Recombinant proteins for therapy.

Recombinant therapeutic proteins have become increasingly important over the past ten years. Numerous products derived from 20 different proteins are already on the market. In this review Peter Buckel discusses the issues surrounding the use of recombinant proteins as therapeutic agents. The first generation proteins for therapy all occur naturally in humans. Protein engineering has brought forth a second generation of products with application-specific properties obtained by fusion, mutation or deletion. The third generation of therapeutic proteins is produced by patients themselves after transfer of the relevant genes. The first successful applications of this gene therapy represent a new milestone in medicine.

Expression of Pseudomonas fluorescens D-galactose dehydrogenase in E. coli.

To investigate the heterologous expression of Pseudomonas genes in Escherichia coli we have cloned P. fluorescens DNA in an E. coli [cosmid] system. A colony bank representing the whole P. fluorescens chromosome was screened immunologically using a modification of the method described by Broome and Gilbert (1978). Radioactive labelling of the antibodies was replaced by conjugation with horseradish peroxidase. Among 523 E. coli colonies one was D-galactose dehydrogenase-positive. The expression of this enzyme in primary clones was lower than in the uninduced Pseudomonas. Subcloning of the D-galactose dehydrogenase gene, in vitro mutagenesis of the DNA, and coupling to a strong E. coli promoter yielded an E. coli strain that produces 90 times more of the enzyme than the induced P. fluorescens.

Establishment of stable mouse myeloma cells constitutively secreting human tissue-type plasminogen activator.

Expression plasmids for human tissue-type plasminogen activator (t-pA) were introduced into mouse myeloma cells and stable cell lines constitutively secreting t-pA established by selection with mycophenolic acid. Expression of t-pA is driven either by the simian virus 40 early promoter or by immunoglobulin regulatory elements of either light or heavy chains of the mouse. The availability of myeloma cells secreting a heterologous protein is of importance for biotechnological applications, because large-scale fermentation of myeloma cells is well established.

High-level expression of recombinant genes in Escherichia coli is dependent on the availability of the dnaY gene product.

We have observed that proteins, such as human tissue-type plasminogen activator, pro-urokinase or gp41 of human immunodeficiency virus, which have a high content of rare codons in their respective genes, are not readily expressed in Escherichia coli. Furthermore induction of these heterologous genes leads to growth inhibition and plasmid instability. Supplementation with tRNA(AGA/AGG(Arg)) by cotransfection with the dnaY gene, which supplies this minor tRNA, resulted in high-level production with greatly improved cell viability and plasmid stability.

Expression of antibody cDNA in murine myeloma cells: possible involvement of additional regulatory elements in transcription of immunoglobulin genes.

Expression vectors for cDNA of the kappa and gamma 1 chains of a monoclonal antibody directed against creatine kinase were introduced into murine myeloma cells. Kappa and gamma 1 cDNA were either under the control of the SV40 early promoter or of the cognate promoters and enhancers of the light- and heavy-chain genes. Secretion of immuno-reactive kappa and gamma 1 chains into the culture medium was demonstrated with the SV40 promoter as well as with the cognate promoters. Expression of gamma 1 cDNA with the SV40 early promoter was about twice as high as with the heavy-chain promoter and enhancer. Expression of kappa cDNA under the control of the SV40 early promoter was about 17 times higher than with the light-chain promoter and enhancer. These expression levels were compared to those of a genomic immunoglobulin (Ig) kappa determinant, including introns. Such an entire kappa gene led to expression of the light chain at levels double those with the kappa cDNA construction using the SV40 promoter and about 35 times as high when using kappa cDNA and the cognate promoter and enhancer. This result might indicate that, besides the cognate promoter and enhancer elements, other intragenic elements are involved in the regulation of Ig expression. However, the SV40 early promoter seems to be able to compensate for the absence of these postulated regulatory elements probably located in the introns.

Amplified expression constructs for human tissue-type plasminogen activator in Chinese hamster ovary cells: instability in the absence of selective pressure.

By linking an expression cassette for human tissue-type plasminogen activator (t-PA) to an amplifiable marker gene, its introduction into Chinese hamster ovary dhfr- cells and subsequent amplification with methotrexate, we have generated cell lines that overproduce the heterologous protein and contain 300-1100 copies of the expression constructs integrated into the hamster genome. We present a detailed investigation of the fate of amplified sequences in the presence and absence of selective pressure by parallel examination of three producer cell lines with respect to relevant parameters. These include the determination of t-PA production upon continuous propagation in culture, the genomic organization of the integrated expression constructs by Southern blotting, and the localization of homogeneously staining regions by in-situ hybridization with biotinylated probes and visualization by interference reflection microscopy. We conclude that in the three cell lines examined, the decrease in production of t-PA in the absence of methotrexate selection is accompanied by decreases in the number of integrated expression constructs and the size of the amplified regions, whereas all these parameters are stable when selective pressure is maintained. The instability is probably due to the head-to-tail mode of integration of the expression constructs in the hamster genome, which increases the frequency of homologous recombination between the integrated plasmids in recombination-proficient cells in the absence of selective pressure.

A new expression system for mammalian cells based on putative replicator sequences of the mouse and a truncated thymidine kinase gene.

We have constructed a new expression vector for mammalian cells. The vector contains a truncated tk gene for amplification under selective conditions, a sequence putatively supporting the replication of plasmid DNA in eukaryotic cells (murine autonomously replicating sequence) and an expression cassette for the cDNA to be studied. As a model cDNA we have used that of human tissue-type plasminogen activator (t-PA). Analysis of Hirt supernatants and chromosomal DNA from L cells, prepared six weeks after isolation of the clones indicated a 50- to 500-fold amplification of the expression construct in the cells. Concomitantly, the expression of t-PA was dramatically increased. Our data are consistent with episomal persistence of the expression construct, with a head-to-tail mode of integration into the mouse genome and with coexistence of both episomal plasmids and head-to-tail integrates. In tk-deficient cell lines other then L-cells, such as mouse mastocytoma or rat hepatoma cells, a strong selection against the persistence of the expression construct was noted. After long-term propagation of the L-cells under selective conditions the expression of the indicator gene continually decreases, but finally a constant plateau level of expression is established. Expression could be restored to the original level by blocking more efficiently the de novo synthesis of nucleosides.

A protease-hypersensitive deletion derivative of human tissue-type plasminogen activator.

We have constructed amplified Chinese hamster ovary cell lines constitutively synthesizing human tissue-type plasminogen activator (t-PA) or a derivative in which the domains homologous to epidermal growth factor and kringle 1 have been removed [delta(G + K1)]. The properties of the secreted proteins were investigated when synthesized in the presence or absence of the serine protease inhibitor aprotinin in the medium. t-PA in the culture supernatants was either single-chain or two-chain protein. The protease activity of both forms was stimulated by fibrin. The biochemical properties of delta(G + K1) were significantly different when harvested from cells grown under different culturing conditions. Protease activity of delta(G + K1) was stimulated ten- to 20-fold by fibrin when harvested from medium with aprotinin, but was stimulated only two- to three-fold when aprotinin was absent from the serum. Characterization of the secreted proteins revealed that the heavy-chain equivalent of delta(G + K1) is degraded when serine protease inhibitor is absent in the culture medium. These results indicate that the functional and biochemical properties of restructured versions of t-PA may depend on the presence of protease(s) in the culture supernatants.

Establishment of a temperature-inducible cell line for human plasminogen activator (tissue-type) by transfection of monkey cells with expression constructs.

An expression construct for human tissue-type plasminogen activator (t-pA) cDNA [containing a simian virus 40 (SV40) origin of replication] was introduced into CV1, COS-7 and COSts2 cells; in the latter cell line the amount of functionally active large T antigen of SV40 is regulated by the temperature. In a transient system, the expression in COSts2 cells at the permissive temperature for large T antigen was improved sixfold compared to COS-7 cells. By cotransfection with a plasmid conferring resistance to G418 into COSts2 cells, a cell line (COSts2Glob t-pA) could be isolated with barely detectable expression of t-pA at the semi-permissive and non-permissive temperature and inducible secretion of t-pA at the permissive temperature. The kinetics of induction, inducibility after continued propagation at the semi-permissive temperature and the influence of the temperature during previous propagation on inducibility were investigated. The biological activity of the secreted material was demonstrated by a functional assay. Inducibility of t-pA by temperature was accompanied by a dramatic increase of the copy number of episomal plasmids (up to 2000 copies per cell).

Functional topology of human tissue-type plasminogen activator: characterization of two deletion derivatives and of a duplication derivative.

Three types of permanent Chinese hamster ovary (CHO) cell lines with different amplified expression constructs that abundantly secrete derivatives of human tissue-type plasminogen activator (t-PA) were established. The first one expresses a deletion derivative in which the kringle 2 domain (K2) has been removed (FGK1L). In the second derivative, the growth-factor-homologous domain (G) has also been deleted (FK1L); a third line expresses a duplication derivative of K2 (FK2K2L) lacking the (G) and kringle 1 (K1). All deletion derivatives were constructed according to the exon-intron organization of the gene. We have analyzed the secreted proteins and the fibrinogen-stimulated plasminogenolytic activity as a function of different culturing conditions (fetal calf serum, aprotinin) of the cells. The specific activities of the two deletion derivatives (FGK1L and FK1L) were only 10-20% of the specific activity of t-PA. Surprisingly, the specific activity of the K2-duplication derivative, FK2K2L, was three times higher than that of t-PA. These data were correlated with the morphological properties of CHO cells constitutively secreting the described derivatives under different culturing conditions. CHO cells secreting the deletion derivatives (FGK1L and FK1L) remained attached to the surface of the petri dishes. Cell lines secreting the duplication derivative FK2K2L detached from the surface even in the presence of the protease inhibitor aprotinin.

Reconstitution of functionally active antibody directed against creatine kinase from separately expressed heavy and light chains in non-lymphoid cells.

We report here for the first time reconstitution and secretion of functionally active antibody in non-lymphoid cells. Expression vectors for the light and the heavy chain of a monoclonal antibody directed against creatine kinase (EC 2.7.3.2) were introduced into COS and CHO Chinese hamster ovary dhfr- cells. Introduction of the expression vectors separately gave rise to immuno-reactive material in the culture supernatants, but only cotransfection of the expression plasmids resulted in secretion of protein with immuno-reactivity against antibodies directed against mouse heavy and light chains as well as specific antigen-binding affinity, as determined by enzyme-linked immunosorbent assay. Secreted kappa and gamma chains from reconstituted antibody were characterized by immunoadsorption and sodium dodecyl sulfate-polyacrylamide gel electrophoresis. In COS cells, reconstituted antibody was transiently secreted; cotransfection of kappa and gamma chain expression plasmids with a dihydrofolate reductase (DHFR)-expression plasmid into CHO dhfr- cells gave rise to stable transformants secreting functionally active antibody.

Cloning and nucleotide sequence of heavy- and light-chain cDNAs from a creatine-kinase-specific monoclonal antibody.

Determination of creatine kinase isoenzymes by inhibition assay is a useful tool for the diagnosis and monitoring of myocardial infarction. We have established several mouse hybridoma lines secreting monoclonal antibodies with creatine kinase M-subunit inhibitory capacity. One of the monoclonal antibodies (MAK33) inhibits creatine kinase-MM by 80% without influencing the activity of creatine kinase-MB. A combination of two monoclonal antibodies increased the inhibition of creatine kinase MM up to 99.4%. Poly(A) + RNA of hybridoma cells secreting MAK33 was isolated and used for cloning cDNA of both heavy and light chains of this antibody. Full-length cDNA clones were obtained by hybridization with gamma 1 and kappa constant region cDNA probes. The complete nucleotide sequences from the variable regions including signal peptide and part of the 5'-untranslated regions have been determined.

Engineering enzymes for clinical diagnosis.

The enzyme creatine amidinohydrolase (creatinase, EC 3.5.3.3.) is a major limiting factor in the enzymatic creatinine determination because of its comparatively poor catalytic activity and stability in the native form. The gene from Pseudomonas putida coding for creatinase was cloned and used for overexpression of the protein in E coli and Pseudomonas. In addition, it was possible by means of 'random' mutagenesis in vivo and subsequent screening using an activity plate assay to isolate creatinase derivatives that are more stable towards detergents and elevated temperature under test kit conditions. This example shows that enzymes can be optimized for use in given assay conditions by mutagenesis of cloned DNA and suitable screening methods.

Identity of a gene responsible for suppression of aminoacyl-tRNA synthetase mutations with rpsT, the structural gene for ribosomal protein S20.

Sepcialized transducing lines of phage lambda carrying segments between thr and car from the E. coli chromosome have been isolated. With help of these phages it has been shown that the gene sups20 (Böck et al., 1974) corresponds to rpsT, the structural gene for ribosomal protein S20.

Toward a new natural medicine.

Treatments summarized under the term "natural medicine," i.e., those offered as an alternative or in addition to conventional medicine, have enjoyed a surge in popularity in recent years. The "natural" descriptor employed in connection with these healing methods is frequently misunderstood, leading to underestimation of the risks arising from incorrect use. However, the essential principle underlying traditional natural medicine, mobilization of the body's own forces against disease, is increasingly being employed in a new, rational form of medicine: molecular medicine. A range of natural endogenous substances for medical use are already available. Human proteins such as erythropoietin can now be produced as medicines in highly pure form with the aid of genetic engineering techniques. Our increasing understanding of the function of our genes and the resulting descriptions of molecular mechanisms underlying disease are also helping us to utilize the body's own construction set. New techniques such as gene therapy will in future enable us to reproduce the natural conditions in the healthy body with increasing specificity in our attempts to cure illnesses. One such application will be the activation of the immune system to combat cancer. The complete decoding of the human genome will not only allow illnesses to be described, and possibly prevented, at an earlier stage. Illnesses will also be able to described more precisely and individually at the molecular level, opening up the possibility of targeted, patient-specific cures.

The D-galactose dehydrogenase gene from Pseudomonas fluorescens: characterization of mutations leading to increased expression in Escherichia coli.

The gene encoding D-galactose dehydrogenase (gld; E.C. 1.1.1.48) from Pseudomonas fluorescens is poorly expressed when cloned into Escherichia coli. Mutagenesis of the wild-type construct leads to a strong expression of gld in the heterologous host. To investigate the mutational events directing the increase in expression we constructed a gld-lacZ translational fusion which facilitated the isolation of mutants by colony screening. From several independent mutants three point mutations could be identified. They were distinguished by the sequence position of their respective single base-pair substitutions in the 5'-untranslated region of the gld gene and the degree of enhancement of enzyme activity of the gene product. The influence of these mutations on gld gene expression was analysed by S1 protection analysis which revealed that their effect was at the level of transcription.

Control of formation of active soluble or inactive insoluble baker's yeast alpha-glucosidase PI in Escherichia coli by induction and growth conditions.

Using standard growth conditions (LB medium, 37 degrees C, induction with 5 mM IPTG) yeast alpha-glucosidase PI expressed under the control of the regulated tac-hybrid promoter results in the synthesis of insoluble aggregated alpha-glucosidase granules in Escherichia coli. Under these conditions active soluble alpha-glucosidase amounts to less than 1% of the heterologously produced protein. However, the amount of soluble active alpha-glucosidase was dramatically increased when the strong tac-hybrid promoter was to a limited extent induced. This was achieved at concentrations of 0.01 mM IPTG or of 1% lactose or lower in a lactose-permease deficient host strain containing the lacIq repressor gene on an R-plasmid. The formation of active soluble alpha-glucosidase was almost 100% when E. coli cells induced in this manner were cultivated under conditions that reduced growth rate, i.e. at decreased temperature, extreme pH values or in minimal and complete media supplemented with different carbon sources.

Suppression of temperature-sensitive aminoacyl-tRNA synthetase mutations by ribosomal mutations: a possible mechanism.

The biochemical basis of suppression of a temperature-sensitive alanyl-tRNA synthetase (alaS) mutation by mutational alterations of the ribosome has been investigated. Measurement of the polyU-dependent polyphenylalanine synthesis showed that ribosomes from the suppressor strains are less active than ribosomes from the unsuppressed aminoacyl-tRNA synthetase mutant. In this system no increased translational ambiguity could be detected for the suppressor ribosomes. This fact and also the findings that the ram-1 mutation is not able to suppress the aminoacyl-tRNA synthetase mutation and that presence of the suppressor allele is not accompanied by a measureably improved alanyl-tRNA synthetase activity argue against the possibility that suppression might be due to increased translational misreading rates of the alanyl-tRNA synthetase mRNA. It has been further found that partial suppression of temperature sensitive growth of the alaS mutation can be achieved by independent ribosomal mutations leading to reduced growth rates because of a mutation to antibiotic resistance. Addition of low concentrations of a variety of antibiotics acting at the ribosomal level can also partially revert the temperature-sensitive phenotype of the alaS mutant. Although the possibility cannot be excluded that suppression is due to the stabilisation or activation of the mutant enzyme by some indirect effect of the suppressor ribosomal mutations, the following working hypothesis is favoured at the moment: It is assumed that limitation of the aminoacyl-tRNA synthetase activity in a certain range of the restrictive temperature causes growth inhibition by the premature termination of polypeptide synthesis at the ribosome or by the unbalanced synthesis of the individual cellular proteins under this condition. The mechanism of suppression by ribosomal mutations is proposed to consist of the release of this growth inhibition by the reduction of the rate of polypeptide synthesis, which would keep amino acid incorporation from exceeding the slow charging of tRNA and thus exhausting the pool of charged tRNA. In the suppressor strains, therefore, growth at the semi-restrictive temperature is no longer limited by the aminoacylation of tRNA but by the translational process at the mutated ribosome. This influence of the ribosomal mutation on the speed of translation could be directly or indirectly coupled with an effect on translational fidelity resulting in the prevention of the binding of uncharged or non-cognate charged tRNA or in the tighter binding of peptidyl-tRNA when cognate aminoacyl-tRNA is limiting.