Báo cáo khoa học: Expression, post-translational modification and biochemical characterization of proteins encoded by subgenomic mRNA8 of the severe acute respiratory syndrome coronavirus

Expression, Post-Translational Modification, and Biochemical Characterization of Proteins Encoded by Subgenomic mRNA8 of the Severe Acute Respiratory Syndrome Coronavirus

Tác giả: Tra M. Le, Hui H. Wong, Felicia P. L. Tay, Shouguo Fang, Choong-Tat Keng, Yee J. Tan, and Ding X. Liu

Lĩnh vực: Sinh học phân tử, Vi sinh vật học

Nội dung tài liệu:
Nghiên cứu này tập trung vào việc phân tích các protein được mã hóa bởi mRNA8 của virus gây hội chứng hô hấp cấp tính nặng (SARS-CoV). mRNA8 của SARS-CoV, đặc biệt là ở các chủng virus lưu hành ở người, có sự biến đổi do mất 29 nucleotide, dẫn đến việc chia một khung đọc mở (ORF) duy nhất thành hai ORF chồng chéo (ORF8a và ORF8b). Các ORF này được dự đoán sẽ mã hóa hai protein nhỏ, protein 8a và 8b, hoặc một protein dung hợp, protein 8ab. Nghiên cứu đã tiến hành biểu hiện các protein này trong các hệ thống in vitro và in vivo, đồng thời khám phá các quá trình biến đổi sau dịch mã như glycosyl hóa và ubiquitination. Kết quả cho thấy protein 8ab trải qua quá trình N-linked glycosylation tại gốc N81 và ubiquitination. Protein 8b, khi thiếu vùng 8a, có xu hướng bị phân hủy nhanh chóng bởi proteasome. Nghiên cứu cũng chỉ ra rằng protein 8b và 8ab có khả năng liên kết với monoubiquitin và polyubiquitin, gợi ý vai trò tiềm tàng của chúng trong cơ chế sinh bệnh của SARS.

Mục lục chi tiết:

• Expression, post-translational modification and biochemical characterization of proteins encoded by subgenomic mRNA8 of the severe acute respiratory syndrome coronavirus
• Keywords
• Correspondence
• Abbreviations
• The most striking difference between the subgenomic mRNA8 of severe acute respiratory syndrome coronavirus isolated from human and some animal species is the deletion of 29 nucleotides, resulting in splitting of a single ORF (ORF8) into two ORFS (ORF8a and ORF8b). ORF8a and ORF8b are predicted to encode two small proteins, 8a and 8b, and ORF8 a single protein, 8ab (a fusion form of 8a and 8b). To understand the functions of these proteins, we cloned cDNA fragments covering these ORFS into expression plasmids, and expressed the constructs in both in vitro and in vivo systems. Expression of a construct containing ORF8a and ORF8b generated only a single protein, 8a; no 8b protein expression was obtained. Expression of a construct containing ORF8 generated the 8ab fusion protein. Site-directed mutagenesis and enzymatic treatment revealed that protein 8ab is modified by N-linked glycosylation on the N81 residue and by ubiquitination. In the absence of the 8a region, protein 8b undergoes rapid degradation by proteasomes, and addition of proteasome inhibitors inhibits the degradation of protein 8b as well as the protein 8b-induced rapid degradation of the severe acute respiratory syndrome coronavirus E protein. Glycosylation could also stabilize protein 8ab. More interestingly, the two proteins could bind to monoubiquitin and polyubiquitin, suggesting the potential involvement of these proteins in the pathogenesis of severe acute respiratory syndrome coronavirus.
• Coronaviruses are important pathogens of human and other animal species. In 2003, a novel coronavirus [severe acute respiratory syndrome coronavirus (SARS-CoV)] was discovered to be the etiologic agent of severe acute respiratory syndrome [1]. In SARS-CoV-infected cells, nine mRNA species are produced, which encode four typical structural proteins, spike (S), membrane (M), envelope (E) and nucleocapsid (N) proteins. In addition, protein 3a, encoded by the first ORF of the subgenomic mRNA3, was recently found to be a minor structural protein [2]. Among other mRNA species, the genome-length mRNA1 codes for approximately 16 functional proteins involved in viral RNA replication, and subgenomic mRNA3, mRNA6, mRNA7, mRNA8 and mRNA9 code for five to eight accessory proteins that share little sequence homology with proteins of other known coronaviruses [1]. Among the accessory proteins, proteins 3a, 7a and 9b have been shown to be expressed, and some of them can elicit antibody responses in patients [3–5]. Although most of these
• Proteins encoded by SARS-CoV mRNA8
• Fig. 1. Diagram showing the six constructs, pF-8a, pF-8b, pF-8a/8b, p8a/b, pF-8ab and pF-8abM, used in this study. The nucleotide positions of ORF8a are shown in bold, and the positions of ORF8b are underlined. Also shown are the nucleotide sequence of the 29 nucleotide insertion, the position of the insertion, the numbers of amino acids for the putative proteins 8a and 8b, and the Flag tag at the N-terminus of each construct.
• accessory proteins are dispensable for viral replication in cultured cells [6], their exact roles in the pathogenesis and virulence of SARS-CoV in its natural hosts are yet to be established.
• Two such accessory proteins, proteins 8a and 8b, are encoded by two overlapping ORFs (8a/b) present in subgenomic mRNA8 (Fig. 1). Subgenomic mRNA8 from most SARS-CoV isolates obtained from animals and patients at early stages of the SARS epidemic contains a single ORF8 with potential to encode a single protein, 8ab [7]. A 29 nucleotide deletion between T27867 and A27868 for strain SG2774 (accession number AY283798) was found in strains collected from human isolates at late stages of the outbreak, resulting in the splitting of the single ORF8 into two overlapping ORFs, ORF8a and ORF8b (Fig. 1) [1,8]. The deletion of the 29 nucleotides and the formation of two overlapping ORFs were predicted to be the consequence of adaptation of SARS-CoV from animals to humans [7,9]. Experimental infection of civets with SARS-CoV isolates BJ01 (without the 29 nucleotides) and GZ01 (with the 29 nucleotides) showed that animals infected with strain BJ01 tend to have a higher average temperature and slightly stronger antibody responses [10]. Two recent studies have shown that the putative protein 8b can downregulate the SARS-CoV E protein in virus-infected cells and induce DNA synthesis in cells expressing the protein [11,12]. However, more detailed characterization of the expression, biochemical properties and functions of the products encoded by these two mRNA isoforms has yet to be done.
• In this study, ORF8a was expressed in vitro in both wheat germ extracts and rabbit reticulocyte lysates as a 5.3 kDa protein. Expression of the protein was also observed in Cos-7 cells, as a Flag-tagged protein. Similarly, translation of ORF8ab was observed in all the translation systems used. In contrast, protein 8b was expressed only when ORF8b was separately cloned. When constructs containing the two overlapping ORFs ORF8a/ORF8b and the single ORF8ab, respectively, were expressed, expression of proteins 8a and 8ab were observed. Protein 8ab was shown to be an N-linked glycosylated protein, and the glycosylation site was identified to be the N81 residue. Proteins 8b and 8ab could be modified by ubiquitination, and in the absence of the 8a region, protein 8b undergoes rapid degradation by proteasome. Addition of the proteasome inhibitors inhibits the degradation of protein 8b as well as the protein 8b-induced rapid degradation of the SARS-CoV E protein. In addition, glycosylation could also stabilize protein 8ab. Furthermore, proteins 8b and 8ab could bind covalently and noncovalently to monoubiquitin and polyubiquitin. As no homology with any known ubiquitin-binding domains (UBDs) was found, they may represent a novel group of ubiquitin-binding proteins.
• Results
• Cloning, expression and post-translational modification of proteins encoded by the SARS-CoV mRNA8
• In some animal and early human isolates, the subgenomic mRNA8 of SARS-CoV was predicted to encode a single ORF (ORF8). Owing to the deletion of 29 nucleotides [between T27867 and A27868 for strain SG2774 (accession number AY283798)], two overlapping ORFs (ORF8a/ORF8b) were found in most human isolates (Fig. 1). ORF8a and ORF8b are predicted to encode two small proteins, 8a and 8b, whereas ORF8 encodes a single protein, 8ab, representing a fused form of proteins 8a and 8b. To understand the functions of these proteins, cDNA fragments covering these ORFs were cloned into pFlag, giving rise to five constructs either with or without a Flag-tag at the N-terminus (Fig. 1). These constructs were then expressed in both in vitro expression systems and in intact cells.
• Fig. 2. Expression of constructs covering the 5′-unique ORFs of the subgenomic mRNA8 of SARS-CoV. (A) Expression of p8a/b (lane 1), pF-8a/b (lane 2), pF-8b (lane 3) and pF-8ab (lane 4) in wheat germ extracts in the presence of [35S]methionine. The in vitro-synthesized products were resolved on SDS/20% polyacrylamide gel and detected by autoradiography. The numbers on the left indicate molecular masses in kilodaltons. (B) Expression of p8a/b (lanes 1 and 5), pF-8a/b (lanes 2 and 6), pF-8b (lanes 3 and 7) and pF-8ab (lanes 4 and 8) in rabbit reticulocyte lysates in the presence of [35S]methionine. The in vitro-synthesized products were resolved on SDS/20% polyacrylamide gel either directly (lanes 1-4) or after immunoprecipitation with polyclonal antibodies to protein 8b (lanes 5-8). Polypeptides were detected by autoradiography. The numbers on the left indicate molecular masses in kilodaltons. (C) Expression of p8a/b (lane 1), pF-8a/b (lane 2), pF-8b (lane 3) and pF-8ab (lane 4) in Cos-7 cells. Cells were infected with the recombinant vaccinia/T7 virus, transfected with each of the four constructs, and harvested at 18 h post-transfection. Polypeptides were resolved on SDS/20% polyacrylamide gel, and analyzed by western blot with antibody to Flag. The numbers on the left indicate molecular masses in kilodaltons.
• When constructs p8a/b, pF-8a/b, pF-8b and pF-8ab were expressed in TnT transcription-coupled translation wheat germ extracts in the presence of [35S]methionine, single protein bands of approximately 5.3, 6.5, 10.2 and 14.4 kDa, representing untagged protein 8a, Flag-tagged protein 8a, Flag-tagged protein 8b and Flag-tagged protein 8ab, respectively, were detected (Fig. 2A, lanes 1-4). Expression of the same four constructs in rabbit reticulocyte lysates in the presence of [35S]methionine gave rise to the same four products (Fig. 2B, lanes 1-4). In addition, a series of bands (ladder bands) with increases of approximately 10 kDa were detected when pF-8b and pF-8ab were expressed in the system (Fig. 2B, lanes 3 and 4). The patterns of these bands suggest that they may represent ubiquitinated forms of proteins 8b and 8ab. To determine whether these bands are related to the 8b region, immunoprecipitation was carried out using rabbit polyclonal antibodies to protein 8b. As shown in Fig. 2B, the 10.2 kDa protein 8b and the 14.4 kDa protein 8ab, together with their corresponding ladder bands, were precipitated with the antibodies (lanes 7 and 8). The fact that these bands were efficiently immunoprecipitated by the antibodies to protein 8b suggests that proteins 8b and 8ab may be modified by ubiquitination.
• Expression of these constructs was then carried out in HeLa cells using the vaccinia/T7 expression system. As shown in Fig. 2C, Flag-tagged proteins 8a, 8b and 8ab were detected in cells transfected with the corresponding constructs (lanes 2–4). Interestingly, a prominent band (8ab*), migrating more slowly than the 14.4 kDa protein 8ab, was detected in cells expressing pF-8ab (Fig. 2C, lane 4). This may represent a post-translationally modified protein 8ab; characterization of this modification was conducted, and the results are presented in a later section. In addition, multiple species of more slowly migrating bands were also detected from cells expressing F-8b and F-8ab (Fig. 2C, lanes 3 and 4).
• Ubiquitination and binding of proteins 8b and 8ab to monoubiquitin and polyubiquitin
• To test further whether proteins 8b and 8ab are post-translationally modified by ubiquitination, coexpression of these proteins with a Myc-tagged ubiquitin in HeLa cells was carried out, and the expression and interaction of the proteins were determined by western blot and coimmunoprecipitation assays. As shown in Fig. 3A, western blot analysis using antibody to Myc resulted in the detection of monoubiquitin and
• Fig. 3. Binding of proteins 8b and 8ab to monoubiquitin and polyubiquitin. (A) Cos-7 cells were transfected with the Myc-tagged ubiquitin (lanes 3, 6 and 9), the Myc-tagged ubiquitin together with the Flag-tagged protein 8b (lanes 1, 4 and 7), and the Myc-tagged ubiquitin together with the Flag-tagged protein 8ab (lanes 2, 5 and 8). Total cell lysates were prepared, separated on SDS/20% polyacrylamide gel, and analyzed either directly by western blot (lanes 1-3) or after coimmunoprecipitation with antibody to Flag (lanes 4-9). The numbers on the left indicate molecular masses in kilodaltons. (B) Cos-7 cells were transfected with the Myc-tagged ubiquitin (lanes 1 and 5), the Flag-tagged protein 8ab (lanes 4 and 8), the Myc-tagged ubiquitin together with the Flag-tagged protein 8b (lanes 2 and 6), and the Myc-tagged ubiquitin together with the Flag-tagged protein 8ab (lanes 3 and 7). Total cell lysates were prepared, separated on SDS/20% polyacrylamide gel, and analyzed either directly by western blot (lanes 1-4) or after coimmunoprecipitation with antibody to Myc (lanes 5-8). The numbers on the left indicate molecular masses in kilodaltons. (C) Pull-down of monoubiquitin and polyubiquitin by GST-8b and GST-8ab fusion proteins. GST, GST-8b and GST-8ab were purified from bacteria, separated on SDS/12% polyacrylamide gel, and visualized by staining with Comassie blue (lanes 1-3). Cos-7 cells expressing the Myc-tagged ubiquitin were labeled with [35S]methionine. Total cell lysates were prepared and precleared by incubation with GST beads at 4 °C for 2 h, and then incubated with beads prebound with GST, GST-8b, and GST-8ab, respectively, at 4 °C for 2 h. After washing three times with RIPA buffer, polypeptides were eluted from the beads by adding 30 µL of the loading dye and boiling for 5 min. The total cell lysates (lane 4) and the eluted polypeptides (lanes 5-7) were separated on 20% SDS/20% polyacrylamide gel and visualized by autoradiography. The numbers on the left of each panel indicate molecular masses in kilodaltons.
• polyubiquitin in cells expressing the Myc-tagged ubiquitin alone (Fig. 3A, lane 3). The monoubiquitin was not detected in cells coexpressing the Myc-tagged ubiquitin with either protein 8b (Fig. 3A, lane 1) or protein 8ab (Fig. 3A, lane 2). Instead, weak polyubiquitin bands were observed (Fig. 3A, lanes 1 and 2). Immunoprecip-
• itation of cell lysates prepared from these transfected cells with antibody to Flag led to the detection of massive free and ubiquitin-conjugated protein 8ab by western blot with antibody to Flag (Fig. 3A, lane 5). However, much lower amounts of protein 8b were detected in cells coexpressing ubiquitin and protein 8b (Fig. 3A, lane 4). Western blot analysis of the same precipitates with antibody to Myc resulted in efficient detection of polyubiquitin bands in cells coexpressing the Myc-tagged ubiquitin and the Flag-tagged protein 8ab (Fig. 3A, lane 8). Much weaker detection of polyubiquitin bands, however, was obtained with cells coexpressing the Myc-tagged ubiquitin and the Flag-tagged protein 8b by western blot analysis with antibody to Myc (Fig. 3A, lane 7). These results strongly suggest that proteins 8b and 8ab are modified by ubiquitination.
• In a similar way