Viele Eu- und Archaebakterien besitzen eine kristalline Protein- oder Glykoproteinschicht (S-Schicht) als äußerste Zellgrenzfläche. Über | eine Reihe einfacher Verfahrensschritte ist es möglich, S-Schichten | zu isolieren und an geeignete poröse Träger zu binden. Auf diese Weise kann ein völlig neuartiger Typ von Ultrafiltrationsmembranen hergestellt werden, dessen aktive Trennschicht sich durch absolute Isoporosität auszeichnet und über ein breites Spektrum proteinchemischer Reaktionen spezifisch modifiziert werden kann.

Ultrafiltrationen von Interferon-ß ergeben im Unterschied zu solchen von Interferon-a unreproduzierbare und teils sehr hohe Aktivitätsverluste. Die Versuchsergebnisse zeigen, daß diese Verluste zum Teil auf einer Adsorption des Proteins an Membranen beruhen. Diafiltrationen mit einer gleichzeitigen Anderung des pH-Wertes führen zu weiteren hohen, proteinspezifischen Verlusten‘ Solche Verluste können jedoch wegen der hohen Produktionskosten für Interferon- 8 nicht akzeptiert werden und daher kommen Membranprozesse bei der Herstellung pharmazeutischer Präparate von Interferon-ß aus tierischen Zellkulturen nur bedingt in Frage. Da Interferon-8 von einigen Kunststoffen stark, von anderen aber wiederum nicht absorbiert wird, sind neue Membranpolymere von Interesse.

Polyethylenglykole wurden mit Cellulose-Acetat-Membranen ultrafiltriert, um den Einfluß hydrodynamischer Verhältnisse auf das Retentionsverhalten zu messen. Dabei wurden Lösungen ohne bzw. mit Rinder-Serumalbumin (BSA) verwendet. Die Reduzierung der Grenzschichtdicke an der Membranoberfläche durch Erhöhung der Rührergeschwindigkeit bewirkt eine Steigerung des Fluxes und des Retentionskoeffizienten. Bei zunehmendem Differentialdruck wird die Retention ebenfalls größer. Bei Diafiltrationen von proteinhaltigen Lösungen wurde ein bisher nicht bekanntes Phänomen beobachtet: Nach einer Zunahme der Retention von 30 bis auf 60% (vergl. reine PEG-Lösung) geht das Rückhaltungsvermögen der Membran mit steigender BSAKonzentration zurück. Dieses Verhalten wird in dem Beitrag diskutiert.

Most proteins present on the surface of eukaryotic cells and proteins secreted by them are glycoproteins. Glycoproteins, generally speaking, may contain asparagine- -linked (N-linked) or serine/threonine linked (0-linked) oligosaccharides. A given protein may contain either one or both types of oligosaccharides. Notwithstanding their common occurrence -glycoproteins are found in bacteria, yeasts and all higher eukaryotes- there is no consensus on the function of protein bound glycans. They are thought to be involved in protection against proteolysis and contribute to or modify the folding of the polypeptide backbone. Protein bound glycans are also thought to constitute signals for intracellular traffic of glycoproteins, and may participate in determining the specificity of interactions between receptors and ligands and the interactions between cells. In present-day biotechnology, the importance of protein bound carbohydrate is likely to come under close scrutiny. Products produced by genetic engineering are often derived from cells or organisms that do not normally produce these substances (e.g. interferon production in bacteria) yet are intended for use in clinical applications. It is therefore of paramount importance to determine what, if any, role is played by protein bound carbohydrate. Parameters such as half-life in vivo, interaction with receptors, immunogenicity and the like may all be affected. Surprisingly enough, the armamentarium available for modifying carbohydrates in a predictable and directed manner is scarce. Through detailed knowledge of glycan biosynthesis, opportunities may be created to fill this gap. This article deals with inhibitors of N-linked oligosaccharide processing. These substances, most of which were described only recently, show promise of allowing the manipulation of N-linked glycan structures in a rather precise manner, and hence might be useful in investigations on the function of carbohydrates. In order to describe the effects of these inhibitors in their proper context, a4 brief overview of N-linked glycan synthesis will be given. Subsequently, the literature on the enzymes involved in N-linked glycan processing will be reviewed in some detail. Since the effects of the inhibitors to be described need not be the same for each of the enzymes involved even if these enzymes have very similar substrate specificities, their description as possibly distinct targets for inhibition is called for. The inhibitors of oligosaccharide trimming themselves will then be described in detail. Finally, the concluding remarks will deal with some possible applications of N-linked glycan processing inhibitors.

Under appropriate conditions synthetic oligonucleotide hybridization probes display essentially absolute hybridization specificity. All of the non-Watson-Crick base pairs, including G-T, have a destabilizing effect. Thus, it is possible to choose stringent conditions of hybridization such that, while a perfectly matched duplex between an oligonucleotide and complementary DNA will form, duplexes mismatched at one or more positions will not. The hybridization of oligonucleotides to regions of DNA containing the single base changes responsible for many human genetic disease, thus, provides a means of detecting these diseases. In an attempt to develop a non-radioactive method for the detection of human genetic diseases, we have prepared biotinylated oligonucleotides by an enzymatic method. An oligonucleotide probe (23-mer) containing a single biotinylated deoxyuridine residue at the 3’-terminus was prepared by a primer extention reaction using E. coli DNA polymerase I (Klenow fragment). The biotinylated probe could be hybridized to a complementary sequence and visualized by Avidin D and biotinylated alkaline phosphatase. A general approach to the enzymatic synthesis of oligonucleotides containing a single biotinylated deoxyuridine at the 3’ end is described.

Tetranucleotides containing the carcinogen-DNA adducts 06-methylguanine and N-(guan-8-yl )-4-aminobiphenyl were synthesized and the structures of these adducted oligomers were characterized. Both tetramers were efficiently incorporated into E. coli bacteriophage M13 DNA containing a four-base gap in the unique recognition site for the restriction endonuclease PstI. Each lesion inhibited cleavage of the respective adducted genome by PstI. After introduction into E. coli, 06-methylguanine induced mutations at the PstI site. In cells with normal 06-methylguanine DNA repair capability, the frequency of mutant phage was 0.4%. When cells were depleted in their ability to repair this lesion, the mutation frequency increased. The highest mutation frequency attained was 20%. In both repair competent and repair depleted cells, the mutants induced by 06-methylguanine were exclusively G to A transitions.

The preparation, starting from glucose, of 3'-[ (2-chlorophenoxy)phosphinylmethyl]- N2-isobutyryl-5'-0-(4-methoxytrityl)-2' ,3'-dideoxyguanosine ts described. Applying established phosphotriester methods, 21 was used to synthesize a phosphonate analogue of the dinucleotide monophosphate d(G-C).

The construction of mutations in the active site of the tyrosyl tRNA synthetase from Bacillus stearothermophilus has allowed us to deduce the relative impörtance of the substrate contacts to transition state binding. The feature dominating the energetics is the exchange reaction with water molecules: thus by deleting a poor H-bonding contact to the substrate we could increase the affinity of the enzyme for substrate. Furthermore by straining the polypeptide backbone by introducing a proline residue, we could improve the interaction of a histidine residue with the substrate. Thus enzymes affinities can be bettered by protein engineering in vitro.

A set of four self-complementary dodecanucleoside undecaphosphates, d[CGNGAATTC(O6Me)GCG] (1), where N = A, C, G, or T, has been synthesized by a phosphoramidite procedure. Each sequence forms a stable duplex, with a Im between 19 and 26°C lower than the Tm Of the "parent" molecule d(CGCGAATTCGCG). The lowest melting sequence is the N=T molecule; the overall order is N = C>A>G> T. Enus 0°-met hylation of guanine creates a region of localized instability in DNA regardless of the base opposite the lesion.

The revolutionary development of molecular biology during the past years has led to a strong interest in synthetic peptides. In this talk several important applications of synthetic peptides are discus sed. Peptides can be synthesized chemically in solution (1) or on solid supports (2,3). Up to a length of approximately 50 residues synthetic peptides have been obtained in (nearly) homogeneous form (4-6).