Human Molecular Genetics, 2003, Vol. 12, No. 9 1045-1053
DOI: 10.1093/hmg/ddg115
© 2003 Oxford University Press
Selenoprotein N: an endoplasmic reticulum glycoprotein with an early developmental expression pattern
1INSERM U582, Institut de Myologie, GH Pitié-Salpêtrière, Paris, France, 2UPR 9002 CNRS, Institut de Biologie Moléculaire et Cellulaire, Strasbourg, France and 3Institute of Molecular Pathology, University of Copenhagen, Copenhagen, Denmark
Received January 2, 2003; Accepted February 27, 2003
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONRESULTSDISCUSSIONMATERIALS AND METHODSREFERENCES |
|---|
Rigid spine muscular dystrophy and the classical form of multiminicore disease are caused by mutations in SEPN1 gene, leading to a new clinical entity referred to as SEPN1-related myopathy. SEPN1 codes for selenoprotein N, a new member of the selenoprotein family, the function of which is still unknown. In a previous study, two isoforms were deduced from SEPN1 transcript analyses. Using polyclonal antibodies directed against SEPN1 and cDNA constructs encoding for the two isoforms, we show that the main SEPN1 gene product corresponds to a 70 kDa protein, containing a single selenocysteine residue. Subcellular fractionation experiments and endoglycosidase H sensitivity indicate that SEPN1 is a glycoprotein-localized within the endoplasmic reticulum. Immunofluorescence analyses confirm this subcellular localization and green fluorescent protein fusion experiments demonstrate the presence of an endoplasmic reticulum-addressing and -retention signal within the N-terminus. SEPN1 is present at a high level in several human fetal tissues and at a lower level in adult ones, including skeletal muscle. Its high expression in cultured myoblasts is also down-regulated in differentiating myotubes, suggesting a role for SEPN1 in early development and in cell proliferation or regeneration.
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONRESULTSDISCUSSIONMATERIALS AND METHODSREFERENCES |
|---|
Recently, we identified mutations in SEPN1 by two independent genetic studies performed on informative families presenting with recessive muscular disorders (1). The probands were assigned either to a class of congenital muscular dystrophy called rigid spine muscular dystrophy (RSMD) (1) or to a congenital myopathy, the ‘classical form’ of multiminicore Disease (MmD) (2). From these studies, a new entity referred to as SEPN1-related myopathy was defined. It is characterized by an early onset of hypotonia and weakness, with predominant axial muscle impairment leading to life-threatening respiratory failure and scoliosis. A variable degree of spinal rigidity was observed in most of the patients. In contrast, the morphological features of the muscle biopsies were rather heterogeneous. Most of them exhibited myopathic changes with variation in fiber size, a limited increase in connective tissue and the presence of multiple, poorly circumscribed, short-length areas of sarcomere disorganization and mitochondria depletion known as minicores. Only some of the biopsies displayed a typical dystrophic pattern, with necrosis and regeneration of muscle fibers, and severe endomysial fibrosis. Other patient biopsies had an intermediate pattern combining morphological characteristics of both groups (2). All the biopsies exhibited a type I fiber predominance.
Selenoprotein N (SEPNl) is the first selenoprotein shown to be responsible for a genetic disorder (1,2). This protein, identified by Lescure et al. (3) contains a selenium atom in the form of a selenocysteine residue, which is characteristic of the selenoprotein family (see 4 for review). Incorporation of this amino acid occurs at a redefined UGA codon and requires the involvement of a stemloop structure formed by the SECIS sequence (for selenocysteine insertion sequence) located in the 3'-UTR of selenoprotein mRNA (5,6). The function of several selenoproteins has been identified: they are all enzymes involved in oxidation–reduction reactions with the selenocysteine residue located in the catalytic center. Indeed the selenol group, ionized at physiological pH, confers more reactivity than the thiol group of cysteine. Although analyses of the amino acid sequence of SEPN1 have revealed a calcium-binding site and a motif similar to that observed in the catalytic site of thioredoxin reductases, another selenoprotein subgroup (7), no function has been assigned to SEPN1 yet.
In a previous study, we established the structure of the human SEPN1 gene, which contains 13 exons spanning 18.5 kb (1). It produces a 4.5 kb transcript with an open reading frame of 1770 nucleotides encoding a 590 amino acid protein. Two isoforms of the SEPN1 gene product were predicted from EST database query and RT–PCR experiments (1). Isoform 1 corresponds to the full-length transcript, whereas exon 3 is spliced out in isoform 2. Both transcripts are detected in skeletal muscle, brain, lung and placenta, with isoform 2 being the predominant transcript. The exon 3 sequence corresponds to an alu cassette and contains a second potential in-frame selenocysteine codon.
In this study, we describe the first characterization and localization of the SEPN1 gene product. With this aim, we developed polyclonal antibodies directed against either the central domain or the C-terminus of SEPN1. We show that a single protein of 70 kDa is detected, and corresponds to the isoform 2 transcript expression. Further experiments demonstrate that SEPN1 is an integral membrane glycoprotein, which localizes in the endoplasmic reticulum. Finally, analyses of the protein distribution reveal a strong expression in fetal tissues and proliferating cells. These results raise the possible involvement of SEPN1 in early muscle formation, in agreement with the clinical features of patients presenting SEPN1-related myopathy.
|
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONRESULTSDISCUSSIONMATERIALS AND METHODSREFERENCES |
|---|
The major SEPN1 gene product is a 70 kDa protein
Two polyclonal antisera were raised against two distinct synthetic peptides corresponding to the common C-terminal region encoded by both SEPN1 transcripts (Fig. 1). They were purified by affinity chromatography, using the immunogenic peptide for antibody 137 or a larger C-terminal region for antibody 143. The specificity of antibodies was first verified by western blot analysis of fibroblasts from controls and RSMD1 patients. A single band of
70 kDa was revealed with antibodies 137 (Fig. 2A, lane 1) and 143 (data not shown). SEPN1 was weakly expressed in HeLa cells (Fig. 2A, lane 3). The 70 kDa band was not detected in fibroblast extracts from a patient with a homozygous 1 bp deletion at position 1446 resulting in a frameshift in SEPN1 at position L482 (Fig. 2A, lane 2). To determine which isoform is expressed, cDNAs encoding both isoforms were transfected into HeLa cells and the cell extracts subjected to western blot analysis. The HeLa endogenous SEPN1 protein was no longer detected due to the decreased amounts of total proteins which were applied (Fig. 2A, lane 4). Antibody 137 revealed a strong signal at 70 kDa in cells transfected with pXJSelN6 expressing isoform 2, which lacks exon 3 (Fig. 2A, lane 6). However, no signal could be detected in cells transfected with pXJSelN7 expressing the full-length isoform 1 (Fig. 2A, lane 5). The same transfection lysates were subjected to immunoblotting with antibody 168, raised against a central domain of SEPN1 (Fig. 2B). No signal could be seen in cells transfected with pXJSelN7, whereas two bands of 70 and 60 kDa were observed in cells transfected with pXJSelN6 (Fig. 2B, lane 6). Since the 60 kDa band was not detected with antibody 137, directed against the C-terminal region, this polypeptide is likely to arise from a premature termination of translation at the UGA selenocysteine codon. Such a phenomenon has been reported in previous studies (3,8) and may be attributed to saturation of the selenoprotein translation machinery under mRNA overexpression conditions. We were unable to detect the endogenous SEPN1 in fibroblasts and HeLa cells with antibody 168. As this antibody was generated from a larger protein domain, it may interact with a structural epitope, making the recognition more dependent on the correct folding of the protein. Altogether, the results indicate that the major SEPN1 gene product is a 70 kDa protein containing a single selenocysteine residue.
|
|
SEPN1 localizes in the endoplasmic reticulum
To gain insight into the localization of SEPN1, subcellular fractionation was performed on control fibroblast cultures. The protocol used allows the enrichment of nuclei in the P1 pellet, mitochondria in the P2 pellet, Golgi membranes in the P3 pellet, and endoplasmic reticulum in the P4 pellet (9). The final S4 supernatant contains the soluble cytosolic proteins. The fractions were analysed by SDS–PAGE and immunoblotted with the purified 137 antiserum (Fig. 3A). The SEPN1 protein was detected in the crude homogenate and in the four pellets, but not in the final S4 supernatant. The quality of the fractionation was controlled by the application of various antibodies raised against other compartment-specific marker proteins: emerin, a transmembrane protein of the nuclear envelope; hsp60, a mitochondrial protein; calnexin, a transmembrane protein of the endoplasmic reticulum (ER); and p23, a protein of the cis-Golgi. SEPN1 displayed a distribution similar to calnexin, as well as a co-sedimentation with p23 in the P3 and P4 fractions, suggesting an ER and/or Golgi localization.
|
To distinguish between these two possibilities, a purification of the rough endoplasmic reticulum (RER) compartment was performed using a simple sucrose density barrier method (9). The western blot analysis revealed the presence of SEPN1 in the RER (Fig. 3B). Subsequent immunoblotting with the compartment-specific markers calnexin and p23 confirmed the RER enrichment and the lack of Golgi contamination in this fraction.
To further confirm the ER localization, immunofluorescence assays were performed on fixed cells. Since our polyclonal antibodies were not sensitive enough to detect SEPN1 in normal cells, the construct pXJSelN6cys encoding SEPN1 was transfected into HeLa cells to transiently overexpress the protein. To increase SEPN1 expression and to avoid translation of the truncated SEPN1 form, the TGA codon was mutated to TGC allowing cysteine instead of selenocysteine incorporation at position 427. Expression of a 70 kDa protein product was verified by western blot analysis (data not shown). A positive staining by SEPN1 antibodies was observed in a fine reticular cytoplasmic and perinuclear pattern consistent with expression in the ER. The double-staining with specific markers of endoplasmic reticulum, mitochondria and golgi apparatus confirmed the ER localization (Fig. 4A). No fluorescence was visible in the plasma membrane.
|
The endoplasmic reticulum addressing and retention signal is located at the N-terminus of the protein
Examination of the N-terminal SEPN1 sequence revealed the presence of two potential ER targeting signals within exon 1: a di-arginine motif at positions 3 and 5, previously characterized as the ER targeting motif of the Iip33 protein (10), and a stretch of hydrophobic amino acids (from L30 to V49) similar to the ER addressing and retention signal of the microsomal cytochrome P450 (11). To verify whether exon 1 contains effective signals, the construct pXJSelN6cys
ex1 encoding the cysteine-modified SEPN1, but lacking the first 61 amino acids corresponding to exon 1, was transfected into HeLa cells. The corresponding expression product was stable and exhibited the expected molecular weight (data not shown). This N-terminal-truncated SEPN1 form displayed a different localization and was then accumulated within the nucleus, suggesting that a major ER addressing signal is located within the first exon (Fig. 4B, a–c). This latter finding was confirmed by the fusion of several SEPN1 N-terminal regions of increasing lengths upstream of the green fluorescent protein (GFP). The subcellular localization of the fusion proteins expressed in the HeLa cells was determined by virtue of the GFP fluorescence. An ER localization was observed with the constructs carrying either the first exon (Fig. 4B, d–f ) or the first exon combined with the second, third and fourth exons (data not shown). These experiments demonstrated the presence of ER-addressing and -retention information in the first exon.
SEPN1 is an integral membrane glycoprotein
In accordance with the above results, the topology predictions on SEPN1 amino acid sequence suggested the presence of one, two or three transmembrane domains (HMMTOP, TMHMM, TEMpred available at the ExPASy Molecular Biology Server: www.expasy.ch; Fig. 1). An experimental confirmation of SEPN1 association with the membrane was provided by a solubilization protocol in which microsomal-enriched fractions from fibroblasts were incubated with carbonate and Triton X-100. Western blot analysis showed that SEPN1 was not extracted by carbonate and only poorly solubilized by Triton X-100 treatment, indicating protein insertion into the membrane (Fig. 5A). By comparison calnexin, which was also retained in the pellet after carbonate incubation, was almost completely solubilized by detergent treatment. A behavior similar to SEPN1 has already been observed for WFS1, another ER integral membrane protein (12), which displayed a poor solubilization even after incubation with 1% Triton.
|
In addition, four putative N-glycosylation sites were predicted in the SEPN1 amino acid sequence at positions 155, 448, 470 and 496 (ScanProsite) (Fig. 1). To confirm the glycosylation status of SEPN1, microsomal fractions from fibroblast extracts were treated with N-glycosidase F (Fig. 5B, lane 2). The treatment led to an increased electrophoretic mobility of SEPN1 in the immunoblot analysis, confirming the presence of N-glycosylation sites. Secondly, the extract was digested with endoglycosidase H. All high-mannose oligosaccharides of glycoproteins are cleaved by this enzyme, but acquire resistance to this treatment once modified during trafficking through the Golgi apparatus (13). The endoglycosidase H sensitivity observed in the immunoblot (Fig. 5B, lane 3) is in agreement with the localization of SEPN1 in the ER. The sensitivity to both enzymes was also observed for the endogenous SEPN1 from fetal skeletal muscle (Fig. 5B, lanes 4–6) and for overexpressed cysteine-modified protein (data not shown).
SEPN1 is mostly expressed in fetal tissues and dividing cells
The specific expression of SEPN1 was investigated in several human tissues. Western blot experiments showed a lower expression of SEPN1 in most adult tissues, including skeletal muscle (Fig. 6A, lanes 1–5), compared to the strong signal detected in all fetal tissues (Fig. 6A, lanes 6–9). The decrease in SEPN1 expression from fetal to adult tissues prompted us to analyse SEPN1 expression during myotube maturation in vitro. Human myoblast cultures were switched into a fusion medium, inducing their differentiation and maturation into myotubes, and collected 1, 3, 5 and 7 days following the medium change. Immunoblotting confirmed a progressive decrease in the level of SEPN1 as the myoblasts fused to become differentiated myotubes (Fig. 7, lanes 2–6). Increasing expression of the fast myosin heavy chain depicted the differentiation process.
|
|
|
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONRESULTSDISCUSSIONMATERIALS AND METHODSREFERENCES |
|---|
We recently demonstrated the implication of the gene encoding selenoprotein N (SEPN1) in congenital muscular dystrophy with rigid spine syndrome and in the ‘classical’ form of multiminicore disease (1,2). The same SEPN1 homozygous mutations have been described in the two groups of patients, suggesting that these entities usually defined by clinical and morphological features, are in fact the same disease now called SEPN1-related myopathy (2). SEPN1 was first identified as a new member of the selenoprotein family (3), but data concerning the protein were rather scarce and no function has yet been ascribed to SEPN1.
To obtain further information, we performed biochemical and cellular characterization using newly developed antibodies. Although two transcripts were detected by RT–PCR experiments, in all the tissues and cell types analysed only one protein of 70 kDa could be revealed with the three antibodies raised against different parts of SEPN1. The observed molecular mass corresponds to the translation product of isoform 2 cDNA, lacking the third exon. In addition, SEPN1 treatment with N-glycosidase F or endoglycosidase H gave rise to a 62 kDa product, which corresponds to the predictive molecular mass of isoform 2. No expression of isoform 1 was obtained by transfection of the full-length cDNA. Therefore, it seems that the additional in-frame UGA codon present in the alternatively spliced exon is recognized as a real stop codon. It should be noted that exon 3 is absent in the genomic sequence of mouse SEPN1 (A. Lescure, M. Rederstorff and A. Krol, unpublished results). In view of these experiments, we conclude that the SEPN1 gene product is a 70 kDa glycosylated protein, containing a single selenocysteine residue.
Subcellular fractionation experiments carried out on fibroblasts and immunofluorescence analysis of transfected cells overexpressing SEPN1 indicate that SEPN1 is an endoplasmic reticulum resident protein. In agreement with this observation, SEPN1 extracted from fibroblasts and fetal muscle, was shown to be sensitive to endoglycosidase H. The rules governing the ER localization of proteins are not fully understood. Nevertheless, some specific targeting signals have been described (reviewed in 14). SEPN1–GFP fusion experiments demonstrated that the first exon is sufficient to address and retain the protein within the ER compartment. The first exon contains two predicted signals consisting in two N-terminal arginine residues at positions 3 and 5 (10), and an hydrophobic sequence located between positions 30 and 49, predicted as part of a transmembrane domain (11). Additional transfection experiments with disruption or mutation in the hydrophobic stretch and dibasic motif have to be performed in order to identify the sequence involved. The ER localization of SEPN1 suggests possible roles of this protein in membrane trafficking, protein processing and/or regulation of ER calcium homeostasis, the latter being especially important for the muscular function (15).
The overall biochemical experiments provide some clues to the topology of the protein. It emerges that SEPN1 is a transmembrane protein, with a striking insertion into the membrane, as illustrated by its poor solubilization in a detergent such as Triton. This result prompts us to hypothesize that SEPN1 may be a part of an integral protein complex. Moreover, the presence of the di-arginine motif suggests the hypothesis that the N-terminus of SEPN1 is cytoplasmic. However, the exact topology as well as the number of transmembrane domains remain to be elucidated.
Northern blot experiments have previously shown that SEPN1 transcripts are distributed in the majority of adult tissues (3). The present study demonstrates that the protein accumulates much less in the adult than in fetal samples, which display a stronger expression level in all the tissues analysed. Additionally, the level of SEPN1 appears to decline abruptly during myotube formation. Thus, SEPN1 seems to be preferentially expressed in proliferating cells or growing tissues. Examination of SEPN1 expression patterns during development by in situ hybridization in a zebrafish embryo revealed that SEPN1 transcripts were first detected in the antero-posterior axis and the notochord, a precursor of the spine, and then in the somites, from which the skeletal muscles arise. At later stages, SEPN1 was more uniformly expressed in all tissues (16). A putative involvement of SEPN1 in early muscle formation is in agreement with the onset of the congenital muscular disease caused by mutations in SEPN1. Indeed, most of the patients present hypotonia at birth with marked neck weakness and a delayed motor development. Besides, SEPN1 expressed at a basal level in post-natal tissues may play a role in muscle cell function and/or regeneration.
In conclusion, this study provides the first characterization of selenoprotein N as an integral membrane glycoprotein of the endoplasmic reticulum. Further analyses need to be performed to elucidate its cellular functions. The discovery of its role should help to understand better the mechanisms underlying the emergence of the disease and open the way to possible therapy. Although SEPN1 is faintly detectable in muscle tissue, the antibodies developed for this project could constitute useful diagnosis tools on patient skin fibroblasts and help the classification of atypical patients into SEPN1-related myopathy group.
|
MATERIALS AND METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONRESULTSDISCUSSIONMATERIALS AND METHODSREFERENCES |
|---|
Antibodies
Two rabbit polyclonal antibodies (137 and 143) directed against two synthetic peptides (KEGLRRGLPLLQP and KPEEIESNLFSFS, Eurogentec) of the C-terminal region of SEPN1 were generated. Antibody 137 was affinity-purified with the immunization peptide covalently bound to an Affi-Gel 10 column (BioRad). Antibody 143 was immunopurified using a recombinant polypeptide protein encompassing the C-terminal 116 residues of SEPN1. To this end, the 412bp cDNA sequence, corresponding to the 348 bp at the 3' end of the SEPN1 coding sequence plus 64 bp of the 3'-UTR, was cloned into the pQE-31 expression vector (QIAexpress Kit, Qiagen). E. coli M15 was transfected with this construct and the His-tagged expression product was purified using a Ni-NTA column. The purified product was then covalently linked to a CNBr-activated Sepharose column to perform affinity chromatography with the polyclonal antibody 143.
A rabbit polyclonal antibody (168) was raised against a 34 kDa recombinant protein, which corresponds to central domain of selenoprotein N (residues 85–427). To this end, a His-tagged polypeptide was produced from a 1029 bp cDNA fragment of SEPN1 isoform 2, cloned into the bacterial expression vector pQE-32 (Qiagen), purified and injected into rabbits. The serum was purified by affinity chromatography on a CNBr-activated Sepharose column substituted with the polypeptide used for immunization.
Monoclonal antibodies against hsp60 (H4149), emerin (4G5) and fast myosin heavy chain (MF20) were purchased from Sigma, Novocastra Laboratories and ATCC, respectively. Anti-Golgi p23 polyclonal antibody was a gift from Dr M. Rojo (17). The polyclonal antibodies against calnexin (SPA860) and desmin (D8281) were purchased at Stressgen and Sigma, respectively. Three peroxidase-conjugated secondary antibodies were used: goat anti-rabbit from Biorad (170-6515) and from DAKO (P 0448), and a goat anti-mouse (P 0447) from DAKO.
Plasmid constructions
To perform GFP fusion with different N-terminal parts of selenoprotein N, SEPN1 cDNA fragments were amplified with oligonucleotides introducing Sma1 and Kpn1 restriction sites at the 5' and 3' ends, respectively. The forward primer used to amplify all fragments is 5'-ccccgctctttcgcttccc-3'. The reverse primers are 5'-ggtacccagggtcttcagcgccagttc-3' for exon 1, 5'-ggtacctagcttctcagcaatgggtttga-3' for exons 1 and 2, 5'-ggtaccgccatctttgctcttggtcat-3' for exons 1, 2 and 4 or 1, 2, 3 and 4. The amplified fragments were purified and recovered by gel excision using Nucleospin Extract (Macherey-Nagel) and inserted into the pGEM-T easy vector (Promega) for sequencing. They were then sub-cloned into the expression vector pXJ GFP C-term using the added restriction site, yielding the pXJex1GFP, pXJex1-2GFP, pXJex1-2-4GFP and pXJex1-2-3-4GFP constructs. Beforehand, the pXJ GFP C-term vector had been obtained as follows: the coding sequence of the alpha GFP cycle 3 mutant (Maxygen) was PCR amplified with forward 5'-ggtaccgctagcaaaggagaagaac-3' and reverse 5'-agatctttatttgtagagctcatc-3' primers, which introduced a KpnI and a BglII site at the 5' and 3' ends, respectively, and the resulting 728 bp KpnI-BglII fragment was sub-cloned into the eukaryotic expression vector pXJ41 (a gift from P. Chambon).
In order to reconstitute full-size cDNAs for SEPN1, a 363 bp SacI–SacII fragment encompassing exon 1, 2 and 4 of SEPN1 was introduced upstream of pSelN, a partial cDNA containing the coding sequence downstream of exon 4 in pBluescript SK(-) (3) to yield pSelN6. Alternatively, a 468 bp SacII–SacII fragment containing exon 1, 2, 3 and 4 of SEPN1 was cloned upstream of pSe1N to give rise to pSelN7. Subsequently, a 4257 bp EcoRI–XhoI fragment from pSelN6, or a 4362 bp EcoRI–XhoI fragment from pSelN7, was cloned into pXJ41, yielding pXJSelN6 or pXJSelN7, respectively. Deletion of exon 1 in pXJSelN6 was achieved by site directed mutagenesis using oligonucleotide 5'-ccagccgcagccatgggcctggcgctgaagaacctg-3' to yield pXJSelN6ex1. In pSelN, the selenocysteine codon TGA was mutated to a TGC cysteine codon by site-directed mutagenesis using oligonucleotide 5'-cagtcctgctgcggttcagggc-3'. The 2127 BamHI fragment containing the TGC codon was substituted for the corresponding fragment into pXJSelN6 or pXJSelN6ex1 to yield pXJSelN6cys and pXJSelN6cysex1.
Cell cultures
Human fibroblasts were cultured from control skin biopsies from a 10-week-old embryo, a 9-year-old child, a 38-year-old adult and from an RSMD1 patient with the homozygous SEPN1 mutation 1446delC, resulting in a frameshift at residue L482 (1).
Fibroblasts were grown in DMEM (Dulbecco's modified Eagle's medium, Life Technologies) supplemented with 10% fetal calf serum (Life Technologies), 20 U/ml streptomycin and penicillin at 37°C in a humidified atmosphere with 5% CO2.
Control human myoblasts were grown in Ham's Nutrient-Mix-F-10 (Life Technologies) supplemented with 20% fetal calf serum (Life Technologies), 20 U/ml streptomycin and penicillin at 37°C in a humidified atmosphere with 5% CO2. For differentiation of myoblasts into myotubes, confluent cultures were switched into fusion medium DMEM supplemented with 10 µg/ml insulin and 100 µg/ml transferrin, 20 U/ml streptomycin and penicillin as described in Edom et al. (18). Cells were collected either prior to, or 1, 3, 5 and 7 days after incubation in a differentiation medium.
HeLa cells (ATCC) were cultured in DMEM, supplemented with 10% heat-inactivated fetal calf serum. 20 U/ml streptomycin and penicillin at 37°C in a humidified atmosphere with 5% CO2.
Transfection protocol
HeLa cells were grown on coverslips in 35 mm tissue culture plates in a growth medium. At 70–80% culture confluence, cells were transfected with 0.4 µg of pXJSelN6cys or pXJSelN6cysex1 with LipofectamineTM Reagent (Invitrogen) according to the manufacturer's instructions. In another set of experiments, HeLa cells were transfected by calcium phosphate precipitation with 0.3 µg pXJGFP, pXJex1GFP, pXJex1-2GFP, pXJex1-2-4GFP or pXJex1-2-3-4GFP, as described by Jordan et al. (19).
Protein extraction and glycosylation analysis
Tissue or cell proteins were extracted by homogenization in a buffer containing 80 mM Tris–HCl pH 6.8/10% SDS/0.12 M sucrose/10 mM EDTA/1 mM PMSF/1 mM benzamidine. The homogenate was incubated for 10 min at 55°C, centrifuged at 300g for 5 min. The protein concentration was determined on the supernatant using the BCA protein Assay (Pierce Chemical Company, Rockford, IL, USA). For glycosylation analysis, protein samples (20–25 µg) were digested for 2 h at 37°C with 12 units of N-glycosidase F (Roche) or overnight with 0.05 units of endoglycosidase H (Roche) according to the manufacturer's instructions.
Subcellular fractionation
Fibroblast extracts were fractionated by differential centrifugations based on a protocol described by Graham (9). Cells were homogenized in 0.25 M sucrose/5 mM Tris–HCl pH 7.4/2 mM and disrupted mechanically by 40 passages through a 22 gauge needle (Terumo 0.7x30/22Gx11/4/Nr12). Cell lysis was checked by Trypan blue staining. The homogenate (H) was centrifuged at 1000g for 10 min (Sigma, 12024-H) to yield S1 and P1. The S1 supernatant was centrifuged at 3000g for 10 min (Sigma, 12024-H) to yield S2 and P2. Pellets P1 and P2 were washed once by resuspension in homogenization buffer, repelleted by centrifugation as before, and resuspended in small volumes of homogenization buffer. The presence of nuclei was checked by Giemsa staining on fractions H, P1 and S1. The S2 supernatant was centrifuged at 10 000g for 20 min (Sigma, 12024-H) and the resulting S3 supernatant was centrifuged at 90 000g for 40 min (Beckman, TLA 45) to yield S4 and P4. Pellets P3 and P4 were resuspended in small volumes of homogenization buffer.
For rough endoplasmic reticulum (RER) purification, cells were processed as above except that the second step (3000g for 10 min) was omitted. The S1 supernatant was directly centrifuged at 10 000g for 20 min (Sigma, 12024-H). The resulting S3 supernatant was overlaid onto a discontinuous sucrose gradient containing 1.3 M Sucrose, 5 mM Tris–HCl pH 8, 15 mM CsCl and 0.6 M Sucrose, 5 mM Tris–HCl pH 8, 15 mM CsCl and centrifuged at 100 000g for 90 min (Beckman, SW 41i). The resulting RER pellet was resuspended in small volumes of homogenization buffer. Fractions were then processed for western blot analysis.
Alkaline treatment
Fibroblasts were homogenized as described in ‘Protein extraction and glycosylation’. The S1 supernatant was divided into three aliquots and centrifuged at 90 000g for 40 min (Beckman, TLA 45). The resulting microsomal-enriched pellets were resuspended in either 0.25 M sucrose/5 mM Tris–HCl pH 7.4/2 mM PMSF, or in the same buffer containing 1% Triton X-100 or in 0.1 M Na2CO3 pH 11.5, respectively. After 30 min incubation at 4°C and centrifugation at 90 000g for 40 min, pellets and supernatants were processed for western blot analysis.
Electrophoresis and western blotting
Subcellular fractions or extracts were subjected to SDS-PAGE and electrophoretically transferred to PVDF membranes (Immobilon-P, Millipore) according to Towbin et al. (20). Membranes were stained with Ponceau red to determine the blotting efficiency.
Indirect immunofluorescence microscopy
Forty-eight hours after transfection, HeLa cells were fixed and permeabilized in methanol for 5 min at -20°C or fixed in 4% paraformaldehyde and permeabilized in 0.5% Triton X-100/PBS for 10 min at room temperature. After rinsing in PBS, non-specific binding sites were blocked in 10% goat serum in PBS for 1 h at room temperature. Cells were then incubated for 2 h with the immunopurified polyclonal antibody 143 (1/50 dilution) in the blocking buffer, followed by washing in PBS and incubation for 1 h with mouse monoclonal antibodies against BiP (1/100, SPA-827, Stressgen), or giantin (1/1000, 324450, Calbiochem), or COX4 [1/100, Bakker et al. (21)] in PBS. The cells were washed in PBS and incubated for 1 h with a mixture of FITC-conjugated goat anti-rabbit (1/200, Jackson) and Alexa Fluor 568 goat anti-mouse (1/1000, Molecular Probes). After washing in PBS, cells were mounted with 10% Mowiol and viewed on a Zeiss Axiophot microscope.
|
ACKNOWLEDGEMENTS |
|---|
We thank Drs Martine Verdière-Sahuqué and Manuel Rojo for helpful discussion and technical advice on fractionation experiments, and antibody purification, immunofluorescence and microscopy, respectively. We are grateful to Christine Loegler for technical assistance, Drs Vincent Mouly and Denis Furling for the gift of human muscular cells, the AFM Banque de Tissus pour la Recherche for providing human tissues and the confocal microscope facility at the Campus Esplanade in Strasbourg. Drs Marc Fiszman, Gill Butler-Browne and Valérie Allamand are thanked for critical reading of the manuscript. This work was supported by funds from the INSERM (French INSERM/AFM Research network on rare disorders), the Association Française contre les Myopathies (AFM) to P.G., the NEYE-Foundation to U.M.W. and the European Commission (contract no. QLG1-CT1999-00870) to P.G. and U.M.W., the Ligue Régionale contre le Cancer and the Association pour la Recherche Contre le Cancer to A.K
|
FOOTNOTES |
|---|
* To whom correspondence should be addressed at: INSERM U582, Institut de Myologie, Groupe Hospitalier Pitié-Salpêtrière, 47 boulevard de l'Hôpital, 75651 Paris cedex 13, France. Tel: +33 142165750; Fax: +33 142165700; Email: p.guicheney{at}myologie.chups.jussieu.fr
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONRESULTSDISCUSSIONMATERIALS AND METHODSREFERENCES |
|---|
- Moghadaszadeh, B., Petit, N., Jaillard, C., Brockington, M., Roy, S.Q., Merlini, L., Romero, N., Estournet, B., Desguerre, I., Chaigne, D. et al. (2001) Mutations in SEPN1 cause congenital muscular dystrophy with spinal rigidity and restrictive respiratory syndrome. Nat. Genet., 29, 17–18.[CrossRef][Web of Science][Medline]
- Ferreiro, A., Quijano Roy, S., Pichereau, C., Moghadaszadeh, B., Goemans, N., Bonnemann, C., Jungbluth, H., Straub, V., Villanova, M., Leroy, J.P. et al. (2002) Mutations of the Selenoprotein N gene, which is implicated in rigid spine muscular dystrophy, cause the classical phenotype of multiminicore disease: reassessing the nosology of early-onset myopathies. Am. J. Hum. Genet., 71, 739–749.[CrossRef][Web of Science][Medline]
-
Lescure, A., Gautheret, D., Carbon, P. and Krol, A. (1999) Novel selenoproteins identified in silico and in vivo by using a conserved RNA structural motif. J. Biol. Chem., 274, 38147–38154.
[Abstract/Free Full Text] -
Hatfield, D.L. and Gladyshev, V.N. (2002) How selenium has altered our understanding of the genetic code. Mol. Cell. Biol., 22, 3565–3576.
[Free Full Text] - Krol, A. (2002) Evolutionarily different mRNA motifs and RNA–protein complexes to achieve selenoprotein synthesis. Biochimie, 84, 765–774.[Medline]
- Lescure, A., Fagegaltier, D., Carbon, P. and Krol, A. (2002) Protein factors mediating selenoprotein synthesis. Curr. Prot. Pept. Sci., 3, 143–151.
-
Zhong. L., Arner, E.S. and Holmgren, A. (2000) Structure and mechanism of mammalian thioredoxin reductase: the active site is a redox-active selenolthiol/selenenylsulfide formed from the conserved cysteine–selenocysteine sequence. Proc. Natl Acad. Sci. USA, 97, 5854–5859.
[Abstract/Free Full Text] -
Guimaraes, M.J., Peterson, D., Vicari, A., Cocks, B.G., Copeland, N.G., Gilbert, D.J., Jenkins, N.A., Ferrick, D.A., Kastelein, R.A., Bazan, J.F. et al. (1996) Identification of a novel selD homolog from eukaryotes, bacteria, and archaea: is there an autoregulatory mechanism in selenocysteine metabolism? Proc. Natl Acad. Sci. USA, 93, 15086–15091.
[Abstract/Free Full Text] - Graham, J. (1984) Isolation of subcellular organelles and membranes. In Rickwood, D. (ed.), Centrifugation (2nd Edition): a Practical Approach. IRL Press, Oxford, pp. 161–182.
- Schutze, M.P., Peterson, P.A. and Jackson, M.R. (1994) An N-terminal double-arginine motif maintains type II membrane proteins in the endoplasmic reticulum. EMBO J., 13, 1696–1705.[Web of Science][Medline]
-
Szczesna-Skorupa, E., Ahn, K., Chen, C., Doray, B. and Kemper, B. (1995) The cytoplasmic and N-terminal transmembrane domains of cytochrome P450 contain independent signals for retention in the endoplasmic reticulum. J. Biol. Chem., 270, 24327–24333.
[Abstract/Free Full Text] -
Takeda, K., Inoue, H., Tanizawa, Y., Matsuzaki, Y., Oba, J., Watanabe, Y., Shinoda, K. and Oka, Y. (2001) WFS1 (Wolfram syndrome 1) gene product: predominant subcellular localization to endoplasmic reticulum in cultured cells and neuronal expression in rat brain. Hum. Mol. Genet., 10, 477–484.
[Abstract/Free Full Text] -
Cai, Y., Maeda, Y., Cedzich, A., Torres, V.E., Wu, G., Hayashi, T., Mochizuki, T., Park, J.H., Witzgall, R. and Somlo, S. (1999) Identification and characterization of polycystin-2, the PKD2 gene product. J. Biol. Chem., 274, 28557–28565.
[Abstract/Free Full Text] - Teasdale, R.D. and Jackson, M.R. (1996) Signal-mediated sorting of membrane proteins between the endoplasmic reticulum and the golgi apparatus. A. Rev. Cell Dev. Biol., 12, 27–54.[CrossRef][Web of Science][Medline]
- Froemming, G.R. and Ohlendieck, K. (2001) The role of ion-regulatory membrane proteins of excitation-contraction coupling and relaxation in inherited muscle diseases. Front. Biosci., 1, D65–74.
- Tfisse, C., Degrave, A., Kryukov, G.V., Gladyshev, V.N., Obrecht-Phlumio, S., Krol, A., Thisse, B. and Lescure, A. (2003) Spatial and temporal expression patterns of seleno-protein genes during embryogenesis in zebrafish. Mech. Dev. (in press).
-
Rojo, M., Pepperkork, R., Emery, G., Kellner, R., Stang, E., Parton, R.G. and Gruenberg, J. (1997) Involvement of the transmembrane protein p23 in biosynthetic protein transport. J. Cell Biol., 139, 1119–1135.
[Abstract/Free Full Text] - Edom, F., Mouly, V., Barbet, J.P., Fiszman, M.Y. and Butler-Browne, G.S. (1994) Clones of human satellite cells can express in vitro both fast and slow myosin heavy chains. Dev. Biol., 164, 219–229.[CrossRef][Web of Science][Medline]
-
Jordan, M., Schallhorn, A. and Wurm, F.M. (1996) Transfecting mammalian cells: optimization of critical parameters affecting calcium-phosphate precipitate formation. Nucl. Acids Res., 24, 596–601.
[Abstract/Free Full Text] -
Towbin, H., Staehelin, T. and Gordon, J. (1979) Electrophoretic transfer of proteins from polyacrylamide gels to nitrocellulose sheets: procedure and some applications. Proc. Natl Acad. Sci. USA, 76, 4350–4354.
[Abstract/Free Full Text] -
Bakker, A., Barthelemy, C., Frachon, P., Chateau, D., Sternberg, D., Mazat, J.P. and Lombes, A. (2000) Functional mitochondrial heterogeneity in heteroplasmic cells carrying the mitochondrial DNA mutation associated with the MELAS syndrome (mitochondrial encephalopathy, lactic acidosis, and strokelike episodes). Pediatr. Res., 48, 143–150.[Web of Science][Medline]
CiteULike 
Connotea 
Del.icio.us What's this?
This article has been cited by other articles:
|
|
 |
|
  J.-C. Zhou, H. Zhao, J.-G. Li, X.-J. Xia, K.-N. Wang, Y.-J. Zhang, Y. Liu, Y. Zhao, and X. G. Lei Selenoprotein Gene Expression in Thyroid and Pituitary of Young Pigs Is Not Affected by Dietary Selenium Deficiency or Excess J. Nutr., June 1, 2009; 139(6): 1061 - 1066. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 A. Bender, P. Hajieva, and B. Moosmann From the Cover: Adaptive antioxidant methionine accumulation in respiratory chain complexes explains the use of a deviant genetic code in mitochondria PNAS, October 28, 2008; 105(43): 16496 - 16501. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
M. J. Jurynec, R. Xia, J. J. Mackrill, D. Gunther, T. Crawford, K. M. Flanigan, J. J. Abramson, M. T. Howard, and D. J. Grunwald Selenoprotein N is required for ryanodine receptor calcium release channel activity in human and zebrafish muscle PNAS, August 26, 2008; 105(34): 12485 - 12490. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 L. Grumolato, H. Ghzili, M. Montero-Hadjadje, S. Gasman, J. Lesage, Y. Tanguy, L. Galas, D. Ait-Ali, J. Leprince, N. C. Guerineau, et al. Selenoprotein T is a PACAP-regulated gene involved in intracellular Ca2+ mobilization and neuroendocrine secretion FASEB J, June 1, 2008; 22(6): 1756 - 1768. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 Y. Zhang, Y. Zhou, U. Schweizer, N. E. Savaskan, D. Hua, J. Kipnis, D. L. Hatfield, and V. N. Gladyshev Comparative Analysis of Selenocysteine Machinery and Selenoproteome Gene Expression in Mouse Brain Identifies Neurons as Key Functional Sites of Selenium in Mammals J. Biol. Chem., January 25, 2008; 283(4): 2427 - 2438. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 M. Rederstorff, V. Allamand, P. Guicheney, C. Gartioux, P. Richard, D. Chaigne, A. Krol, and A. Lescure Ex vivo correction of selenoprotein N deficiency in rigid spine muscular dystrophy caused by a mutation in the selenocysteine codon Nucleic Acids Res., January 17, 2008; 36(1): 237 - 244. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
B. A. Carlson, M. E. Moustafa, A. Sengupta, U. Schweizer, R. Shrimali, M. Rao, N. Zhong, S. Wang, L. Feigenbaum, B. J. Lee, et al. Selective Restoration of the Selenoprotein Population in a Mouse Hepatocyte Selenoproteinless Background with Different Mutant Selenocysteine tRNAs Lacking Um34 J. Biol. Chem., November 9, 2007; 282(45): 32591 - 32602. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 S. E. Riechman, R. D. Andrews, D. A. MacLean, and S. Sheather Statins and Dietary and Serum Cholesterol Are Associated With Increased Lean Mass Following Resistance Training J. Gerontol. A Biol. Sci. Med. Sci., October 1, 2007; 62(10): 1164 - 1171. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 B. Moghadaszadeh and A. H. Beggs Selenoproteins and Their Impact on Human Health Through Diverse Physiological Pathways. Physiology, October 1, 2006; 21(5): 307 - 315. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 J. Kohrle, F. Jakob, B. Contempre, and J. E. Dumont Selenium, the Thyroid, and the Endocrine System Endocr. Rev., December 1, 2005; 26(7): 944 - 984. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 S. L. Venance, W. J. Koopman, B. A. Miskie, R. A. Hegele, and A. F. Hahn Rigid spine muscular dystrophy due to SEPN1 mutation presenting as cor pulmonale Neurology, January 25, 2005; 64(2): 395 - 396. [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||