In vitro translation in a cell-free system from Trypanosoma brucei yields glycosylated and glycosylphosphatidylinositol-anchored proteins.

African trypanosomes escape many cellular and unspecific immune reactions by the expression of a protective barrier formed from a repertoire of several hundred genes encoding immunologically distinct variant surface glycoproteins (VSGs). All mature VSGs are glycosylphosphatidylionositol-anchored and N-glycosylated. To study trypanosome-specific post-translational modifications of VSG, a cell-free system capable of in vitro translation, translocation into the rough endoplasmic reticulum, N-glycosylation and glycosylphosphatidylinositol-anchor addition was established using lysates of the bloodstream form of Trypanosoma brucei. Monitoring protein synthesis by [35S]methionine incorporation, labeled protein bands were readily detected by fluorography following SDS/PAGE. Appearance of these bands increased during a time-course of 45 min and was sensitive to cycloheximide but not chloramphenicol treatment. Efficiency of this system, in terms of incorporation of radiolabeled amino acids into newly formed proteins, is similar to reticulocyte lysates. The system does not, however, allow initiation of protein synthesis. Depending on the clone used, immunoprecipitation revealed one or two newly formed VSG bands. Upon digestion with N-glycosidase F these bands resulted in a single band of a lower apparent molecular mass, indicating that newly synthesized VSG underwent translocation and glycosylation in the cell-free system. Biotinylation of VSG and a combination of precipitation with immobilized avidin and detection of VSG using antibodies specific for clones and cross-reacting determinants revealed that newly formed VSG contained the glycosylphosphatidylinositol anchor.     

Folding of firefly luciferase during translation in a cell-free system.

In vitro synthesis of firefly luciferase and its folding into an enzymatically active conformation were studied in a wheat germ cell-free translation system. A novel method is described by which the enzymatic activity of newly synthesized luciferase can be monitored continuously in the cell-free system while this protein is being translated from its mRNA. It is shown that ribosome-bound polypeptide chains have no detectable enzymatic activity, but that this activity appears within a few seconds after luciferase has been released from the ribosome. In contrast, the renaturation of denatured luciferase under identical conditions occurs with a half-time of 14 min. These results support the cotranslational folding hypothesis which states that the nascent peptides start to attain their native tertiary structure during protein synthesis on the ribosome.     

The synthesis of vitellogenins by the digestive gland of Helix aspersa: evidence from cell-free translation of mRNA.

All the RNAs were extracted separately from the gonads and digestive glands of the garden snail Helix aspersa. After separation by oligo-dT-cellulose chromatography, the mRNA were evaluated using optical density measurements and agarose gel electrophoretic analysis and then used in a cell-free translation system: a rabbit reticulocyte lysate added with 35S methionine. The in vitro synthesized proteins were characterized by immunoprecipitation with polyclonal antibodies which were raised against mature oocyte proteins. The proteins were then evaluated quantitatively using liquid scintillation counting and qualitatively using SDS polyacrylamide gel electrophoresis and fluorography. The results demonstrated that the digestive gland of the garden snail is a vitellogenin synthesis site.     

Insulin modulation of human apolipoprotein B mRNA translation: studies in an in vitro cell-free system from HepG2 cells.

Insulin modulation of apolipoprotein B gene expression was studied at the translational level by the use of a cell-free translation system from a hepatoma cell-line, HepG2. Extracts of HepG2 cells lysed with lysolecithin were found to have high in vitro protein synthesizing activity utilizing endogenous mRNA. The level of peptide chain initiation was high, as suggested by a significant inhibition of translation by edeine. The translation products of endogenous mRNA in HepG2 cell-free lysate were probed with anti-apolipoprotein B antibodies to investigate its synthesis. A 550 kilodalton (kDa) polypeptide was selected by a polyclonal antibody, as well as a monoclonal antibody, against the C-terminal end of apolipoprotein B molecule. This in vitro synthesized polypeptide was also found to compare well in size with the in vivo product. The HepG2 lysate was also shown to efficiently synthesize in vitro a number of other proteins including albumin, apolipoprotein E, apolipoprotein A1, and actin. The in vitro synthesis of polypeptides as large as 500 kDa was unexpected and has not previously been demonstrated in a cell-free system. The HepG2 translation system was used to investigate the effect of insulin on the in vitro translation of apolipoprotein B. Lysates prepared from HepG2 cells treated with insulin were found to have lower translational activity (by an average of 52.3%) for apolipoprotein B compared with lysates from control untreated cells. In vitro synthesis of actin and apolipoprotein E were unaffected under these conditions. The insulin-stimulated decline in in vitro apolipoprotein B synthesis was not due to a change in apolipoprotein B mRNA levels as determined by slot- and Northern-blot analyses, suggesting that the inhibitory effect of insulin may be exerted partly at the level of apolipoprotein B mRNA translation.     

Behavior of cysteine mutants of human lysozyme in de novo synthesis and in vivo secretion

To investigate the mechanism of disulfide-bond-coupled de novo folding of human lysozyme, we have constructed 23 mutant enzymes in which cysteine residue(s) were replaced by alanine(s). The mutant genes were translated in vitro in a system composed of rabbit reticulocyte lysate, canine pancreatic microsomal vesicles and oxidized glutathione. This system allows the formation of intramolecular disulfide bonds in translation products translocated into the microsomal lumen. The mobilities of the translation products were analyzed by SDS/PAGE in nonreducing conditions. Some mutant lysozymes were found to form a compact conformation with native-like mobility in the presence of SDS. The de novo formation of the SDS-resistant compact conformation of each mutant correlated well with its efficiency of secretion by Saccharomyces cerevisiae. Our results suggest that the de novo synthesized products reflect the conformational states in vivo to some extent, and that the formation of SDS-resistant compact conformation can be regarded as a necessary condition for allowing lysozyme to be secreted. In addition, the analysis of a mutant C116A (Cys116----Ala) under different oxidative conditions suggests two distinct pathways for the disulfide-bond-coupled formation of the compact conformation.     

Biosynthesis of intestinal microvillar proteins. Processing of aminopeptidase N by microsomal membranes.

The biosynthesis of small-intestinal aminopeptidase N (EC 3.4.11.2) was studied in a cell-free translation system derived from rabbit reticulocytes. When dog pancreatic microsomal fractions were present during translation, most of the aminopeptidase N synthesized was found in a membrane-bound rather than a soluble form, indicating that synthesis of the enzyme takes place on ribosomes attached to the rough endoplasmic reticulum. The microsomal fractions process the Mr-115 000 polypeptide, which is the primary translation product of aminopeptidase N, to a polypeptide of Mr 140 000. This was found to be sensitive to the action of endo-beta-N-acetylglucosaminidase H (EC 3.2.1.96), showing that aminopeptidase N undergoes transmembrane glycosylation during synthesis. The position of the signal sequence in aminopeptidase N was determined by a synchronized translation experiment. It was found that microsomal fractions should be added before about 25% of the polypeptide was synthesized to ensure processing to the high-mannose glycosylated form. This suggests that the signal sequence is situated in the N-terminal part of the aminopeptidase N. The size of the cell-free translation product in the absence of microsomal fractions was found to be similar to that on one of the forms of the enzyme obtained from tunicamycin-treated organ-cultured intestinal explants.     

Efficient synthesis of a disulfide-containing protein through a batch cell-free system from wheat germ.

We have developed a highly productive cell-free protein synthesis system from wheat germ, which is expected to become an important tool for postgenomic research. However, this system has not been optimized for the synthesis of disulfide-containing proteins. Thus, we searched here for translation conditions under which a model protein, a single-chain antibody variable fragment (scFv), could be synthesized into its active form. Before the start of translation, the reducing agent dithiothreitol, which normally is added to the wheat germ extract but which inhibits disulfide formation during translation, was removed by gel filtration. When the scFv mRNA was incubated with this dithiothreitol-deficient extract, more than half of the synthesized polypeptide was recovered in the soluble fraction. By addition of protein disulfide isomerase in the translation solution, the solubility of the product was further improved, and nearly half of the soluble polypeptides strongly bound to the antigen immobilized on an agarose support. This strong binding component had a high affinity as shown by surface-plasmon resonance analysis. These results show that the wheat germ cell-free system can produce a functional scFv with a simple change of the reaction ingredients. We also discuss protein folding in this system and suggest that the disulfide bridges are formed cotranslationally. Finally, we show that biotinylated scFv could be synthesized in similar fashion and immobilized on a solid surface to which streptavidin is bound. SPR measurements for detection of antigens were also possible with the use of this immobilized surface.     

Synthesis and processing of glycoprotein D of herpes simplex virus types 1 and 2 in an in vitro system.

We carried out studies of in vitro translation and processing of glycoprotein D (gD) of herpes simplex virus types 1 and 2 by using mRNA from cells infected for 6 h and a reticulocyte lysate translation system. Polypeptides of 49,000 daltons were immunoprecipitated with anti-gD-1 sera. Each in vitro-synthesized molecule had the same methionine tryptic peptide profile as the respective in vivo precursors, pgD-1 and pgD-2. In addition, the polypeptides synthesized in vitro were larger than the corresponding molecules synthesized in the presence of tunicamycin. This suggested that each of the gD polypeptides synthesized in vitro contained a transient N-terminal signal sequence. When the translation mixture was supplemented with pancreatic microsomes, each of the gD polypeptides was converted cotranslationally to a larger-molecular-weight form. Processing involved addition of three N-asparagine-linked oligosaccharides and removal of the signal peptide. When trypsin was added after in vitro processing, a polypeptide which was 3,000 daltons smaller than the in vitro-modified form of gD was immunoprecipitated. Experiments with endo-beta-N-acetylglucosaminidase H showed that this polypeptide still contained the three N-asparagine-linked oligosaccharides. Two monoclonal antibodies, 57S (group V) and 17O (group VII), were used to further orient gD in microsomes. The group V determinant was located in the trypsin-sensitive 3,000-dalton fragment, and the group VII determinant was located in the portion of gD which was protected from trypsin. We concluded that gD is oriented with the three glycosylation sites inside the vesicles and that 3,000 daltons containing the group V determinant are located outside. Immunofluorescence studies indicated that the group V determinant of gD is inside the plasma membrane of herpes simplex virus-infected cells and that the group VII determinant is outside. This cellular orientation is consistent with predictions based on the in vitro experiments.     

In vitro translation and processing of rat kidney gamma-glutamyl transpeptidase

Rat kidney gamma-glutamyl transpeptidase is composed of two nonidentical glycosylated subunits. The enzyme is localized on the lumenal surface of the brush-border membranes of proximal tubule epithelial cells; it is attached to the membranes via an NH2-terminal segment of the larger of the two subunits. Tissue-labeling experiments followed by immunoprecipitation with antibodies directed against the enzyme and its two subunits demonstrate that a glycosylated single chain precursor (Mr = 78,000), containing the elements of both the subunits, is initially synthesized. Pulse-chase studies in the presence of pactamycin, and inhibitor of protein synthesis initiation, indicate that the larger of the two subunits is located at the NH2 terminus of the Mr = 78,000 precursor. The initial events in the biosynthesis and processing of gamma-glutamyl transpeptidase were investigated by in vitro translation of rat kidney mRNA. Such translation results in the synthesis of a Mr = 63,000 unglycosylated polypeptide which has been shown immunologically to contain the domains for both subunits. The Mr = 63,000 species is processed to a Mr = 78,000 core-glycosylated polypeptide when translation of mRNA is carried out in the presence of dog pancreas microsomes. This processing does not appear to be associated with cleavage of an NH2-terminal leader sequence. The Mr = 78,000 polypeptide is integrated into the microsomal membranes with an orientation that is analogous to that found on the brush-border membranes. Glycosylation and membrane integration of transpeptidase are cotranslational events. Upon longer incubation, the Mr = 78,000 species sequestered within the microsomal vesicles is cleaved to species corresponding in size to the two subunits of the kidney enzyme.     

Differential translation in normal and adenovirus type 5-infected human cells and cell-free systems.

When uninfected or adenovirus 5-infected KB cells are exposed to hypertonic medium, the incorporation of radioactive amino acids into protein decreases in both, but more severely in the uninfected cells. Although the effect of hypertonic medium on the synthesis of specific polypeptides varies, the translation of viral polypeptides as a class is less inhibited. The same patterns of proteins are synthesized regardless of the solute used in the hypertonic medium. The mechanism by which hypertonic conditions exert their effect on whole cells was investigated in K cell-free systems. It was possible to simulate the differential patterns of protein synthesis obtained in whole cells in hypertonic medium by increasing ion concentrations in cell-free extracts which are capable of initiating polypeptide chains on exogenous templates. However, in cell lysates which only elongate proteins, the same patterns were not obtained. Certain host and viral polypeptides displayed striking responses to increased ionic conditions in whole cells and cell-free systems. The synthesis of a host 44K protein, actin, appeared to be most sensitive; lower-molecular-weight proteins were fairly resistant. Among the viral proteins, the synthesis of 100K was inhibited, but most notable was the marked resistance of the synthesis of polypeptide IX. Possible mechanisms for differential synthesis and their significance are considered.     

Product of in vitro translation of the Rous sarcoma virus src gene has protein kinase activity.

In vitro translation of Rous sarcoma virus virion RNA resulted in the synthesis of a protein kinase which, when immunoprecipitated with antitumor serum, phosphorylated the immunoglobulin heavy chain. Even though in vitro translation of virion RNA resulted in the synthesis of a number of polypeptides which were recognized by antitumor serum, control experiments demonstrated that an immunoprecipitable protein kinase activity was found only when an immunoprecipitable p60src, the polypeptide product of the src gene, was synthesized. A protein kinase with similar properties was therefore intimately associated with p60src which was synthesized in vitro in the reticulocyte lysate, just as it is with p60src which is obtained from transformed chick and mammalian cells. It is therefore highly unlikely that this association is artifactual. ts NY68 is a mutant of Rous sarcoma virus which is able to transform cells at 36 but not at 41 degrees C. In vitro translation of ts NY68 virion RNA at 30 degrees C resulted in efficient synthesis of immunoprecipitable p60src, but very inefficient synthesis of an immunoprecipitable protein kinase. The p60src obtained by in vitro translation of wild-type virion RNA was more than 20-fold more active as a protein kinase than was that obtained from ts NY68 RNA. The correlation in the case of ts NY68 of a deficiency in protein kinase activity with an inability to transform cells at high temperature suggests that the protein kinase activity associated with p60src is indeed critical to cellular transformation.     

In vitro biosynthesis of the beta-subunit of the Na+/K+-ATPase in developing brine shrimp: glycosylation and membrane insertion

We demonstrate here translation, glycosylation, and membrane insertion of the beta-subunit of the Na+/K+-ATPase of the developing brine shrimp, Artemia, in a reticulocyte lysate translation system. The apparent molecular weight of the primary translation product as determined by SDS-PAGE is 33,000 +/- 1000 (n = 7). When microsomal membranes are present during the entire translation period, a new band with an apparent molecular weight of 37,000 +/- 1000 (n = 7) appears. This change in apparent molecular weight is due to the addition of about two N-linked oligosaccharides. The temporal relationship between protein synthesis and glycosylation have also been examined. Glycosylation and membrane insertion could be achieved if membranes were added after completion of about 70% of the peptide chain. However, glycosylation did not occur if membranes were added after the completion of translation of the beta-subunit. The beta-subunit was synthesized on membrane-bound polysomes, where about two N-linked oligosaccharides were added to the growing polypeptide chain. These studies demonstrate that in vitro translation systems will be useful for studying the biosynthesis of the beta-subunit of the brine shrimp, which is a good model system to examine the developmental regulation of the Na+/K+-ATPase.     

Physiology of ammonium assimilation in Neurospora crassa.

In Neurospora crassa the assimilation of high and low concentrations of ammonium occurs by two different pathways. When the fungi are growing exponentially on ammonium excess, this compound is fixed by a glutamic dehydrogenase and an octameric glutamine synthetase (GS). The synthesis of this GS polypeptide (beta) is regulated by the nitrogen source present in excess; being higher on glutamate, intermediate on ammonium, and lower on glutamine. When N. crassa is growing in fed-batch ammonium-limited cultures a different polypeptide of GS (alpha), arranged as a tetramer, is synthesized. In both conditions synthesis in vivo correlates with the data obtained with an in vitro translation system primed with N. crassa RNA. This different expression of alpha and beta GS polypeptides was also observed when the cultures were shifted from excess to low nitrogen, and vice versa. By agarose gel electrophoresis in the presence of methylmercury hydroxide, some separation of different mRNAs that direct the in vitro synthesis of alpha and beta GS polypeptides has been accomplished. Data are presented that establish the operation of the tetrameric alpha GS and of glutamate synthase in the assimilation of ammonium in low concentration.     

Expression of the mammalian mitochondrial genome. Stability of mitochondrial translation products as a function of membrane potential

Polypeptide decay has been measured as a function of membrane potential. Mitochondrial translation products were pulse-labeled in vitro with [35S]methionine using isolated rat heart mitochondria in the presence of an energy-generating system. The relative rate of protein degradation was estimated from the specific activity (counts/min/mg of protein) of the labeled translation products following the addition of unlabeled methionine (chase). To modulate membrane potential, inhibitors of oxidative phosphorylation were used singly or in combination; their effect was monitored by following uptake of the nonmetabolizable lipophilic cation triphenylmethylphosphonium. When the potential was dissipated, the rate of polypeptide decay increased and vice versa. These results suggest that the stability of mitochondrial translation products is linked to a process(es) that is dependent upon delta psi; likely candidates include synthesis and/or assembly of mitochondrial gene products.     

Mode of action of bottromycin A2: effect of bottromycin A2 on polysomes

When bottromycin A2 was added to an in vitro protein synthesis system carried out by naturally occurring polysomes, it inhibited protein synthesis effectively. Examination of the 3 steps of peptide chain elongation revealed that the binding of aminoacyl-tRNA to the polyribosomes was inhibited by bottromycin A2. In contrast, we concluded that the peptide bond formation and the translocation steps in this system were not inhibited by bottromycin A2 on the basis of the following observations: (1) The break-down of polysomes, which is dependent on EFG, puromycin and RR (ribosome releasing) factor, was insensitive to bottromycin A2; (2) The puromycin dependent release of polypeptide from polysomes, with or without EFG, was not inhibited by bottromycin A2. Thus bottromycin specifically interferes with proper functioning of the A sites of polysomes. This is consistent with the results obtained using the model system with synthetic polynucleotides.     

Erythrocyte membrane protein band 3: its biosynthesis and incorporation into membranes

Band 3, a transmembrane protein that provides the anion channel of the erythrocyte plasma membrane, crosses the membrane more than once and has a large amino terminal segment exposes on the cytoplasmic side of the membrane. The biosynthesis of band 3 and the process of its incorporation into membranes were studied in vivo in erythroid spleen cells of anemic mice and in vitro in protein synthesizing cell-free systems programmed with polysomes and messenger RNA (mRNA). In intact cells newly synthesized band 3 is rapidly incorporated into intracellular membranes where it is glycosylated and it is subsequently transferred to the plasma membrane where it becomes sensitive to digestion by exogenous chymotrypsin. The appearance of band 3 in the cell surface is not contingent upon its glycosylation because it proceeds efficiently in cells treated with tunicamycin. The site of synthesis of band 3 in bound polysomes was established directly by in vitro translation experiments with purified polysomes or with mRNA extracted from them. The band-3 polypeptide synthesized in an mRNA- dependent system had the same electrophoretic mobility as that synthesized in cells treated with tunicamycin. When microsomal membranes were present during translation, the in vitro synthesized band-3 polypeptide was cotranslationally glycosylated and inserted into the membranes. This was inferred from the facts that when synthesis was carried out in the presence of membranes the product had a lower electrophoretic mobility and showed partial resistance to protease digestion. Our observations indicate that the primary translation product of band-3 mRNA is not proteolytically processed either co- or posttranslationally. It is, therefore, proposed that the incorporation of band 3 into the endoplasmic reticulum (ER) membrane is initiated by a permanent insertion signal. To account for the cytoplasmic exposure of the amino terminus of the polypeptide we suggest that this signal is located within the interior of the polypeptide. a mechanism that explains the final transmembrane disposition of band 3 in the plasma membrane as resulting from the mode of its incorporation into the ER is presented.     

Physical and functional heterogeneity in TYMV RNA: evidence for the existence of an independent messenger coding for coat protein.

Turnip yellow mosaic virus RNA can be separated into two distinct components of 2 times 10(6) and 300 000 daltons molecular weight after moderate heat treatment in the presence of SDS or EDTA. The two species cannot have arisen by accidental in vitro degradation of a larger RNA, as they both possess capped 5' ends. Analysis of the newly synthesized proteins resulting from translation of each RNA by a wheat germ extract shows that the 300 000 molecular weight RNA can be translated very efficiently into coat protein. When translated in vitro the longer RNA gave a series of high molecular weight polypeptides but only very small amounts of a polypeptide having about the same mass as the coat protein. Thus our results suggest that the small RNA is the functional messenger for coat protein synthesis in infected cells.     

Multiple apomucin translation products from human respiratory mucosa mRNA.

Poly(A)-rich RNA was purified from a pool of five human tracheobronchial mucosa. After in vitro translation in a reticulocyte lysate and immunoprecipitation of the translated products, using either a polyclonal antiserum or a monoclonal antibody to deglycosylated respiratory mucin peptides, the products were characterized by SDS/PAGE. The respiratory mucin precursors migrated as a very large smear from almost the top of the resolving polyacrylamide gel to an area corresponding to a molecular mass of about 100 kDa. After hybridization with mucin cDNA probe TH 29 described by Crepin et al. [Crepin, M., Porchet, N., Aubert, J. P. & Degand, P. (1990) Biorheology 27, 471-484] respiratory mucin mRNAs also appeared polydisperse. Although degradation or incomplete translation of high-molecular-mass mRNA cannot be entirely ruled out, these results suggest that human respiratory apomucins consist of a family of peptides which share some common epitopes. This possibility is in agreement with (a) the diversity of mucin precursors observed previously with pulse/chase experiments performed with explants of human respiratory mucosa and (b) the polydispersity of secreted respiratory mucins observed by electron microscopy.