Cell-free synthesis of enzymically active tissue-type plasminogen activator. Protein folding determines the extent of N-linked glycosylation

Identification, characterization, and engineering of glycosylation in thrombolyticsa

Cardiovascular diseases, such as myocardial infarction, ischemic stroke, and pulmonary embolism, are the most common causes of disability and death worldwide. Blood clot hydrolysis by thrombolytic enzymes and thrombectomy are key clinical interventions. The most widely used thrombolytic enzyme is alteplase, which has been used in clinical practice since 1986. Another clinically used thrombolytic protein is tenecteplase, which has modified epitopes and engineered glycosylation sites, suggesting that carbohydrate modification in thrombolytic enzymes is a viable strategy for their improvement. This comprehensive review summarizes current knowledge on computational and experimental identification of glycosylation sites and glycan identity, together with methods used for their reengineering. Practical examples from previous studies focus on modification of glycosylations in thrombolytics, e.g., alteplase, tenecteplase, reteplase, urokinase, saruplase, and desmoteplase. Collected clinical data on these glycoproteins demonstrate the great potential of this engineering strategy. Outstanding combinatorics originating from multiple glycosylation sites and the vast variety of covalently attached glycan species can be addressed by directed evolution or rational design. Directed evolution pipelines would benefit from more efficient cell-free expression and high-throughput screening assays, while rational design must employ structure prediction by machine learning and in silico characterization by supercomputing. Perspectives on challenges and opportunities for improvement of thrombolytic enzymes by engineering and evolution of protein glycosylation are provided.

A Cell-free Expression Pipeline for the Generation and Functional Characterization of Nanobodies

Cell-free systems are well-established platforms for the rapid synthesis, screening, engineering and modification of all kinds of recombinant proteins ranging from membrane proteins to soluble proteins, enzymes and even toxins. Also within the antibody field the cell-free technology has gained considerable attention with respect to the clinical research pipeline including antibody discovery and production. Besides the classical full-length monoclonal antibodies (mAbs), so-called "nanobodies" (Nbs) have come into focus. A Nb is the smallest naturally-derived functional antibody fragment known and represents the variable domain (V_HH, ∼15 kDa) of a camelid heavy-chain-only antibody (HCAb). Based on their nanoscale and their special structure, Nbs display striking advantages concerning their production, but also their characteristics as binders, such as high stability, diversity, improved tissue penetration and reaching of cavity-like epitopes. The classical way to produce Nbs depends on the use of living cells as production host. Though cell-based production is well-established, it is still time-consuming, laborious and hardly amenable for high-throughput applications. Here, we present for the first time to our knowledge the synthesis of functional Nbs in a standardized mammalian cell-free system based on Chinese hamster ovary (CHO) cell lysates. Cell-free reactions were shown to be time-efficient and easy-to-handle allowing for the "on demand" synthesis of Nbs. Taken together, we complement available methods and demonstrate a promising new system for Nb selection and validation.

Synthetic Biology Goes Cell-Free

Cell-free systems (CFS) have recently evolved into key platforms for synthetic biology applications. Many synthetic biology tools have traditionally relied on cell-based systems, and while their adoption has shown great progress, the constraints inherent to the use of cellular hosts have limited their reach and scope. Cell-free systems, which can be thought of as programmable liquids, have removed many of these complexities and have brought about exciting opportunities for rational design and manipulation of biological systems. Here we review how these simple and accessible enzymatic systems are poised to accelerate the rate of advancement in synthetic biology and, more broadly, biotechnology.

Co- and Post-Translational Protein Folding in the ER

The biophysical rules that govern folding of small, single-domain proteins in dilute solutions are now quite well understood. The mechanisms underlying co-translational folding of multidomain and membrane-spanning proteins in complex cellular environments are often less clear. The endoplasmic reticulum (ER) produces a plethora of membrane and secretory proteins, which must fold and assemble correctly before ER exit - if these processes fail, misfolded species accumulate in the ER or are degraded. The ER differs from other cellular organelles in terms of the physicochemical environment and the variety of ER-specific protein modifications. Here, we review chaperone-assisted co- and post-translational folding and assembly in the ER and underline the influence of protein modifications on these processes. We emphasize how method development has helped advance the field by allowing researchers to monitor the progression of folding as it occurs inside living cells, while at the same time probing the intricate relationship between protein modifications during folding.

Optimization of a miniaturized fluid array device for cell-free protein synthesis

Cell-free protein synthesis (CFPS), which entails synthesizing proteins outside of intact cells, is conducted in several formats with the continuous-exchange cell-free (CECF) format generally having the greatest protein expression yields. With this format, continuous chemical exchange occurs through a dialysis membrane separating a reaction solution from a feeding solution containing supplemental nutrient/energy molecules. Here, we describe the optimization of the miniaturized fluid array device (µFAD) by studying the effects of structural and experimental parameters responsible for the heightened chemical exchange across the dialysis membranes and enhanced protein expression capabilities of the high-throughput device. The interface area and number between the reaction and feeding solutions have a direct impact on protein expression, with a 1.6% enhancement in protein expression yield with each square millimeter increase in area and a 20% decrease with each additional interface. For nutrient/energy availability, an increasing solution volume ratio and height difference increase protein expression yield until the expression yield plateaus at a volume ratio of 20 to 1 (feeding to reaction solution) and a solution height difference of 2 mm. This yield can be further increased by 7% every 30 min with feeding solution replacement. Of the studied experimental factors (feeding solution stirring, device shaking, and temperature increase), feeding solution stirring has a significant effect on protein expression in this device. In the optimized system, green fluorescent protein (GFP), ß-glucuronidase (GUS), ß-galactosidase (LacZ), luciferase, and tissue plasminogen activator (tPA) expression increased 77.8-, 212-, 3.66-, 463-, and 5.43-fold, respectively, compared to the conventional batch format in a standard microplate. These results highlight the significance of structural/experimental conditions on the productive expression of proteins in the CECF format.

A continuous-exchange cell-free protein synthesis system based on extracts from cultured insect cells

In this study, we present a novel technique for the synthesis of complex prokaryotic and eukaryotic proteins by using a continuous-exchange cell-free (CECF) protein synthesis system based on extracts from cultured insect cells. Our approach consists of two basic elements: First, protein synthesis is performed in insect cell lysates which harbor endogenous microsomal vesicles, enabling a translocation of de novo synthesized target proteins into the lumen of the insect vesicles or, in the case of membrane proteins, their embedding into a natural membrane scaffold. Second, cell-free reactions are performed in a two chamber dialysis device for 48 h. The combination of the eukaryotic cell-free translation system based on insect cell extracts and the CECF translation system results in significantly prolonged reaction life times and increased protein yields compared to conventional batch reactions. In this context, we demonstrate the synthesis of various representative model proteins, among them cytosolic proteins, pharmacological relevant membrane proteins and glycosylated proteins in an endotoxin-free environment. Furthermore, the cell-free system used in this study is well-suited for the synthesis of biologically active tissue-type-plasminogen activator, a complex eukaryotic protein harboring multiple disulfide bonds.

Glycosylation: impact, control and improvement during therapeutic protein production

The emergence of the biopharmaceutical industry represented a major revolution for modern medicine, through the development of recombinant therapeutic proteins that brought new hope for many patients with previously untreatable diseases. There is a ever-growing demand for these therapeutics that forces a constant technological evolution to increase product yields while simultaneously reducing costs. However, the process changes made for this purpose may also affect the quality of the product, a factor that was initially overlooked but which is now a major focus of concern. Of the many properties determining product quality, glycosylation is regarded as one of the most important, influencing, for example, the biological activity, serum half-life and immunogenicity of the protein. Consequently, monitoring and control of glycosylation is now critical in biopharmaceutical manufacturing and a requirement of regulatory agencies. A rapid evolution is being observed in this context, concerning the influence of glycosylation in the efficacy of different therapeutic proteins, the impact on glycosylation of a diversity of parameters/processes involved in therapeutic protein production, the analytical methodologies employed for glycosylation monitoring and control, as well as strategies that are being explored to use this property to improve therapeutic protein efficacy (glycoengineering). This work reviews the main findings on these subjects, providing an up-to-date source of information to support further studies.

Protein glycosylation control in mammalian cell culture: past precedents and contemporary prospects

Protein glycosylation is a post-translational modification of paramount importance for the function, immunogenicity, and efficacy of recombinant glycoprotein therapeutics. Within the repertoire of post-translational modifications, glycosylation stands out as having the most significant proven role towards affecting pharmacokinetics and protein physiochemical characteristics. In mammalian cell culture, the understanding and controllability of the glycosylation metabolic pathway has achieved numerous successes. However, there is still much that we do not know about the regulation of the pathway. One of the frequent conclusions regarding protein glycosylation control is that it needs to be studied on a case-by-case basis since there are often conflicting results with respect to a control variable and the resulting glycosylation. In attempts to obtain a more multivariate interpretation of these potentially controlling variables, gene expression analysis and systems biology have been used to study protein glycosylation in mammalian cell culture. Gene expression analysis has provided information on how glycosylation pathway genes both respond to culture environmental cues, and potentially facilitate changes in the final glycoform profile. Systems biology has allowed researchers to model the pathway as well-defined, inter-connected systems, allowing for the in silico testing of pathway parameters that would be difficult to test experimentally. Both approaches have facilitated a macroscopic and microscopic perspective on protein glycosylation control. These tools have and will continue to enhance our understanding and capability of producing optimal glycoform profiles on a consistent basis.

Aberrant localization and underglycosylation of highly accumulating single-chain Fv-Fc antibodies in transgenic Arabidopsis seeds

Production of high-value recombinant proteins in transgenic seeds is an attractive and economically feasible alternative to conventional systems based on mammalian cells and bacteria. In contrast to leaves, seeds allow high-level accumulation of recombinant proteins in a relatively small volume and a stable environment. We demonstrate that single-chain variable fragment (scFv)-Fc antibodies, with N-terminal signal sequence and C-terminal KDEL tag, can accumulate to very high levels as bivalent IgG-like antibodies in Arabidopsis thaliana seeds and illustrate that a plant-produced anti-hepatitis A virus scFv-Fc has similar antigen-binding and in vitro neutralizing activities as the corresponding full-length IgG. As expected, most scFv-Fc produced in seeds contained only oligomannose-type N-glycans, but, unexpectedly, 35-40% was never glycosylated. A portion of the scFv-Fc was found in endoplasmic reticulum (ER)-derived compartments delimited by ribosome-associated membranes. Additionally, consistent with the glycosylation data, large amounts of the recombinant protein were deposited in the periplasmic space, implying a direct transport from the ER to the periplasmic space between the plasma membrane and the cell wall. Aberrant localization of the ER chaperones calreticulin and binding protein (BiP) and the endogenous seed storage protein cruciferin in the periplasmic space suggests that overproduction of recombinant scFv-Fc disturbs normal ER retention and protein-sorting mechanisms in the secretory pathway.

Definition of the bacterial N-glycosylation site consensus sequence

The Campylobacter jejuni pgl locus encodes an N-linked protein glycosylation machinery that can be functionally transferred into Escherichia coli. In this system, we analyzed the elements in the C. jejuni N-glycoprotein AcrA required for accepting an N-glycan. We found that the eukaryotic primary consensus sequence for N-glycosylation is N terminally extended to D/E-Y-N-X-S/T (Y, X not equalP) for recognition by the bacterial oligosaccharyltransferase (OST) PglB. However, not all consensus sequences were N-glycosylated when they were either artificially introduced or when they were present in non-C. jejuni proteins. We were able to produce recombinant glycoproteins with engineered N-glycosylation sites and confirmed the requirement for a negatively charged side chain at position -2 in C. jejuni N-glycoproteins. N-glycosylation of AcrA by the eukaryotic OST in Saccharomyces cerevisiae occurred independent of the acidic residue at the -2 position. Thus, bacterial N-glycosylation site selection is more specific than the eukaryotic equivalent with respect to the polypeptide acceptor sequence.

Determination and Characterization of Site-Specific N-Glycosylation Using MALDI-Qq-TOF Tandem Mass Spectrometry: Case Study with a Plant Protease

MALDI tandem mass spectrometry analysis on a hybrid quadrupole-quadrupole time-of-flight (Qq-TOF) instrument was used in combination with two-dimensional gel electrophoresis, proteolytic digestion, and liquid chromatography for identification and structural characterization of glycosylation in a novel glycoprotein, pathogenesis-related subtilisin-like proteinase P69B from tomato. Glycopeptide fractions from microcolumn reversed-phase HPLC deposited on MALDI targets were identified from MS by their specific m/z spacing patterns (203, 162, 146 u) between glycoforms. In most cases, MS/MS spectra of [M + H]+ ions of glycopeptides featured peaks useful for determining sugar compositions, peptide sequences, and thus probable glycosylation sites. Furthermore, peptide-related product ions could readily be used in database search procedures to identify the glycoprotein. Four out of five predicted glycosylation sites were biologically relevant and occupied by five N-linked glycan side chains each. In addition, the fragmentation efficiency allowed detection of further modification of methionine-containing glycoforms with either oxidized or iodoacetamide alkylated methionine. The high resolution furnished by MALDI-Qq-TOF allowed rapid and sensitive structural characterization of site-specific N-glycosylation from a limited quantity of material and revealed heterogeneity at different levels, including different glycan side-chain modifications, and heterogeneity of oligosaccharide structures on the same glycosylation site.

The past, present and future of cell-free protein synthesis

Recent technical advances have revitalized cell-free expression systems to meet the increasing demands for protein synthesis. Cell-free systems offer several advantages over traditional cell-based expression methods, including the easy modification of reaction conditions to favor protein folding, decreased sensitivity to product toxicity and suitability for high-throughput strategies because of reduced reaction volumes and process time. Moreover, improvements in translation efficiency have resulted in yields that exceed a milligram of protein per milliliter of reaction mix. We review the advances on this expanding technology and highlight the growing list of associated applications.

Enhancing multiple disulfide bonded protein folding in a cell-free system

A recombinant plasminogen activator (PA) protein with nine disulfide bonds was expressed in our cell-free protein synthesis system. Due to the unstable and reducing environment in the initial E. coli-based cell-free system, disulfide bonds could not be formed efficiently. By treating the cell extract with iodoacetamide and utilizing a mixture of oxidized and reduced glutathione, a stabilized redox potential was optimized. Addition of DsbC, replacing polyethylene glycol with spermidine and putrescine to create a more natural environment, adding Skp, an E. coli periplasmic chaperone, and expressing PA at 30 degrees C increased the solubility of the protein product as well as the yield of active PA. Taken together, the modifications enabled the production of more than 60 microg/mL of bioactive PA in a simple 3-h batch reaction.

Effect of shear stress on intrinsic CHO culture state and glycosylation of recombinant tissue-type plasminogen activator protein

Shear stress in suspension culture was investigated as a possible manipulative parameter for the control of glycosylation of the recombinant tissue-type plasminogen activator protein (r-tPA) produced by recombinant Chinese hamster ovary (CHO) cell culture, grown in protein-free media. Resulting fractions of partially glycosylated, Type II, and fully glycosylated, Type I, r-tPA protein were monitored as a direct function of the shear characteristics of the culture environment. The shear-induced response of CHO culture to levels of low shear stress, where exponential growth was not obtained, and to higher levels of shear stress, which resulted in extensive cell death, were examined through manipulation of the bioreactor stirring velocity. Both apparent and intrinsic cell growth, metabolite consumption, byproduct and r-tPA production, and r-tPA glycosylation, from a variable site-occupancy standpoint, were monitored throughout. Kinetic analyses revealed a shear-stress-induced alteration of cellular homeostasis resulting in a nonlinear dependency of metabolic yield coefficients and an intrinsic cell lysis kinetic constant on shear stress. Damaging levels of shear stress were used to investigate the shear dependence of cell death and lysis, as well as the effects on the intrinsic growth rate of the culture. Kinetic models were also developed on the basis of the intrinsic state of the culture and compared to traditional models. Total r-tPA production was maximized under moderate shear conditions, as was the viable CHO cell density of the culture. However, Type II r-tPA production and the fraction of Type II glycoform production ratio was maximized under damaging levels of shear stress. Analyses of biomass production yield coefficients coupled with a plug-flow reactor model of glycan addition in the endoplasmic reticulum (ER) were used to propose an overall mechanism of decreased r-tPA protein site-occupancy glycosylation with increasing shear stress. Decreased residence time of r-tPA in the ER as a result of increased protein synthesis related to shear protection mechanisms is proposed to limit contact of site Asn184 with the membrane-bound oligosaccharyltransferase enzyme in the ER.

N-Glycosylation at two sites critically alters thiazide binding and activity of the rat thiazide-sensitive Na(+):Cl(-) cotransporter

The rat thiazide-sensitive Na-Cl cotransporter (rNCC) is expressed in the renal distal convoluted tubule and is the site of action of an important class of antihypertensive agents, the thiazide diuretics. The amino acid sequence contains two potential N-linked glycosylation consensus sites, N404 and N424. Either enzymatic deglycosylation or tunicamycin reduced the cotransporter to its core molecular weight (113 kD). Glycosylation site single mutants expressed in oocytes ran as thick bands at 115 kD, consistent with the high-mannose glycoprotein. The double mutant produced the single thin 113-kD band seen in the deglycosylated cotransporter. Functional expression of cotransporters in Xenopus laevis oocytes revealed that the mutants displayed drastically decreased thiazide-sensitive (22)Na(+) uptake compared with wild-type NCC. Analysis of enhanced green fluorescence protein (EGFP)-tagged cotransporters demonstrated that this decrease in function is predominantly secondary to decreased surface expression. The elimination of glycosylation in the double mutant increased thiazide sensitivity by more than two orders of magnitude and also increased Cl(-) affinity. Thus, we have demonstrated that rNCC is N-glycosylated in vivo at two sites, that glycosylation is essential for efficient function and surface expression of the cotransporter, and that the elimination of glycosylation allows much greater access of thiazide diuretics to their binding site.

Protein folding during cotranslational translocation in the endoplasmic reticulum

To test how far into the protein-conducting channel of the translocon complex a nascent polypeptide domain must move before it can fold, we analyzed the folding of in vitro translated products of truncated mRNAs encoding the Semliki Forest virus capsid protease domain (Cp) during translocation into microsomes. Cp folded when the C-terminal linker connecting it to the peptidyltransferase center was 64 amino acids or longer. This means that to fold, Cp must exit the translocon channel. With an uncleaved signal sequence, about one out of four of the Cp domains could undergo folding with a C-terminal linker of only 38-66 amino acids. This suggested that the constraint imposed on folding by the translocon complex may be less stringent for signal-anchored membrane proteins.

Multiple cell culture factors can affect the glycosylation of Asn-184 in CHO-produced tissue-type plasminogen activator

Human tissue-type plasminogen activator (t-PA) contains a variably occupied glycosylation site at Asn-184 in naturally produced t-PA and in t-PA produced in recombinant Chinese hamster ovary (CHO) cells. The presence of an oligosaccharide at this site has previously been shown to reduce specific activity and fibrin binding. In this report, the site occupancy of t-PA is shown to increase gradually over the course of batch and fed-batch CHO cultures. Additional cell culture factors, including butyrate and temperature, are also shown to influence the degree of glycosylation. In each of these cases, conditions with decreased growth rate correlate with increased site occupancy. Investigations using quinidine and thymidine to manipulate the cell cycle distribution of cultures further support this correlation between site occupancy and growth state. Comparison of the cell cycle distribution across the range of cell culture factors investigated shows a consistent relationship between site occupancy and the fraction of cells in the G(0)/G(1) phase of the cell cycle. These results support a correlation between growth state and site occupancy, which fundamentally differs from site occupancy trends previously observed and illustrates the importance of the growth profile of CHO cultures in producing consistently glycosylated recombinant glycoproteins.

Glycosylation of the hepatitis C virus envelope protein E1 is dependent on the presence of a downstream sequence on the viral polyprotein

The addition of N-linked oligosaccharides to Asn-X-(Ser/Thr) sites is catalyzed by the oligosaccharyltransferase, an enzyme closely associated with the translocon and generally thought to have access only to nascent chains as they emerge from the ribosome. However, the presence of the sequon does not automatically ensure core glycosylation because many proteins contain sequons that remain either nonglycosylated or glycosylated to a variable extent. In this study, hepatitis C virus (HCV) envelope protein E1 was used as a model to study the efficiency of N-glycosylation. HCV envelope proteins, E1 and E2, were released from a polyprotein precursor after cleavage by host signal peptidase(s). When expressed alone, E1 was not efficiently glycosylated. However, E1 glycosylation was improved when expressed as a polyprotein including full-length or truncated forms of E2. These data indicate that glycosylation of E1 is dependent on the presence of polypeptide sequences located downstream of E1 on HCV polyprotein.

Glycosylation of the overlapping sequons in yeast external invertase: effect of amino acid variation on site selectivity in vivo and in vitro

Yeast invertase contains 14 sequons, all of which are glycosylated to varying degrees except for sequon 5 which is marginally glycosylated, if at all. This sequon overlaps with sequon 4 in a sequence consisting of Asn92-Asn93-Thr94-Ser95(Reddy et al., 1988, J. Biol. Chem., 263, 6978-6985). To determine whether glycosylation at Asn93is sterically hindered by the oligosaccharide on Asn92, the latter amino acid was converted to a glutamine residue by site-directed mutagenesis of the SUC2 gene in a plasmid vector which was expressed in Saccharomyces cerevisiae. A glycopeptide encompassing sequons 3 through 6 was purified from a tryptic digest of the mutagenized invertase and sequenced by Edman degradation, which revealed that Asn93 of sequon 5 contained very little, if any, carbohydrate, despite the elimination of sequon 4. When Ser and Thr were inverted to yield Asn-Asn-Ser-Thr carbohydrate was associated primarily with the second sequon, in agreement with numerous studies indicating that Asn-X-Thr is preferred to Asn-X-Ser as an oligosaccharide acceptor. However, when the invertase overlapping sequons were converted to Asn-Asn-Ser-Ser, both sequons were clearly glycosylated, with the latter sequon predominating. These findings rule out steric hindrance as a factor involved in preventing the glycosylation of sequon 5 in invertase. Comparable results were obtained using an in vitro system with sequon-containing tri- and tetrapeptides acceptors, in addition to larger oligosaccharide acceptors.

The oligosaccharyltransferase complex from yeast

N-Glycosylation of eukaryotic secretory and membrane-bound proteins is an essential and highly conserved protein modification. The key step of this pathway is the en bloc transfer of the high mannose core oligosaccharide Glc3Man9GlcNAc2 from the lipid carrier dolichyl phosphate to selected Asn-X-Ser/Thr sequences of nascent polypeptide chains during their translocation across the endoplasmic reticulum membrane. The reaction is catalysed by the enzyme oligosaccharyltransferase (OST). Recent biochemical and molecular genetic studies in yeast have yielded novel insights into this enzyme with multiple tasks. Nine proteins have been shown to be OST components. These are assembled into a heterooligomeric membrane-bound complex and are required for optimal expression of OST activity in vivo in wild type cells. In accord with the evolutionary conservation of core N-glycosylation, there are significant homologies between the protein sequences of OST subunits from yeast and higher eukaryotes, and OST complexes from different sources show a similar organisation as well.

Post-translational N-glycosylation of a truncated form of a peptide processing enzyme

Peptidylglycine alpha-amidating monooxygenase (PAM) catalyzes the carboxyl-terminal amidation of bioactive peptides through a two-step reaction involving the monooxygenase and lyase domains. PAM undergoes endoproteolytic cleavage in neuroendocrine cells in the lyase domain. To determine which of the two possible paired basic sites is utilized, truncated PAM proteins ending at these sites were stably expressed in Chinese hamster ovary cells. While characterizing the truncation mutants, it became apparent that N-glycosylation occurred post-translationally at the single site localized near the carboxyl terminus of the lyase domain. The post-translational N-glycosylation of this site does not require the newly synthesized protein to remain tightly bound to membranes. Both malfolded, secretion incompetent proteins and fully active, secreted proteins were subject to post-translational N-glycosylation.

The amino acid following an asn-X-Ser/Thr sequon is an important determinant of N-linked core glycosylation efficiency

Many eukaryotic proteins are modified by Asn-linked (N-linked) glycosylation. The number and position of oligosaccharides added to a protein by the enzyme oligosaccharyltransferase can influence its expression and function. N-Linked glycosylation usually occurs at Asn residues in Asn-X-Ser/Thr sequons where X not equal Pro. However, many Asn-X-Ser/Thr sequons are not glycosylated or are glycosylated inefficiently. Inefficient glycosylation at one or more Asn-X-Ser/Thr sequons in a protein results in the production of heterogeneous glycoprotein products. These glycoforms may differ from one another in their level of expression, stability, antigenicity, or function. The signals which control the efficiency of N-linked glycosylation at individual Asn residues have not been fully defined. In this report, we use a site-directed mutagenesis approach to investigate the influence of the amino acid at the position following a sequon (the Y position, Asn-X-Ser/Thr-Y). Variants of rabies virus glycoprotein containing a single Asn-X-Ser/Thr sequon at Asn37 were generated. Variants were designed with each of the twenty common amino acids at the Y position, with either Ser or Thr at the hydroxy (Ser/Thr) position. The core glycosylation efficiency of each variant was quantified using a cell-free translation/glycosylation system. These studies reveal that the amino acid at the Y position is an important determinant of core glycosylation efficiency.

Calnexin and calreticulin bind to enzymically active tissue-type plasminogen activator during biosynthesis and are not required for folding to the native conformation

The roles of the endoplasmic-reticulum lectins calnexin and calreticulin in the folding of tissue-type plasminogen activator (tPA) have been investigated using an in vitro translation system that reconstitutes these processes as they would occur in the intact cell. Using co-immunoprecipitation of newly synthesized tPA with antibodies to calnexin and calreticulin, it was demonstrated that the interaction of tPA with both lectins was dependent upon tPA glycosylation and glucosidase trimming. When tPA was synthesized in the presence of semi-permeabilized cells under conditions preventing complex formation with calnexin and calreticulin, the translation product had a specific plasminogenolytic activity identical with that when synthesized under conditions permitting interactions with both lectins. Furthermore, complexes of tPA bound to calnexin and calreticulin were shown to be enzymically active. These results demonstrate that calnexin and calreticulin can form a stable interaction with correctly folded tPA; however, such interactions are not required for the synthesis of enzymically active tPA.

Regulation of receptor binding affinity of influenza virus hemagglutinin by its carbohydrate moiety

The hemagglutinin (HA) of the fowl plague virus (FPV) strain of influenza A virus has two N-linked oligosaccharides attached to Asn123 and Asn149 in the vicinity of the receptor binding site. The effect of these carbohydrate side chains on the binding of HA to neuraminic acid-containing receptors has been analyzed. When the oligosaccharides were deleted by site-specific mutagenesis, HA expressed from a simian virus 40 vector showed enhanced hemadsorbing activity. Binding was so strong under these conditions that erythrocytes were no longer released by viral neuraminidase and that release was significantly reduced when neuraminidase from Vibrio cholerae was used. Similarly, when these oligosaccharides were removed selectively from purified viruses by N-glycosidase F, such virions were unable to elute from receptors, although they retained neuraminidase activity. Thus, release of FPV from cell receptors depends on the presence of the HA glycans at Asn123 and Asn149. On the other hand, receptor binding was abolished when these oligosaccharides were sialylated after expression in the absence of neuraminidase (M. Ohuchi, A. Feldmann, R. Ohuchi, and H.-D. Klenk, Virology 212:77-83, 1995). These observations indicate that the receptor affinity of FPV HA is controlled by oligosaccharides adjacent to the receptor binding site.

Glycosylation: heterogeneity and the 3D structure of proteins

Glycoproteins generally exist as populations of glycosylated variants (glycoforms) of a single polypeptide. Although the same glycosylation machinery is available to all proteins that enter the secretory pathway in a given cell, most glycoproteins emerge with characteristic glycosylation patterns and heterogeneous populations of glycans at each glycosylation site. The factors that control the composition of the glycoform populations and the role that heterogeneity plays in the function of glycoproteins are important questions for glycobiology. A full understanding of the implications of glycosylation for the structure and function of a protein can only be reached when a glycoprotein is viewed as a single entity. Individual glycoproteins, by virtue of their unique structures, can selectively control their own glycosylation by modulating interactions with the glycosylating enzymes in the cell. Examples include protein-specific glycosylation within the immunoglobulins and immunoglobulin superfamily and site-specific processing in ribonuclease, Thy-1, IgG, tissue plasminogen activator, and influenza A hemagglutinin. General roles for the range of sugars on glycoproteins such as the leukocyte antigens include orientating the molecules on the cell surface. A major role for specific sugars is in recognition by lectins, including chaperones involved in protein folding. In addition, the recognition of identical motifs in different glycans allows a heterogeneous population of glycoforms to participate in specific biological interactions.

Getting the glycosylation right: implications for the biotechnology industry

Glycosylation is the most extensive of all the posttranslational modifications, and has important functions in the secretion, antigenicity and clearance of glycoproteins. In recent years major advances have been made in the cloning of glycosyltransferase enzymes, in understanding the varied biological functions of carbohydrates, and in the accurate analysis of glycoprotein heterogeneity. In this review we discuss the impact of these advances on the choice of a recombinant host cell line, in optimizing cell culture processes, and in choosing the appropriate level of glycosylation analysis for each stage of product development.

The amino acid at the X position of an Asn-X-Ser sequon is an important determinant of N-linked core-glycosylation efficiency

N-Linked glycosylation is a common form of protein processing that can profoundly affect protein expression, structure, and function. N-Linked glycosylation generally occurs at the sequon Asn-X-Ser/Thr, where X is any amino acid except Pro. To assess the impact of the X amino acid on core glycosylation, rabies virus glycoprotein variants were generated by site-directed mutagenesis with each of the 20 common amino acids substituted at the X position of an Asn-X-Ser sequon. The efficiency of core glycosylation at the sequon in each variant was quantified in a rabbit reticulocyte lysate cell-free translation system supplemented with canine pancreas microsomes. The presence of Pro at the X position completely blocked core glycosylation, whereas Trp, Asp, Chi, and Leu were associated with inefficient core glycosylation. The other variants were more efficiently glycosylated, and several were fully glycosylated. These findings demonstrate that the X amino acid is an important determinant of N-linked core-glycosylation efficiency.

IgM polymerization inhibits the Golgi-mediated processing of the mu-chain carboxy-terminal glycans

Secreted glycoproteins generally contain oligosaccharides of the complex type. However, several molecules have been described in which individual glycans are processed differently from one another. Folding, assembly and oligomerization could affect the maturation of certain glycans by hindering them to the Golgi processing machinery. We have tested this possibility by analysing a panel of engineered murine mu chains secreted as mu2L2 monomers or as polymers, and having or not the carboxy-terminal glycan (Asn563). In secreted IgM polymers, Asn563 bears high-mannose oligosaccharides, typical of endoplasmic reticulum resident proteins, while complex sugars are found at the other four sites (Brenckle and Kornfeld, 1980 Arch. Biochem. Biophys. 243, 605-618). Polymeric and monomeric IgM contain mu chains whose glycans are processed differently. We show here that this is mainly due to the differential processing at the Asn563 glycan, which undergoes Golgi-mediated processing when IgM are secreted in the monomeric form. These results indicate that the oligomerization-dependent accessibility to the sugar modifying enzymes can be one of the key features that dictate the extent of oligosaccharide processing in multimeric glycoproteins. The presence of high mannose glycans at Asn563 implies that IgM polymerization takes place before encountering mannosidase II, likely in a pre-Golgi compartment.

Folding of human intestinal lactase-phlorizin hydrolase

The folding of human intestinal prolactase-phlorizin hydrolase (pro-LPH) has been analyzed in a cell-free transcription/translation system. In the presence of the thiol oxidant GSSG, disulfide bond formation in pro-LPH can be promoted concomitant with the binding of the molecule to a conformation-specific monoclonal anti-LPH antibody. Under these conditions, pro-LPH does not bind to the molecular chaperone BiP. In the absence of GSSG, on the other hand, pro-LPH does not bind to the monoclonal anti-LPH antibody, but can be immunoprecipitated with a polyclonal antibody that is directed against a denatured form of the enzyme. In this case, interaction of pro-LPH with immunoglobulin heavy chain binding protein can be discerned. The results demonstrate the existence of intramolecular disulfide bonds that are essential for the promotion of pro-LPH to a native conformation. Furthermore, BiP is involved in the folding events of pro-LPH.

Partial glycosylation of antithrombin III asparagine-135 is caused by the serine in the third position of its N-glycosylation consensus sequence and is responsible for production of the beta-antithrombin III isoform with enhanced heparin affinity

Two antithrombin III (ATIII) isoforms occur naturally in human plasma. The alpha-ATIII isoform has four N-linked oligosaccharides attached to asparagines 96, 135, 155, and 192. The beta-ATIII isoform lacks carbohydrate on asparagine-135 (N135), which is near the heparin binding site, and binds heparin with higher affinity than does alpha-ATIII. Two isoforms are also produced when the normal human ATIII cDNA sequence is expressed in baculovirus-infected insect cells, and the recombinant beta' isoform similarly binds heparin with higher affinity than the recombinant alpha' isoform. Consensus sequences (CSs) of the ATIII N-glycosylation sites are N-X-S for 135 and N-X-T for 96, 155, and 192. On the basis of database and in vitro glycosylation studies suggesting that N-X-S CSs are utilized less efficiently than N-X-T CSs, we hypothesized that the beta-ATIII isoform might result from inefficient core glycosylation of the N135 N-X-S CS due to the presence of a serine, rather than a threonine, in the third position. ATIIIs with N-X-S, N-X-T, and N-X-A consensus sequences were expressed in baculovirus-infected insect cells. In contrast to the N-X-S sequence, which expressed a mixture of alpha' and beta' molecules, the N-X-T variant produced alpha' exclusively, while the N-X-A variant produced beta' exclusively. Thus, serine in the third position of the N135 CS is responsible for its "partial" glycosylation and leads to production of beta-ATIII.(ABSTRACT TRUNCATED AT 250 WORDS)

The hydroxy amino acid in an Asn-X-Ser/Thr sequon can influence N-linked core glycosylation efficiency and the level of expression of a cell surface glycoprotein

N-Linked glycosylation usually occurs at the sequon, Asn-X-Ser/Thr. In this sequon, the side chain of the hydroxy amino acid (Ser or Thr) may play a direct catalytic role in the enzymatic transfer of core oligosaccharides to the Asn residue. Using recombinant variants of rabies virus glycoprotein (RGP), we examined the influence of the hydroxy amino acid on core glycosylation efficiency. A variant of RGP containing a single Asn-X-Ser sequon at Asn37 was modified by site-directed mutagenesis to change the sequon to either Asn-X-Cys or Asn-X-Thr. The impact of these changes on core glycosylation efficiency was assessed by expressing the variants in a cell-free transcription/translation/glycosylation system and in transfected tissue culture cells. Substitution of Cys at position 39 blocks glycosylation, whereas substitution of Thr dramatically increases core glycosylation efficiency of Asn37 in both membrane-anchored and secreted forms of RGP. The substitution of Thr for Ser also dramatically enhances the level of expression and cell surface delivery of RGP when the sequon at Asn37 is the only sequon in the protein. Novel forms of membrane-anchored and secreted RGP which are fully glycosylated at all three sequons were also generated by substitution of Thr at position 39.

The translocation, folding, assembly and redox-dependent degradation of secretory and membrane proteins in semi-permeabilized mammalian cells

We describe here a semi-permeabilized cell-system which reconstitutes the efficient synthesis, translocation, folding, assembly and degradation of membrane and secretory proteins. Cells grown in culture were treated with the detergent digitonin which selectively permeabilized the plasma membrane leaving the cellular organelles, such as the endoplasmic reticulum (ER) and trans-Golgi network intact. These permeabilized cells were added to an in vitro translation system, either wheatgerm or reticulocyte lysate, supplemented with RNA coding for either membrane or secretory proteins. Efficient translocation and modification of proteins by these cells was demonstrated by protease protection, photocross-linking of nascent chains to components of the translocation apparatus and by post-translational modifications such as glycosylation or hydroxylation. A comparison was made between the ability of semi-permeabilized cells and microsomal vesicles to fold and assemble proteins. The results show that the intact ER within these cells can assemble proteins much more efficiently than vesicularized ER. Furthermore, the semi-permeabilized cells carried out the redox-dependent degradation of tissue-type plasminogen activator. This system has all the advantages of conventional cell-free systems, including speed and, importantly, the ability to manipulate the components of the assay, while retaining intracellular organelles and, therefore, allowing cellular processes to occur as they would in the intact cell.

Intracellular folding of tissue-type plasminogen activator. Effects of disulfide bond formation on N-linked glycosylation and secretion

The addition of N-linked core oligosaccharides to membrane and secretory glycoproteins occurs co-translationally at asparagine residues in the tripeptide sequon Asn-Xaa-Ser/Thr soon after translocation of the nascent polypeptide into the lumen of the endoplasmic reticulum. However, the presence of the sequon does not automatically ensure core glycosylation, as many proteins contain sequons that remain either unglycosylated or glycosylated to a variable extent. To investigate whether intracellular protein folding can influence sequon utilization, we have expressed tissue-type plasminogen activator (t-PA) in cell culture in the presence of mild concentrations of the reducing agent dithiothreitol to prevent co-translational disulfide bond formation in the endoplasmic reticulum. We show that conditions that prevent disulfide bond formation lead to complete glycosylation of a sequon that otherwise undergoes variable glycosylation in untreated cells. This demonstrated that folding and disulfide bond formation of t-PA determines its extent of core N-linked glycosylation. When dithiothreitol was removed from the cells, the reduced and overglycosylated t-PA formed disulfide bonds, folded, and was secreted. We also show t-PA present within cells is more susceptible to reduction with low concentrations of dithiothreitol than secreted t-PA.

The alpha subunit of the Saccharomyces cerevisiae oligosaccharyltransferase complex is essential for vegetative growth of yeast and is homologous to mammalian ribophorin I

Oligosaccharyltransferase mediates the transfer of a preassembled high mannose oligosaccharide from a lipid-linked oligosaccharide donor to consensus glycosylation acceptor sites in newly synthesized proteins in the lumen of the rough endoplasmic reticulum. The Saccharomyces cerevisiae oligosaccharyltransferase is an oligomeric complex composed of six nonidentical subunits (alpha-zeta), two of which are glycoproteins (alpha and beta). The beta and delta subunits of the oligosaccharyltransferase are encoded by the WBP1 and SWP1 genes. Here we describe the functional characterization of the OST1 gene that encodes the alpha subunit of the oligosaccharyltransferase. Protein sequence analysis revealed a significant sequence identity between the Saccharomyces cerevisiae Ost1 protein and ribophorin I, a previously identified subunit of the mammalian oligosaccharyltransferase. A disruption of the OST1 locus was not tolerated in haploid yeast showing that expression of the Ost1 protein is essential for vegetative growth of yeast. An analysis of a series of conditional ost1 mutants demonstrated that defects in the Ost1 protein cause pleiotropic underglycosylation of soluble and membrane-bound glycoproteins at both the permissive and restrictive growth temperatures. Microsomal membranes isolated from ost1 mutant yeast showed marked reductions in the in vitro transfer of high mannose oligosaccharide from exogenous lipid-linked oligosaccharide to a glycosylation site acceptor tripeptide. Microsomal membranes isolated from the ost1 mutants contained elevated amounts of the Kar2 stress-response protein.

Substitution of asparagine residues in Aspergillus awamori glucoamylase by site-directed mutagenesis to eliminate N-glycosylation and inactivation by deamidation

Aspergillus awamori glucoamylase is a secreted glycoprotein containing N-linked carbohydrate recognition sites at Asn-171, Asn-182 and Asn-395. Site-directed mutagenesis was performed at Asn-182 and Asn-395 to determine whether these residues were N-glycosylated by Saccharomyces cerevisiae, to investigate the function of any glycans linked to them, and to determine the effect of their deamidation on glucoamylase thermostability. Asn-171 and Asn-395, but not Asn-182, were N-glycosylated. Deletion of the glycan N-linked to Asn-395 did not affect specific activity, but greatly decreased enzyme secretion and thermostability. The mutant lacking the N-glycan linked to Asn-395 was synthesized very slowly, and was more associated with cell membrane components and susceptible to proteinase degradation than were wild-type or other mutant glucoamylases. Its secreted form was 30-fold less thermostable than wild-type enzyme at pH 4.5. Replacement of Asn-182 by Gln to eliminate deamidation at this site did not change glucoamylase specific activity or thermostability, while replacement by Asp decreased specific activity about 25%, but increased thermostability moderately at pH 4.5 below 70 degrees C. Both mutations of Asn-182 increased glucoamylase production.

The effect of protein synthesis inhibitors on the glycosylation site occupancy of recombinant human prolactin

The relationship between synthesis and N-linked glycosylation site occupancy of recombinant human prolactin produced from C127 cells was studied with the aid of a battery of protein synthesis inhibitors. Non-lethal concentrations of sodium fluoride, gougerotin, puromycin, anisomycin, and emetine did not alter site occupancy, but low concentrations (< 10 micrograms ml-1) of cycloheximide increased the fraction of secreted prolactin bearing oligosaccharide from 20% to 80% of the total. Cycloheximide is an inhibitor of the elongation step of protein synthesis. The observed increase in glycosylation site occupancy upon addition of cycloheximide is consistent with the current opinion that the initial glycosylation event occurs cotranslationally during a limited time period. Cycloheximide may extend this time period by reducing elongation rate. However, the absence of any effect from treatment with other inhibitors of elongation suggests that cycloheximide is unique in its behavior on this system.

Effect of lipid supplements on the production and glycosylation of recombinant interferon-gamma expressed in CHO cells

The effects of lipids on the glycosylation of recombinant human interferon-gamma expressed in a Chinese Hamster Ovary cell line were investigated in batch culture. Lipids form an essential part of the N-glycosylation pathway, and have been shown to improve cell viability. In control (serum-free) medium the proportion of fully-glycosylated interferon-gamma deteriorated reproducibly with time in batch culture, but the lipoprotein supplement ExCyte was shown to minimise this trend. Partially substituting the bovine serum albumin content of the medium with a fatty-acid free preparation also improved interferon-gamma glycosylation, possibly indicating that oxidised lipids carried on Cohn fraction V albumin may damage the glycosylation process.

Efficiency of N-linked core glycosylation at asparagine-319 of rabies virus glycoprotein is altered by deletions C-terminal to the glycosylation sequon

In N-linked core glycosylation, the oligosaccharide Glc3Man9GlcNAc2 is transferred to the tripeptide sequon Asn-X-Ser/Thr. However, this process must be regulated by additional protein signals, since many sequons are either poorly glycosylated or not glycosylated at all. Since N-linked glycosylation can influence protein structure and function, understanding these signals is essential for the design and expression of recombinant glycoproteins. Core glycosylation usually occurs cotranslationally in the rough endoplasmic reticulum (RER) during translocation of nascent proteins. Since only regions of a protein immediately near to a sequon or N-terminal to it are thought to be in the RER when core glycosylation occurs, most models predict that regions C-terminal to the sequon do not influence this process. We tested whether regions C-terminal to a sequon can influence its core glycosylation. Full-length (505 amino acid) rabies virus glycoprotein (RGP) mutants, each containing only one of the three sequons normally present in RGP, were used for these studies. Using a cell-free system, the core glycosylation efficiency at each sequon was determined. Termination codons were then introduced into these mutants at defined sites to produce C-terminal truncations, and the effect of each of these truncations on the core glycosylation efficiency at each sequon was assessed. While deletion of the C-terminal transmembrane and cytoplasmic domains did not affect core glycosylation, more extensive C-terminal deletions did result in altered core glycosylation in a site-specific fashion. Specifically, C-terminal truncations resulting in proteins containing 386 or 344 amino acids decreased the efficiency of core glycosylation at Asn319. This demonstrates that core glycosylation efficiency can be influenced by the presence or absence of regions in a protein more than 68 amino acids C-terminal to a specific glycosylation site.