Autoantibodies against a serine tRNA-protein complex implicated in cotranslational selenocysteine insertion.

Abstract

We describe an autoantibody specificity present in a subgroup of patients with a severe form of autoimmune chronic active hepatitis. These antibodies precipitate a 90-nucleotide RNA from human whole cell extracts and recognize a 48-kDa polypeptide in immunoblotting assays. The RNA is a UGA suppressor serine tRNA that carries selenocysteine (tRNAlserJs'¢), as shown by sequence analysis. The protein does not appear to be seryl-tRNA synthetase; rather, it is an excellent candidate for a factor involved in cotranslational selenocysteine incorporation in human cells.Autoantibodies reactive with cellular constituents are often detected in sera from patients with autoimmune disease. Autoantibodies that immunoprecipitate tRNA have been detected primarily in sera from patients with polymyositis (1-3) and, less frequently, in patients with systemic lupus erythematosus (4) and scleroderma (5). The reactive antigens are usually tRNA-associated proteins; in only a few cases are they the tRNAs themselves (4, 6). Some of these protein antigens have been identified as aminoacyl-tRNA synthetases (3, 6-10).Other well-characterized components of the protein synthetic machinery [for example, ribosomes (11), 5S ribonucleoprotein (12), and elongation factor la (13)] also act as autoantigens. One translational event, the incorporation of selenocysteine into a growing polypeptide chain, is not well-understood in eukaryotes, and to our knowledge, the factors involved in this process have not been implicated in autoimmunity. The synthesis of selenoproteins in Escherichia coli involves several reactions (for review, see ref. 14). Briefly, selenocysteine incorporation is cotranslational and requires an unusual suppressor aminoacyl-tRNA (selenocysteyl-tRNAsec, where Sec is selenocysteine) and a specific translation factor (the product of the selB gene) that substitutes for elongation factor Tu. Insertion is directed by certain UGA triplets that in other contexts act as termination codons. tRNASec is initially charged with serine to form seryltRNASec and is then converted to selenocysteyl-tRNAsec by the action of the selA (selenocysteine synthase) and selD gene products. In mammalian cells, selenocysteine incorporation also occurs at specific UGA codons, and a suppressor serine tRNA that carries selenocysteine (tRNA[Serlsec) has been detected (15). However, complete elucidation of the eukaryotic pathway (and identification of the factors involved) has not yet been achieved.Here we report that an RNA-protein complex is the target of an autoantibody specificity found in the sera of a subset of patients with autoimmune chronic active hepatitis (AI-CAH). The RNA corresponds to a previously identified human suppressor serine tRNA, and the antigenic protein may be a factor involved in the pathway of cotranslational incorporation of selenocysteine into specific polypeptide chains.MATERIALS AND METHODS Subjects. Sera from 35 patients with AI-CAH were collected over a period of 4 years at the Sant Pau and Valle Hebron Hospitals in Barcelona. ...

Identification of a protein component of a mammalian tRNASec complex implicated in the decoding of UGA as selenocysteine

“…Northern blot analysis verifies that tRNA Sec is the 90-nt RNA that copurifies with SECp43+ A human tRNA Sec -specific probe was generated by PCR and used to probe a blot containing RNA fractions from the specific antibody column (lane 6) or control BSA column (lane 5)+ Arrowhead marks the position of the 90-nt RNA X+ A sucrose gradient fraction of HeLa S100 extract (lane Fr 8) contains bulk tRNA+ Ab and BSA column samples identical to those used in lanes 5 and 6 were labeled with 32 pCp as shown at left (lanes 1,2)+ Lanes MW: pBR322 MspI digest+ FIGURE 5. Immunoaffinity selection of tRNA Sec is blocked in the presence of excess recombinant SECp43, but not by a control RNAbinding protein+ Top panel: 32 pCp-labeled RNA species eluted from specific antibody columns (Ab) or control BSA columns (BSA)+ The columns were pretreated with buffer (lanes Ϫ), excess control protein (U2AF), or excess SECp43 as indicated (top)+ Bottom panel: Northern blot of samples of top panel performed with a DNA probe specific for tRNA Sec + Arrowheads mark position of tRNA Sec + elements and charged tRNA Sec + SECp43 might function as such an adaptor+ Here, protein overlay analysis demonstrates a specific association of recombinant SECp43 with a 48-kDa protein species in HeLa S100 extracts+ It is likely that the 48-kDa SECp43-interacting protein observed in the present study is identical to the 48-kDa antigen detected by patient autoantibodies, which was shown in a previous report to immunoprecipitate tRNA Sec (Gelpi et al+, 1992)+ We have attempted to test this possibility; however, the patient antibodies are no longer available+ It is also intriguing to consider the suggestion of Shen et al+ (1998) that the SECISinteracting protein dbpB might be the same as the 48-kDa autoantigen reported in the Gelpi et al+ (1992) study+ If the 48-kDa protein observed in the present study is the same as that reported in these previous studies, this would point to an adaptor role for SECp43 in the mechanism of selenocysteine insertion+ An alternative possibility is that the 48-kDa protein observed in the experiment of Figure 6 is a molecular bridge that supports the interaction of tRNA Sec and SECp43+ SECp43 and its related antibody should prove to be useful reagents to test these ideas+…”

“…Current evidence supports a model in which the recognition of UGA as a selenocysteine codon, rather than a stop codon, is dependent upon the presence of mRNA secondary structures termed selenocysteine insertion sequence (SECIS) elements (Heider et al+, 1992; for review see Stadtman, 1996;Low & Berry, 1996;Atkins et al+, 1999)+ SECIS elements are RNA stemloop structures found immediately downstream of the selenocysteine-specific UGA codon in bacterial mRNAs, whereas, in mRNAs of eukaryotes and archaebacteria, SECIS elements are located in the 39 untranslated region, frequently at a considerable distance from the selenocysteine codon+ An outline of the biochemical pathway for selenocysteine insertion in bacteria has been elucidated by genetic and biochemical studies in which the products of the selA, sel B, sel C, and sel D genes are shown to be required for the insertion of selenocysteine at the di-rection of SECIS elements (Leinfelder et al+, 1988a(Leinfelder et al+, , 1988b(Leinfelder et al+, , 1989Forchhammer et al+, 1989Forchhammer et al+, , 1990)+ A key feature of this pathway involves the function of a specialized selenocysteine tRNA (tRNA Sec ; product of the sel C gene), which is initially charged with serine and subsequently converted to selenocysteyl-tRNA Sec (Baron & Böck, 1995)+ In correspondence to these factors, only tRNA Sec and the selenophosphate synthetase gene (homolog of sel D) have been identified in eukaryotes (Lee et al+, 1989;Low et al+, 1995)+ An important unanswered question is how, for eukaryotes, do the SECIS elements in the 39 untranslated region signal, at a distance, the cotranslational incorporation of selenocysteine at UGA codons? In particular, an understanding of the molecular link between tRNA Sec and SECIS RNA elements has remained elusive, in part, because the eukaryotic homolog of SELB has not yet been identified+ In a model proposed by Ringquist et al+ (1994) it is held that SELB delivers bacterial selenocysteyl-tRNA Sec to a ribosome-SECIS RNA complex+ This model implicitly accounts for the recognition of a selenocysteine codon by coupling the required SECIS RNA recognition event to the delivery of the charged tRNA+ In eukaryotes, the known elongation factor, EF-1a, fails to deliver selenocysteyltRNA Sec to the ribosome, and a partially characterized factor is implicated for this role (Jung et al+, 1994;Yamada, 1995)+ Recent progress has been made in the identification and cloning of a mammalian protein, dbpB, which recognizes SECIS RNA elements (Shen et al+, 1998)+ Nonetheless, a variety of proteins that have been detected in association with mammalian SECIS RNA elements, or by autoimmune antibodies that recognize tRNA Secprotein complexes, have yet to be fully characterized (Gelpi et al+, 1992;Shen et al+, 1995;…”

B cells and monocytes are not developmentally affected in a case of reticular dysgenesis

“…Immunoprecipitations of small RNAs were performed as described [12]. Immunoprecipitated 32 P-labelled RNAs were extracted, ethanol-precipitated, electrophoresed on polyacrylamide denaturing gels and subjected to autoradiography.…”

New Developments in Selenium Biochemistry: Selenocysteine Biosynthesis in Eukaryotes and Archaea

“…Each search resulted in known components of the Sec insertion machinery and an additional protein that might be SecS in mammals. This additional protein had previously been detected in patients with autoimmune chronic hepatitis as an autoimmune factor that co-precipitated tRNA [Ser]Sec in cell extracts of these individuals and was known as the SLA [19]. SLA, which is a PLPdependent transferase [20], also bound other protein components involved with Sec metabolism [22,23].…”

“…In addition, because no sequences with homology to SelA could be detected in eukaryotes, there was some question as to the identity of the Sec synthase (SecS) in mammals. The soluble liver antigen (SLA), which was initially identified as a 48-kDa protein bound to Sec tRNA [Ser]Sec and was targeted by antibodies in patients with an autoimmune chronic hepatitis [19], was reported to occur as a separate family within a larger superfamily of diverse PLP-dependent transferases [20] and proposed as a possible SecS in mammals [3,[20][21][22]. SLA was also found to exist in a protein complex with other factors involved in Sec biosynthesis and/or its insertion into protein providing further evidence that this protein is involved in selenium metabolism [22,23].…”