Isolation of overproducing recombinant mammalian cell lines by a fast and simple selection procedure.

The expression of genes transfected into mammalian cells is critically dependent on both the copy number and the site of integration. This conclusion was derived from the transfection of a plasmid containing three linked genes, which guarantees the same dose of each gene in the recipient cell clones. The selection pressure imposed by the usual concentrations of the neomycin analogue G418 or puromycin alone is low and yields cell clones with widely differing transcription rates. Time-consuming screening procedures, which are generally applied to obtain high-yield cell lines, can be obviated by the simultaneous transfer of two drug-resistance genes (neo plus dhfr, or neo plus pac). By a combined selection procedure only cells with high-level expression for the first marker will survive upon exposure to increasing concentrations of the second antibiotic. Generally, these cells also exhibit high expression levels for the non-selected gene. Many of the selected clones exhibit increased copy numbers of the respective gene but some of them owe their high productivity to a favorable position of only one or a few copies of the gene. Screening for high-yield clones using the combined selection protocol introduced here provides rapid and simple access to authentic and mutant gene products.

Expression of immunogenically reactive diphtheria toxin fusion proteins under the control of the pR promoter of bacteriophage lambda.

The tox228 gene encoding the non-toxic, immunologically cross-reactive CRM228 mutant diphtheria toxin (DT) has been cloned downstream of the PR promoter and the cro translational initiation region of bacteriophage lambda carried by plasmid pCQV2 (Queen, 1983). Efficient transcription but no appreciable amount of a translational product corresponding to complete DT could be detected in Escherichia coli hosts. Deletion of 320 bp from the C-terminal region of the B-fragment of DT, and fusion of the truncated tox228 gene to lacZ yielded several hybrid beta-galactosidases (beta Gal) in an E. coli lon- strain in addition to beta Gal. The various DT fragments fused to beta Gal were immunologically reactive and were identified with antibodies specifically directed against the A- or the B-fragment of DT. Antibodies raised against the DT-beta Gal fusion proteins in guinea pigs cross-reacted with wild-type DT and its B-fragment and protected Vero cells in tissue culture against the lethal action of DT. Immunized guinea pigs survived upon injection of a five-fold lethal dose of wild-type DT.

Expression and characterization of human CD4:immunoglobulin fusion proteins.

Different chimeric antibody-like molecules consisting of the four human CD4 extracellular domains (amino acids 1-369) fused to different parts of human IgG1 and IgM heavy-chain constant regions have been created and expressed in mammalian cells. For both IgG1 and IgM fusion proteins, the best expression in COS cells was observed for molecules lacking the CH1 domain of the heavy-chain constant region. The chimeric molecules are potent inhibitors of human immunodeficiency virus (HIV) infection and HIV-mediated cytotoxicity. A CD4:IgG1 hinge fusion protein, which was analyzed in more detail, binds efficiently to HIV gp160 and human Fc receptors and shows complement-assisted inhibition of viral propagation in culture. Half-life studies after intravenous application of the latter human fusion protein into mice and monkeys showed significant prolongation of serum survival compared to soluble CD4. An IgG2b murine homolog of the human CD4:IgG1 hinge fusion protein was prepared and evaluated in mice, where it was found to be nontoxic and to have no detectable effect on the humoral response to soluble antigen.

Welcome to the Conference

No abstract

Characterization of recombinant factor XIIIa produced in Saccharomyces cerevisiae.

Recombinant factor XIIIa (FXIIIa), produced in Saccharomyces cerevisiae, was recovered as a fully active cytosolic component and rigorously compared to natural F XIIIa from human placenta with respect to physicochemical and functional properties. Identical parameters were found in SDS polyacrylamide gel electrophoresis, analytical ultracentrifugation and HPLC gel filtration, and all spectral characteristics including derivative UV absorbance, fluorescence and circular dichroism were identical. Similarly, the interaction of both proteins with polyclonal antibodies directed against the entire FXIIIa or its N-terminal 4 kD activation peptide were identical. Furthermore, thrombin cleavage and fibrin cross-linking showed indistinguishable patterns. The only difference we observed was with respect to endgroup analysis. The recombinant protein is homogeneous, whereas placental FXIIIa shows multiple electrophoretic bands caused by microheterogeneity in the C-terminal part of the protein.

Limited proteolysis as a tool to study the kinetics of protein folding: conformational rearrangements in acid-dissociated lactic dehydrogenase as determined by pepsin digestion.

Noncovalent aggregation as a side reaction competing with the reconstitution of oligomeric enzymes is enhanced by slow conformational changes within the partially unfolded subunits. This has been shown for lactic dehydrogenase from pig muscle after acid dissociation [G. Zettlmeissl, R. Rudolph, and R. Jaenicke (1981) Eur. J. Biochem. 121, 169-175]. The present experiments confirm previous spectroscopic evidence (from circular dichroism) applying pepsin digestion and subsequent analysis of the fragments on sodium dodecyl sulfate-polyacrylamide gradient gels. The susceptibility of certain fragmentation sites toward pepsin digestion changes with increasing incubation at acid pH, in accordance with a slow M1 leads to M2 transition of the acid-dissociated monomers. Constant pulses of pepsin at varying times after transferring native enzyme to pH 2.3 yield distinct changes in the fragmentation pattern consisting of undigested monomers (Mr = 35,000) plus 12 fragments ranging from 31,000 to 5000. Short digestion of the M2 species at low concentrations of pepsin preferentially yields 25,000 and 10,500 fragments (molar ratio pepsin:lactic dehydrogenase = 1:24). The time-dependent decrease of monomers upon incubation in 0.1 M sodium phosphate, pH 2.3, at 20 degrees C strictly parallels the formation of the two fragments. The quantitative kinetic analysis of the changes in peptide pattern yields a first-order rate constant k1 = 8 +/- 2 X 10-4 s-1. The observed increase in proteolytic susceptibility is in the time range of the above mentioned decrease in the far-ultraviolet circular dichroism, and the parallel decrease in the yield of reactivation. The results suggest that during the M1 leads to M2 transition at acid pH a specific interdomain cleavage site is becoming exposed. As taken from the molecular weight of the two main fragments the trp 225-lys 226 peptide bond is the most probable candidate for this cleavage site.

Rate-determining folding and association reactions on the reconstitution pathway of porcine skeletal muscle lactic dehydrogenase after denaturation by guanidine hydrochloride.

Reactivation of tetrameric porcine skeletal muscle lactic dehydrogenase after dissociation and extensive unfolding of the monomers by 6 M guanidine hydrochloride (Gdn . HCl) is characterized by sigmoidal kinetics, indicating a complex mechanism involving rate-limiting folding and association steps. For analysis of the association reactions, chemical cross-linking with glutaraldehyde may be used [Hermann, R., Jaenicke, R., & Rudolph, R. (1981) Biochemistry 20, 2195-2201]. The data clearly show that the formation of a dimeric intermediate is determined by a first-order folding reaction of the monomers with k1 = (8.0 +/- 0.1) x 10(-4) s-1. The rate constant of the association of dimers to tetramers which represents the second rate-limiting step on the pathway of reconstitution after guanidine denaturation, was then determined by reactivation and cross-linking experiments after dissociation in 0.1 M H3PO4 containing 1 M Na2SO4. The rate constant for the dimer association (which is the only rate-limiting step after acid dissociation) was k2 = (3.0 +/- 0.5) x 10(4) M-1 s-1. On the basis of the given two rate constants, the complete reassociation pattern of porcine lactic dehydrogenase after dissociation and denaturation in 6 M Gdn . HCl can be described by the kinetic model (formula: see text).

Effects of low concentrations of guanidine . HCl on the reconstitution of lactic dehydrogenase from pig muscle in vitro. Evidence for guanidine binding to the native enzyme.

The presence of low concentrations of guanidine . HCl has a pronounced effect on the overall rate of reactivation of lactic dehydrogenase from pig muscles after preceding dissociation and deactivation in various denaturants. The obseverd attenuation is a function of the amount of guanidine . HCl present during reconstitution. At a given guanidine concentration in the reactivation buffer the yield, but not the rate of reactivation, is influenced by the extent of denaturation caused initially in the process of deactivation and dissociation. As a possible explanation for the influence of guanidine . HCl on the kinetics of reconstitution, binding of the ligand to intermediates of folding and association is considered. This hypothesis is corroborated by the observation that guanidine . HCl in the relevant concentration range does bind to native lactic dehydrogenase without inactivating the enzyme or disrupting its quaternary structure. A kinetic model comprising guanidine binding to both the native enzyme and structured intermediates is proposed to describe the observed effects of guanidine . HCl on the rate of reactivation. In addition, the dissociation constants for guanidine binding to intermediates of reconstitution and to native lactic dehydrogenase are estimated.

Reconstitution of lactic dehydrogenase. Noncovalent aggregation vs. reactivation. 2. Reactivation of irreversibly denatured aggregates.

Noncovalent aggregation is a side reaction in the process of reconstitution of oligomeric enzymes (e.g., lactic dehydrogenase) after preceding dissociation, denaturation, and deactivation. The aggregation product is of high molecular weight and composed of monomers which are trapped in a minium of conformational energy different from the one characterizing the native enzyme. This energy minimum is protected by a high activation energy of dissociation such that the aggregates are perfectly stable under nondenaturing conditions, and their degradation is provided only by applying strong denaturants, e.g., 6 M guanidine hydrochloride at neutral or acidic pH. The product of the slow redissolution process is the monomeric enzyme in its random configuration, which may be reactivated by diluting the denaturant under optimum conditions of reconstitution. The yield and the kinetics of reactivation of lactic dehydrogenase from pig skeletal muscle are not affected by the preceding aggregation-degradation cycle and are independent of different modes of aggregate formation (e.g., by renaturation at high enzyme concentration or heat aggregation). The kinetics of reactivation may be described by one single rate-determining bimolecular step with k2 = 3.9 x 10(4) M-1 s-1 at zero guanidine concentration. The reactivated enzyme consists of the native tetramer, characterized by enzymatic and physical properties identical with those observed for the enzyme in its initial native state.

Characterization of cDNA coding for human factor XIIIa.

A cDNA library prepared from human placenta has been screened for sequences coding for factor XIIIa, the enzymatically active subunit of the factor XIII complex that stabilizes blood clots through crosslinking of fibrin molecules. Two oligonucleotides, based on the amino acid sequences of tryptic peptides of factor XIIIa, were used as hybridization probes. Of 0.36 X 10(6) independent recombinants, 1 clone was identified that hybridized to both probes. The insert of 1704 base pairs coded for the amino-terminal 541 amino acid residues of the mature factor XIIIa molecule. Blot-hybridization analysis using this cDNA as a probe showed that the factor XIIIa mRNA from placenta has a size of approximately 4000 bases. The insert was used to rescreen cDNA libraries and to identify further factor XIIIa-specific sequences. The total length of the isolated factor XIIIa cDNA is 3905 bases, and it codes for a protein of 732 amino acids. In spite of the presence of factor XIII in blood plasma, we could not identify a leader sequence typical for secreted proteins.

A CD 4: immunoglobulin fusion protein with antiviral effects against HIV.

Antibody-like molecules consisting of the human CD 4 extracellular domain fused to human IgG1 heavy chain constant regions were genetically constructed and expressed in a BHK cell stable transfectant. Purified chimeric antibodies bound to HIV particles as it was shown by immuno electron microscopy, inhibited fusions of HIV-1-infected cells with uninfected cells, neutralized HIV-1, and were able to inhibit the spread of a cellular HIV-1 infection in CD 4+ cells. Plaque reduction assays with CD 4(+)-transfected Hela-cells showed a comparable inhibition of HIV-1 and HIV-2. Inhibitory functions were enhanced in the presence of complement. HIV-1- and HIV-2-infected CD 4+ cells were efficiently lysed by a slow, complement-dependent mechanism, whereas uninfected CD 4+ cells and HLA-DR+ cells were not affected.

Nucleotide sequence of a cluster of early and late genes in a conserved segment of the vaccinia virus genome.

The nucleotide sequence of a 7.6 kb vaccinia DNA segment from a genomic region conserved among different orthopox virus has been determined. This segment contains a tight cluster of 12 partly overlapping open reading frames most of which can be correlated with previously identified early and late proteins and mRNAs. Regulatory signals used by vaccinia virus have been studied. Presumptive promoter regions are rich in A, T and carry the consensus sequences TATA and AATAA spaced at 20-24 base pairs. Tandem repeats of a CTATTC consensus sequence are proposed to be involved in the termination of early transcription.

Diphtheria toxin promoter function in Corynebacterium diphtheriae and Escherichia coli.

The expression of the diphtheria tox228 gene encoding the nontoxic, serologically related CRM228 mutant diphtheria toxin has been analyzed in Corynebacterium diphtheriae and Escherichia coli. The diphtheria toxin promoter has been used to direct the expression of beta-galactosidase in E.coli, and the efficiency of promotion has been compared to that obtained with the lac promoter. Expression in C.diphtheriae is known to be dependent on the absence of iron, and we present for the first time direct evidence that this regulation occurs at the level of transcription. The 5' end of toxin mRNA maps at the same position in C.diphtheriae and E.coli, suggesting identical sequences to be recognized by C.diphtheriae and E.coli RNA polymerase. The diphtheria toxin promoter carries at position -34 a TTGATT sequence closely related to the E.coli -35 consensus sequence and in the -14 to -8 region a set of overlapping sequences with complete or partial homology to the E.coli -10 consensus sequence.

Molecular biology of proteins involved in blood coagulation.

Normal hemostasis in humans requires the interaction of a large number of plasma glycoproteins with blood platelets and vascular endothelial cells. Many of the plasma glycoproteins which participate in blood coagulation are zymogens of enzymes that interact in a stepwise manner in a series of reactions. In the last years most of these glycoproteins have been purified from human plasma by standard techniques. Some of them are used as therapeutics for restoring coagulation disorders. The knowledge about the plasma proteins involved in blood coagulation was greatly increased after cloning and sequencing of the respective complementary DNAs. Furthermore, recombinant DNA technology is used for the alternative production of several coagulation factors. It is the aim of this article to give an overview about the molecular biology of the enzymes and cofactors involved in blood coagulation.

Efficient expression system for human antithrombin III in baby hamster kidney cells.

A DNA fragment carrying the enhancer of an immediate early gene of human cytomegalovirus (hCMV) was tested as a transcriptional control element by fusion downstream of a transcription unit for human antithrombin III (AT III) driven by the Simian virus (SV40) enhancer and promoter. Measurement of transient AT III expression in baby hamster kidney cells (BHK) shows that by the presence of the hCMV enhancer the synthesis of AT III is increased considerably (three to fourfold). The AT III expression vectors carrying the hCMV enhancer were used to establish stable BHK cell-lines by a new and fast G418/methotrexate selection protocol, which express up to 12 micrograms AT III/10(6) cells/24h after 40-50 days. The given system might be generally useful for the fast expression of recombinant proteins in mammalian cells, e.g. the screening of altered AT III molecules obtained by site directed mutagenesis.

Denaturation-renaturation of the fibrin-stabilizing factor XIII a-chain isolated from human placenta. Properties of the native and reconstituted protein.

The denaturation-renaturation transition between the native and unfolded states of the dimeric blood coagulation factor XIIIa has been examined by far-UV circular dichroism, fluorescence spectroscopy, activity measurements, sedimentation equilibrium analysis, and size exclusion high performance liquid chromatography. Guanidine hydrochloride and urea-dependent denaturation in the absence and in the presence of 5mM dithioerythritol or glutathione (5mM GSH) exhibit biphasic transitions. The first stage represents a sharp transition characterized by a change in secondary structure without subunit dissociation. This step is accompanied by the irreversible loss of biological activity. The second transition reflects the dissociation and complete unfolding of the protein to a random coil. After loss of biological activity no reactivation can be accomplished under any of the following conditions: (i) denaturation and renaturation under reducing or non-reducing conditions, (ii) variation of the protein concentration and temperature, (iii) addition of specific ligands (Ca2+, substrate), (iv) presence of stabilizing and/or destabilizing agents. Attempts to renature the protein under standard conditions (0.1 M Tris/HCl pH 7.5-9.0, 5mM DTE, 5mM EDTA) lead to refolding intermediates which exhibit a strong tendency to aggregate. A soluble product of reconstitution can be obtained by refolding at low protein concentration, low temperature, and in the presence of small amounts of destabilizing agents such as arginine or urea in the renaturation buffer at pH 7.5 to 9. The spectroscopic and hydrodynamic characterization of the partially reconstituted (non-native inactive) protein shows that partially reconstituted factor XIIIa exhibits the fluorescence properties and the dimeric structure of the native protein.(ABSTRACT TRUNCATED AT 250 WORDS)

Isolation, physicochemical properties, and folding of octopine dehydrogenase from Pecten jacobaeus.

Two types (isoenzymes) of octopine dehydrogenase (A and B) from Pecten jacobaeus adductor muscle were purified to homogeneity, applying affinity chromatography as an efficient final step of purification. Both forms of the enzyme differ in their electrophoretic mobility. All other physico-chemical and enzymatic properties, as well as the folding behaviour were found to be identical. Interconversion of one form into the other was not detectable. Sedimentation equilibrium, gel permeation chromatography, and NaDodSO4/polyacrylamide gel electrophoresis yield a relative molecular mass of 45000 +/- 1500 for both native and denatured enzyme. The unfolding transition at varying guanidine X HCl concentrations is characterized by a two-step profile: at 0.4-0.8 M, partial unfolding is parallelled by inactivation; at 2.0-2.4 M the residual structure is destroyed in a second unfolding step. Beyond 2.8 M no further changes in fluorescence emission and dichroic absorption are observed. At 0.4-1.8 M guanidine X HCl, partial unfolding is superimposed by aggregation. The emission maximum of the intrinsic protein fluorescence at 327 nm is shifted to 352 nm upon denaturation in 6 M guanidine X HCl. Changes in the far-ultraviolet circular dichroism indicate complete loss of the overall backbone structure in this denaturant, including the native helix content of about 33%. Denaturation in 6 M guanidine X HCl, as monitored by the decrease of protein fluorescence, is fast (less than 8s). Upon reactivation after short denaturation, about 25% of the activity is recovered in a fast initial phase (less than 20s). The product of this phase has a similar stability towards destabilizing additives or proteases as the native enzyme. The slow phase of reactivation, which predominates after long-term denaturation, is determined by a single first-order reaction characterized by tau = 29 +/- 3 min (20 degrees C). This reaction must be a relatively late event on the folding pathway, preceded by the fast formation of a structured intermediate, as indicated by the immediate recovery of the native fluorescence. The structural rearrangements, which are rate-limiting for reactivation after long-term denaturation, are characterized by a high energy of activation (112 +/- 8 kJ/mol). The slow reactivation step is compatible in rate with the first-order folding reactions involved in the reconstitution of several oligomeric dehydrogenases [c.f. R. Jaenicke and R. Rudolph (1983) Colloq. Ges. Biol. Chem. Mosbach 34, 62-90].

Autoantibodies in HIV-infected hemophilia patients against different epitopes on CD4+ lymphocytes and recombinant CD4.

We studied 684 sera obtained from 20 hemophilia patients with AIDS/AIDS-related complex (ARC), 89 asymptomatic HIV+, 76 HIV- hemophilia patients and 151 healthy controls for antibodies against recombinant CD4 (rCD4). Twenty-two percent of AIDS/ARC patients, 10% of asymptomatic HIV+ patients, 17% of HIV-patients, and 1% of healthy controls had anti-rCD4 antibodies. Purified anti-rCD4 antibodies did not react with human CD4+ lymphocytes. This may explain why formation of anti-rCD4 antibodies correlated neither with the occurrence of autoantibodies against CD4+ lymphocytes nor with a decrease in CD4+ cell counts. Antibodies that were eluted from CD4+ lymphocytes after sequential adsorption and elution with separated CD8+ and CD4+ cells reacted with CD4+ lymphocytes of only some healthy individuals, suggesting diversity of CD4 expression.

Decreased affinity of recombinant antithrombin for heparin due to increased glycosylation.

Recombinant antithrombin produced by baby hamster kidney (BHK) or Chinese hamster ovary (CHO) cells was separated into two fractions, containing comparable amounts of protein, by affinity chromatography on matrix-linked heparin. Fluorescence titrations showed that the more tightly binding fraction had a heparin affinity similar to that of plasma antithrombin (Kd approximately 20 nM), whereas the affinity of the more weakly binding fraction was nearly 10-fold lower (Kd approximately 175 nM). Analyses of the heparin-catalysed rate of inhibition of thrombin further showed that the fractions differed only in their affinity for heparin and not in the intrinsic rate constant of either the uncatalysed or the heparin-catalysed inactivation of thrombin. The recombinant antithrombin fraction with lower heparin affinity migrated more slowly than both the fraction with higher affinity and plasma antithrombin in SDS/PAGE under reducing conditions, consistent with a slightly higher apparent relative molecular mass. This apparent size difference was abolished by the enzymic removal of the carbohydrate side chains from the proteins. Such removal also increased the heparin affinity of the weakly binding fraction, so that it eluted from matrix-linked heparin at a similar position to the deglycosylated tightly binding fraction or plasma antithrombin. Analyses of N-linked carbohydrate side chains showed that the weakly binding fraction from CHO cells had a higher proportion of tetra-antennary and a lower proportion of biantennary oligosaccharides than the tightly binding fraction. We conclude that the recombinant antithrombin produced by the two cell lines is heterogeneously glycosylated and that the increased carbohydrate content of a large proportion of the molecules results in a substantial decrease in the affinity of these molecules for heparin. These findings are of particular relevance for studies aimed at characterizing the heparin-binding site of recombinant antithrombin by site-directed mutagenesis.

Inactivation of recombinant plasmid DNA from a human erythropoietin-producing mouse cell line grown on a large scale.

Experiments were carried out to assess the survival of recombinant plasmid DNA during large-scale production of recombinant human erythropoietin (rhuEPO) in a fermentation pilot plant. The analyses revealed DNA-degrading activities in the fermentation broth and in the waste-water, leading to rapid destruction of plasmid DNA added to medium or waste-water. The capability of the plasmid-DNA-spiked samples to transform competent bacteria was drastically reduced. The DNA-degrading activity in the waste-waters could be blocked by addition of EDTA or by boiling, indicating the presence of DNA-degrading enzymes (DNases). No plasmid-specific DNA sequences were detected in waste-water samples by in-vitro amplification with Taq-polymerase. Genomic DNA preparations of cell debris collected from waste-water samples only contained degraded plasmid DNA. Furthermore, it was shown that intact plasmid DNA could be degraded to fragments of less than 1000 bp by incubation at 121 degrees C for 20 min, leading to a decrease in the plasmid-specific transforming capacity by a factor of 10(3) per minute. Thus, DNA from the rhuEPO production pilot plant was efficiently inactivated at three different levels: (i) in the fermentation medium (DNase), (ii) in the waste-water container (DNase), and (iii) by heat inactivation for 20 min at 120 degrees C. These results indicate that the probability of delivery of recombinant DNA into the environment is extremely low in such biotechnological production processes.