Selenocysteine insertion sequence binding protein 2L is implicated as a novel post-transcriptional regulator of selenoprotein expression

Abstract

The amino acid selenocysteine (Sec) is encoded by UGA codons. Recoding of UGA from stop to Sec requires a Sec insertion sequence (SECIS) element in the 3' UTR of selenoprotein mRNAs. SECIS binding protein 2 (SBP2) binds the SECIS element and is essential for Sec incorporation into the nascent peptide. SBP2-like (SBP2L) is a paralogue of SBP2 in vertebrates and is the only SECIS binding protein in some invertebrates where it likely directs Sec incorporation. However, vertebrate SBP2L does not promote Sec incorporation in in vitro assays. Here we present a comparative analysis of SBP2 and SBP2L SECIS binding properties and demonstrate that its inability to promote Sec incorporation is not due to lower SECIS affinity but likely due to lack of a SECIS dependent domain association that is found in SBP2. Interestingly, however, we find that an invertebrate version of SBP2L is fully competent for Sec incorporation in vitro. Additionally, we present the first evidence that SBP2L interacts with selenoprotein mRNAs in mammalian cells, thereby implying a role in selenoprotein expression.

Figure 1. Capitella CT-SBP2L is able to promote Sec incorporation.

(A) Diagram of the luciferase Sec incorporation reporter construct. (B) 5 fmol of in vitro translated CT-SBP2 (blue bars) or Capitella CT-SBP2L (red bars) were added to a rabbit reticulocyte lysate assay programmed with a luciferase reporter construct containing a Sec codon at position 258 and the SECIS element indicated. The data is reported as the mean ± standard deviation of three experiments (except for the mutant GPX4 SECIS element, which was only tested once). (C) The data from (B) was replotted to show the fold enhancement of Sec incorporation upon the addition of either CT-SBP2 (blue bars) or Capitella CT-SBP2L (red bars).

Figure 2. CT-SBP2L does not promote Sec incorporation.

(A) Domain architecture of human SBP2 and SBP2L. The conserved Sec incorporation domains (SID) and RNA binding domains (RBD) are indicated. The C-terminal (CT−) portions of the proteins used through this study are underlined. (B) A representative gel showing [³⁵S]-Met labeled translation products of the Sec incorporation reporter mRNA in the presence of the indicated recombinant proteins. Full-length (FL) product is the result of Sec incorporation while the termination product (Term) results from translation termination at the in-frame UGA/Sec codon.

Figure 3. SBP2L is a universal SECIS binding protein.

Representative EMSA gels of CT-SBP2 and CT-SBP2L with the human GPX2 (A), SelW (B), SelH (C), and SelV (D) SECIS elements.

Figure 4. Summary of SECIS binding data.

(A) The mean apparent dissociation constants (K_d) for CT-SBP2 and CT-SBP2L obtained from the values reported in Table 1. (B) Scatterplot of CT-SBP2L SECIS dissociation constants as a function of CT-SBP2 SECIS dissociation constants. (C) Mean dissociation constants of CT-SBP2 and CT-SBP2L for Form 1 and Form 2 SECIS elements. (D) Total luciferase activity from a luciferase Sec incorporation reporter construct containing the human SelV SECIS element. A range of recombinant CT-SBP2 or CT-SBP2L (10–160 nM) was added as indicated.

Figure 5. SBP2 and SBP2L compete for SECIS binding in vitro.

(A) EMSA of 20 nM wild-type (WT) or mutant (G721R) CT-SBP2L with the indicated SECIS elements. (B) Top: UV Cross-linking (X-link) of CT-SBP2 and CT-SBP2L. Bottom: Quantitation of CT-SBP2 signal in the presence of increasing concentrations of CT-SBP2L. (C) EMSA of the SelV SECIS element with CT-SBP2 in the presence of increasing amounts of CT-SBP2L. (D) Sec incorporation activity in the presence of 8 nM CT-SBP2 plus 160 nM wild-type or G721R mutant CT-SBP2L. Luciferase activity from 1 nM luciferase mRNA harboring the SelV SECIS element is normalized to a reaction containing no CT-SBP2L (left bar).

Figure 6. The SBP2L SID and RBD do not stably interact.

(A) 6xHis-Xpress tagged rat SBP2 SID was incubated with FLAG tagged rat SBP2 RBD and wild-type (WT), core deleted (ΔAUGA), or loop mutant (TTT) rat GPX4 SECIS as indicated and complexes were immunoprecipitated (IP) with α-FLAG agarose. Top western blot: 50% pellet; middle western blot: 7.5% supernatant; bottom panel: RNA extracted from 50% of the supernatant was resolved on an agarose gel. (B) Same as in panel A but 6xHis-Xpress tagged SBP2L SID was incubated with 6xHis-FLAG tagged SBP2L RBD.

Figure 7. Overexpressing CT-SBP2L does not alter selenoprotein expression.

HEK293 cells were transiently transfected with empty vector, wild-type (WT), or G721R CT-SBP2L expression vectors and labeled for 24 h with ⁷⁵Se. (A) Equal amounts of cell lysates were analyzed by western blotting for SBP2L and β-actin. (B) Equal amounts of cell lysates were fractionated by SDS-PAGE. The gel was stained (left) to visualize total protein and ⁷⁵Se labeled proteins were detected by phosporimaging (right).

Figure 8. Association of selenoprotein mRNAs with SBP2L in mammalian cells.

(A) Qualitative RT-PCR analysis of selenoprotein mRNAs (GPX4 and TR1) and β-actin mRNA extracted from pre-immune (PI) and immune (Imm) immunoprecipitations (IP) of SBP2L from U87MG cytoplasmic extracts. (B) Qualitative RT-PCR analysis of selenoprotein mRNAs (GPX1 and GPX4) and β-actin mRNA extracted from pre-immune (PI) and immune (Imm) immunoprecipitations (IP) of SBP2 and SBP2L from PC3 cytoplasmic extracts. RT, reverse transcriptase.