of selenoprotein in HEK293T was not significantly improved with insertion of SECIS elements in its 3'UTR. As well, supplementation of sodium selenite in media did not established obvious changes on protein production. Overall, my current laboratory trials indicate that SEPP1 mRNAs are still being synthesized in cell but the translation activities has been terminated in the cytoplasm.

There are many factors which could lead to protein termination (see figure 5). Other trans-acting elements are thought to initiate the translational elongation events then facilitate the mRNA transcription (Driscoll & Copeland, 2003). Evidence has demonstrated that the SECIS binding protein (SBP2) is recruited for Sec insertion in vitro (Copeland et. al., 2000). Therefore, I think that the decoding of UGA to selenocysteine not only requires the interaction with SECIS at downstream sequence, but the binding of other proteins is also crucial. The involvement of SBP2 is vital in selonoprotein synthesis, with belief that this specialized factor is essential for binding of ribosome to the Sec-tRNASec. To ensure that the sec insertion is processive, future trial is suggested to co-transfect the SBP2-flanked plasmid and SECIS-3'UTR-containing plasmid into HEK293T (see figure 6).

Figure 5: Selenoprotein expression system of this project and the existing factors which may influence the protein production.

Figure 6: Suggestion for next optimization work.


Åkesson, B., Bellew, T., & Burk, R. F. (1994). Purification of selenoprotein P from human plasma. Biochimica et Biophysica Acta (BBA)-Protein Structure and Molecular Enzymology, 1204(2), 243-249.

Bermano, G., Arthur, J. R., & Hesketh, J. E. (1996). Selective control of cytosolic glutathione peroxidase and phospholipid hydroperoxide glutathione peroxidase mRNA stability by selenium supply. FEBS letters, 387(2), 157-160.

Burk, R. F., & Hill, K. E. (2009). Selenoprotein P-expression, functions, and roles in mammals. Biochimica et Biophysica Acta (BBA)-General Subjects, 1790(11), 1441-1447.

Burk, R. F., & Hill, K. E. (1994). Selenoprotein P. A selenium-rich extracellular glycoprotein. The Journal of nutrition, 124(10), 1891.

Copeland, P. R., Fletcher, J. E., Carlson, B. A., Hatfield, D. L., & Driscoll, D. M. (2000). A novel RNA binding protein, SBP2, is required for the translation of mammalian selenoprotein mRNAs. The EMBO journal, 19(2), 306-314.

Driscoll, D. M., & Copeland, P. R. (2003). Mechanism and regulation of selenoprotein synthesis. Annual review of nutrition, 23(1), 17-40.

Forchhammer, K., Leinfelder, W., & Böck, A. (1989). Identification of a novel translation factor necessary for the incorporation of selenocysteine into protein. Nature, 342(6248), 453-456.

Li, C. L., Nan, K. J., Tian, T., Sui, C. G., & Liu, Y. F. (2007). Selenoprotein P mRNA expression in human hepatic tissues. WORLD JOURNAL OF GASTROENTEROLOGY, 13(16), 2363.

Nasim, M. T., Jaenecke, S., Belduz, A., Kollmus, H., Flohé, L., & McCarthy, J. E. (2000). Eukaryotic selenocysteine incorporation follows a nonprocessive mechanism that competes with translational termination. Journal of Biological Chemistry, 275(20), 14846-14852.

Stoytcheva, Z., Tujebajeva, R. M., Harney, J. W., & Berry, M. J. (2006). Efficient incorporation of multiple selenocysteines involves an inefficient decoding step serving as a potential translational checkpoint and ribosome bottleneck. Molecular and cellular biology, 26(24), 9177-9184.

Tujebajeva, R. M., Harney, J. W., & Berry, M. J. (2000). Selenoprotein P expression, purification, and immunochemical characterization. Journal of Biological Chemistry, 275(9), 6288-6294.

Appendix 1: pCDNA 3.1 vector map