Phospholipid hydroperoxide glutathione peroxidase (PHGPx or GPx4) was originally described as a selenoprotein that protects liposomes and biomembranes from peroxidation and was thus implicated in the protection of biomembranes against oxidative stress . Three distinct mRNAs differing in their 5′ ends are produced by the GPx4 gene and encode for the mitochondrial, the cytosolic, and the nuclear GPx4, respectively . Since the definition of the primary structure of cytosolic GPx4, several hundreds of humongous sequences have been deposited in databanks. The cytosolic GPx4 is homologous to the previously known tetrameric Sec-containing cytosolic GPx1 . Both enzymes use glutathione (GSH) as a reductant, but only GPx4 reduces lipid hydroperoxides in membranes besides small hydroperoxides such as H2O2 or free fatty acid hydroperoxides .
Over the last couple of years, it became evident that GPx4 protects cells against various apoptotic stimuli, such as prooxidants, DNA damaging agents, glucose depletion, and ultraviolet irradiation [5-7]. The specific function of GPx4 depends mostly on the selenocysteine (Sec) in the peptide. Sec is the 21st amino acid  and is encoded by UGA, a codon that typically dictates translational termination. The intricate mammalian Sec incorporation mechanism imposes a major obstacle in obtaining recombinant GPx4. The Sec insertion sequence (SECIS) in the 3′-untranslated region (3′-UTR) of the selenoprotein-coding mRNA and SECIS binding protein 2 (SBP2) was demonstrated to be essential for the efficient incorporation of Sec into peptides [9, 10]. The SBP2 is believed to recruit Sec-tRNA(Ser)Sec-specific elongation factor, which is specific for Sec-tRNA(Ser)Sec and delivers it to the ribosome .
Recently, an efficient vector pSelExpress1  for overexpression of selenoprotein was developed, and it was demonstrated that a Sec-containing glutathione S-transferase could be obtained using this expression vector [13, 14].
In this study, human GPx4 (hGPx4) gene was cloned into pSelExpress1-leader (PL) and transfected into HEK293T cells. Recombinant hGPx4 was purified from the cell culture supernatant and analyzed using Western blotting. Previous work suggests that hGPx4 purified from human placental cytosol catalyze as in the “ping–pong mechanism” . Our results showed that the purified recombinant hGPx4 has a high glutathione peroxidase (GPx) activity and a similar reaction mechanism to that of native GPx4.
Materials and Methods
The vector pSelExpress1 was a generous gift from Vadim N. Gladyshev. Reduced nicotinamide adenine dinucleotide phosphate (NADPH), ethylenediaminetetraacetic acid (EDTA), and glutathione reductase were purchased from Sigma (St. Louis, MO, USA). MiniBEST Plasmid Purification Kit was purchased from TAKARA (Mountain View, CA, USA). Tryptone, yeast extract, Zeocin, culture medium [Dulbecco's modified Eagle's medium (DMEM)], and fetal bovine serum (FBS) were purchased from Gibco (Rockville, MD, USA). All antibodies involved in Western blotting analysis were purchased from Pharmacia. All the other chemicals were obtained from Beijing Chemical Factory, China, and are of analytical grade.
Amplification of hGPx4 Sequence and Construction of the Eukaryotic Expression Plasmid pSelExpress1-leader-GPx4 (PL-GPX4)
hGPx4 gene was amplified from the complementary DNA (cDNA) library of human hepatoma cell line (HepG2) using primers 5′-CCCAAGCTTAACATGGCAATGTGCGCGTCC-3′ and 5′-GCTCTAGACTAATGATGATGATGATGATGGAAATAGTGGGGCAG-3′. The amplified hGPx4 gene fragment was electrophoresed in 0.8% agarose gel stained with ethidium bromide, and then the fragment was visualized by Tanon-1600 figure gel image processing system and analyzed by GIS 1D gel image system software (Tanon, Shanghai, China). Plasmid pSelExpress1 served as the vector for expression of recombinant hGPx4 in this study. The vector pSelExpress1 contained the human cytomegalovirus immediate-early promoter, the human elongation factor 1α-subunit promoter, and SBP2, which could enhance the high-level constitutive expression of recombinant selenoproteins. The murine Ig κ-chain leader sequence from pSecTag2A plasmid had been introduced into pSelExpress1 plasmid to allow secretory expression of the target protein in our previous work , yielding plasmids PL. As a result, there are two HindIII restriction enzyme recognition sites, and we used site-directed mutagenesis to eliminate the one in the upstream of the leader sequence. The hGPx4 gene digested with HindIII and XbaI could be ligated into the vector in the downstream of the leader sequence. The mutant plasmid was generated using the mutagenic primers (forward primers: 5′-CTCACTATAGGGAGACCCCAGCTTACCATGGAGACAGAC-3′; reverse primers: 5′-GTCTGTCTCCATGGTAAGCTGGGGTCTCCCTATAGTGAG-3′), and the mutation was verified by automated Sanger sequencing. hGPx4 gene fragment was digested with HindIII and XbaI and subcloned into the HindIII and XbaI sites of pSelExpress1 vector. The resulting plasmid was designated as PL-GPx4.
Cell Culture and Transient Transfection of Human HEK293T Cells
Human HEK293T cells were cultured in DMEM (high glucose) with 10% FBS at 37 °C in 5% CO2. Cells were seeded in growth medium without antibiotics, and then the medium was replaced with serum-free DMEM before transfecting with the plasmid PL or PL-GPx4 using Lipofectamine 2000 according to the manufacturer's protocol. The medium was replaced with DMEM supplemented with 10 μM sodium selenite and 10% FBS 6 h after transfection, and then the cells were incubated for an additional 48 h. The cell culture supernatant was collected for purification.
Purification of Recombinant hGPx4
Recombinant hGPx4 with His-tag in its C-terminal was purified by immobilized metal affinity chromatography (IMAC) as mentioned previously . The supernatant was filtered with 0.45 μM filter and then incubated with Ni–agarose resin, which was equilibrated with buffer A (50 mM sodium phosphate, 300 mM NaCl, pH 7.4) beforehand for 2 h at 4 °C on a rotary mixer. The resin was then packed under gravity into a small column and washed extensively with buffer A to remove the unbound proteins. Then the column was eluted with buffer A containing 50 and 300 mM of imidazole. The final purified recombinant hGPx4 was in buffer A containing 300 mM of imidazole and identified by Western blotting. The protein concentration was determined by the Bradford method.
Assay of GPx Activity
Activity of recombinant hGPx4 was determined using the method as described previously . The reaction was carried out at 37 °C in 700 μL of solution containing 50 mM sodium phosphate buffer (pH 7.4), 1 mM EDTA, 1 mM sodium azide, 1 mM GSH, 0.25 mM NADPH, 1 U of glutathione reductase, and 10–50 nM of recombinant hGPx4. Recombinant hGPx4 was preincubated with GSH, NADPH, and glutathione reductase for 5 min. The reaction was initiated by the addition of 0.5 mM H2O2. The activity was determined by the decrease of NADPH absorption at 340 nm. Appropriate controls were performed without recombinant hGPx4 and subtracted. The activity was expressed as micromoles of NADPH oxidized per minute. The value is presented by µmol/min/mg protein.
Assay of Steady-State Kinetics of Recombinant hGPx4
The assay of kinetics of recombinant hGPx4 was similar to that of seleno-LuGST1-1 . The initial rates were determined by following the decrease of NADPH absorption at 340 nm at several concentrations of one substrate, whereas the concentration of the second substrate was kept constant. All kinetic experiments were performed in 700 μL of the reaction solution containing 50 mM sodium phosphate buffer (pH 7.4), 1 mM EDTA, 1 U of glutathione reductase, 0.25 mM NADPH, 10–50 nM of recombinant hGPx4, and varying concentrations of GSH and H2O2. After the recombinant hGPx4 was preincubated with GSH, glutathione reductase, and NADPH for 5 min, the reaction was initiated by the addition of H2O2. The nonenzymatic reaction affecting the measurement of the initial rate was taken into account and subtracted to give the kinetic values.
Amplification of hGPx4 Gene
The polymerase chain reaction (PCR) was carried out using the primer mentioned above and the cDNA library of HepG2 as the template. The PCR products were electrophoresed in 0.8% agarose gel containing ethidium bromide and visualized by Tanon-1600 figure gel image processing system. The results show that there is a specific band that appears at about 500 bp (Fig. 1).
Construction and Identification of hGPx4 Expression Vector
The vector PL and hGPx4 gene fragment were digested with HindIII and XbaI. The hGPx4 PCR product digested with HindIII and XbaI was ligated into the precut vector PL to make PL-GPx4. Construction of PL-GPx4 was confirmed by dual-enzyme digestion of HindIII and XbaI. There is a band of about 500 bp that appears after digestion of PL-GPx4 (Fig. 2). It suggests that hGPx4 gene had been successfully subcloned into plasmid PL. Sequencing result is consistent with the expected gene sequence (data not shown).
Purification and Identification of Recombinant hGPx4
Primers were designed to incorporate His-tag in the C-terminal of the protein expressed, which made it convenient to purify the recombinant hGPx4 by IMAC. The purified recombinant hGPx4 was identified by Western blotting. As shown in Fig. 3, a specific protein was detected, and its molecular weight was about 21.0 kDa, which was consistent with the calculated value of hGPx4. This result indicates that the recombinant hGPx4 has been successfully expressed in the eukaryotic expression system.
Assay of GPx Activity of Recombinant hGPx4
Activity of the recombinant hGPx4 is listed in Table 1. Recombinant hGPx4 shows remarkable GPx activity of 30.6 µmol/min/mg protein, which is close to that of native GPx4 purified from human placental cytosol (34.8 µmol/min/mg) but lower than that of native GPx4 purified from pig heart (140 µmol/min/mg). The activity of recombinant hGPx4 was completely lost when treated with excess iodoacetate in the presence of GSH, indicating the presentation of the enzyme-bound selenol in the catalytic cycle .
Table 1. Gpx activities of recombinant GPx4 and native GPx4 of different resource
Assay of Steady-State Kinetics of Recombinant hGPx4
To study the possible catalytic mechanism of recombinant hGPx4, the initial rate was determined at several concentrations of the two substrates. When the concentration of one substrate was fixed, the initial reaction rate was measured under different concentrations of the other substrate. The double-reciprocal plots of the initial rate versus concentration of the varied substrate exhibited families of parallel lines as shown in Fig. 4. These results indicate that the catalytic mechanism of recombinant hGPx4 is similar to that of GPx1, which follows a ping–pong mechanism. The steady-state rate equation for this mechanism is given in the following equation :
In steady-state kinetics equation, V0 is the initial velocity of the enzymatic reaction, [E]0 is the total concentration of the enzyme, k+1 is the rate constant for the reaction of the reduced enzyme with the H2O2, and is the forward rate constant for the reduction of the oxidized enzyme by GSH. The rate constants k+1 and are experimentally available from the steady-state kinetics and calculated from the slope and intercept, respectively. The k+1 value for H2O2 obtained from recombinant hGPx4 is 2.9 × 106 M−1 min−1, and the value for GSH is 1.2 × 106 M−1 min−1.
Successful expression of selenoprotein is difficult because of the intricate and low-efficient translation machinery for the incorporation of Sec into protein. Nevertheless, some prior reports have described expression methods of mammalian selenoproteins. One such method is using cysteine auxotrophic strain system, and all the cysteine in the protein will be replaced by Sec [17, 21]. As a result, the unavoidable structural changes led to functional loss of the selenoprotein expressed . Another mostly used method is using the mammalian cell lines such as HEK293 and HepG2 as in the eukaryotic expression system [23, 24]. Success in expressing selenoproteins using this method depends mostly on the inherent characteristics of the host cell line. However, low translation efficiency makes it difficult to get the sufficient quantity of recombinant selenoprotein. To achieve high-level, constitutive expression of recombinant selenoprotein, Novoselov et al.  developed the expression vector pSelExpress1, which contains AUGA form of Toxoplasma SelT SECIS element in the 3′-UTR of the target gene. Most importantly, the vector also constitutively expresses the C-terminal functional domain of rat SBP2. As we know, the SECIS element in the 3′-UTR and the SBP2 are both essential for the recognition of UGA as Sec rather than terminal codon in mammalian selenoprotein. Besides that element, there are many more cis- and trans-acting factors from the host cell that participate in the regulation of Sec incorporation event. Using this vector, selenium-containing glutathione transferase has been successfully expressed in the HEK293T cell line. The murine Ig κ-chain leader sequence in the N-terminal and the His-tag in the C-terminal of the selenoprotein expressed make it convenient to isolate and purify selenoprotein from the cell culture supernatant [13, 14].
To our knowledge, this was the first time to express and characterize an enzymatically active recombinant hGPx4 from an expression system based on HEK293T cell line using pSelExpress1 vector. So far, more than 40 species of selenoproteins have been discovered. GPx family is an important class of antioxidant selenoenzymes . As one of the eight GPxs in mammals, GPx4 plays a major role in the regulation of phospholipid hydroperoxide levels and is the sole monomeric enzyme . GPx4 was first purified from pig liver in 1982 by Ursini et al.  and characterized as a protein efficiently protecting liposomes and biomembranes from peroxidative degradation in the presence of GSH. Almost 30 years after the discovery of GPx4 and a huge number of published studies, this enzyme is still hardly to reach the sufficient quantity expression in mammalian cells for purification and characterization of catalytic activity and mechanism. The intricate and low-efficient translation machinery for the incorporation of Sec into protein makes it difficult to study the function of GPx4 in vitro. Previous functional studies depend on knocking down the intracellular GPx4 by RNA interference or the overexpression of intracellular GPx4 [5-7, 27] by transfecting cells with plasmids encoding GPx4. These methods could hardly be used in the enzymatic properties study of GPx4 and the preparation of medicinal GPx4. Because of the important role of GPx4, a large number of reports focus on the methods of designing proteins or small molecules to imitate GPx function. The major strategy for generating a GPx mimic needs the creation of GSH substrate binding sites and the introduction of catalytic group [28, 29]. It is difficult to simultaneously satisfy both conditions. There would be no such problems if we could express the hGPx4 as it naturally contains the substrate binding site and the catalytic site. Our results show that recombinant hGPx4 has GPx activity.
Furthermore, it is important to elucidate the catalytic mechanism for recombinant hGPx4. As we know, the catalytic mechanism is similar in all GPxs, involving redox shuttling of the Sec residue within the active site . The dissociated selenol from Sec is oxidized by hydroperoxides to yield a selenenic acid. The selenenic acid reacts with a free thiol, typically GSH, to form an intermediate selenodisulfide, which in turn is resolved by a second GSH molecule to yield a selenol anion. In other words, the reaction comprises two independent events: oxidation of the reduced enzyme by hydroperoxides and reduction of the oxidized enzyme by GSH. The rate constant k+1 and are for the oxidative reaction and the sum of the two reductive steps of the catalytic cycle, respectively. In this study, we tested if the recombinant hGPx4 had the catalytic process similar to that of native GPx when the substrates GSH and H2O2 are provided. The steady-state kinetic analysis of the recombinant hGPx4 indicates that the rate constant is 1.2 × 106 M−1 min−1, similar to that of natural GPx4 purified from human placental cytosol (3.4 × 106 M−1 min−1) . This finding suggested that the recombinant hGPx4 has almost same affinity for GSH to that of native GPx4. To date, the rate constant k+1 for H2O2 of human resource GPx4 has not been reported when H2O2 and GSH were provided as substrates. In this work, the rate constant k+1 for H2O2 was calculated to be 2.9 × 106 M−1 min−1, which is much lower than that of well-known radical scavenger GPx1 . This indicates that hGPx4 is less reactive to H2O2 than GPx1. The result is not surprising because of the broad substrate specificity of GPx4 .
In conclusion, we achieved the overexpression and purification of recombinant hGPx4 in eukaryotic cell and the further characterization of enzymatic properties. In the process of our experiments, the recombinant hGPx4 was found to exhibit a strong tendency toward protein polymerization. Mass spectral analysis of tryptic digest fragments of native GPx4 polymers suggested that Sec46 might play a major role in the polymerization process . Crystallographic data of GPx4 (Sec46Cys) indicated that Cys10, Cys46, and Cys66 are located at monomer–monomer interfaces, and it has been suggested that the surface-exposed cysteines play a role in the polymerization . In the future, we will try to prevent polymerization and to improve the catalytic activity of recombinant hGPx4 through computer simulations and site-directed mutagenesis on the basis of this research.
The authors thank Professor Vadim N. Gladyshev for providing pSelExpress1. This work is supported by the Fundamental Science Funds of Jilin University (No. 450060445268) and the National Natural Science Funds, China (Nos. 30970633, 31270851).