Strategies to avoid virus transmissions by biopharmaceutic products.

The use of biopharmaceutical products offers an opportunity for the treatment of many diseases. Biotechnical manufacturing using recombinant mammalian cell lines is the most appropriate method today for the production of biopharmaceutical protein drugs for the treatment of human and animal diseases. However, mammalian cell line derived products have a potential risk for virus transmission to patients who are treated with these biopharmaceutical products. The reliability that biological products are free of any viruses requires a combination of several strategies: The use of well-characterized cell bank systems and, if feasible, the avoidance of biological raw materials for the cultivation of these mammalian cell lines and the production of biopharmaceuticals. Further on, the purification process for biopharmaceuticals has to be validated for its ability to efficiently remove and inactivate any potential virus contamination and, where applicable, also unconventional transmissible agents, such as BSE. In addition, the biopharmaceutical product itself can be tested for the presence of viruses. Like other manufacturing processes, biotechnical production processes have to be performed in compliance with current Good Manufacturing Practices (cGMP).

Adhesion of neural cells to extracellular matrix constituents. Involvement of glycosaminoglycans and cell adhesion molecules.

Single cell suspensions of early postnatal mouse cerebellum adhere to substrate-bound culture supernatants of the teratocarcinoma cell line PF-HR9 and can be inhibited to adhere by antibodies to the neural cell adhesion molecules L1 and N-CAM. Adhesion can also be inhibited by the glycosaminoglycans heparin and heparan sulfate, and less by chondroitin sulfate or hyaluronic acid. Heparinase treatment of cells, but not of HR9 substrate, reduces adhesion. Adhesion does not appear to be mediated by laminin, a constituent of HR9 extracellular matrix, since L1 and N-CAM antibodies do not interfere with cell adhesion on EHS sarcoma laminin as substrate and since antibodies to EHS sarcoma laminin partially inhibit adhesion to HR9 extracellular matrix which contains laminin. Of the other extracellular matrix constituents analysed in HR9 culture supernatants (collagen type IV, a heparan sulfate proteoglycan and fibronectin) none could be shown to promote adhesion, when coated as substrate, suggesting that yet unidentified compounds are responsible for L1- or N-CAM-mediated cell adhesion. These experiments show for the first time that extracellular matrix constituents can act as binding partners for the neural cell adhesion molecules L1 and N-CAM.

Experimental modification of postnatal cerebellar granule cell migration in vitro.

Histotypic migration of [3H]thymidine pulse-labeled granule cell neurons in cerebellar folium explants was monitored in the presence of antibodies to cell adhesion molecules and quantified by automatic image analysis. When explants were cultured in the presence of monovalent antibody fragments to cell adhesion molecules L1 and N-CAM, an inhibition of cell migration of 33.3 +/- 4.4% and 13.9 +/- 2.1%, respectively, was observed. In the presence of an equimolar mixture of monovalent antibody fragments to L1 antigen and N-CAM no additive effects in inhibition of cell migration were seen. Antibodies to the L2 carbohydrate epitope which is common to L1, N-CAM and other cell surface glycoproteins showed a similarly small effect on cell migration as antibodies to N-CAM. Monoclonal antibodies to cell surface antigen M2 and polyclonal antibodies to mouse liver membranes reacting with the surface of all cerebellar cell types did not alter the migratory behavior of granule cells. Cultivation of explants in the presence of neuraminidase, ganglioside binding toxins, as well as glycosaminoglycans and glycosaminoglycan degrading enzymes, also did not modify the extent of cell migration under the culture conditions used.

Glycosaminoglycans of rat cerebellum: II. A developmental study.

Total and individual glycosaminoglycans (GAGs) were determined in rat cerebellum in tissue explants at various postnatal ages. The major constituents of GAGs were chondroitin sulfate (CS), hyaluronic acid (HA), and heparan sulfate (HS). Dermatan sulfate (DS) and keratan sulfate (KS) could not be detected and therefore each amounts to less than 5% of all GAGs at all ages studied. HA was the prominent GAG during postnatal development and only a minor constituent at adult ages, whereas CS was the predominant GAG in adulthood. HS remained relatively constant throughout development. The incorporation of [3H]glucosamine into individual GAGs was highest for HS at postnatal day 6, whereas HA showed intermediate and CS the lowest levels of incorporation during the first postnatal week. All major GAGs showed the lowest incorporation values at adult ages.

Glycosaminoglycans of rat cerebellum: I. Quantitative analysis of the main constituents at postnatal day 6.

Isolated glycosaminoglycans (GAGs) were quantified biochemically in the cerebella of 6-day-old rats. 14C-Labeled hyaluronic acid (HA) and chondroitin-4-sulfate (C-4-S), added prior to isolation of GAGs from tissue, served as internal standards to allow correction for unknown losses during the purification procedure and exact quantification of GAGs in the intact tissue. Three main constituents--HA, chondroitin sulfate (CS), and heparan sulfate (HS)--were found at concentrations of 1.82, 1.52, and 0.76 micrograms/mg protein amounting to 44%, 37%, and 19% of the total GAG fraction, respectively. Incorporation of [3H]glucosamine precursor into GAGs was higher for HS (56% of incorporated precursor) and lower for HA (29%) and CS (15%). The specific activities of individual GAGs were 64.7 nCi/micrograms for HS, 14.2 for HA, and 8.3 for CS.

Experimental approaches to guarantee minimal risk of potential virus in purified monoclonal antibodies.

Regarding biological products, increasing awareness of potential side effects have placed great importance not only at protein purity regarding other proteins but on the removal of biologicals such as DNA and especially virus the importance of which may not be known. Monoclonal antibodies (Mab) have come to be an important class of molecules obtained from hybridoma cells, i.e., nonrecombinant cells in culture. It has been noted during the last years, that with rare exceptions hybridoma cell lines contain retrovirus like particles. The infectious nature of the EM-visible particles has been tested for, however, in most cases not been substantiated. In order to bring these valuable biological reagents, Mab's, to good use in man for imaging or therapy, the remaining concern about a potential retroviral infection has to be reduced to an acceptable minimum. We describe experimental approaches for the validation of chromatographic and ultrafiltration steps used in the production of monoclonal antibodies to remove and inactivate murine retrovirus. Present day biotechnological manufacturing processes have been devised incorporating a number of strategic preventive measures that have found wide spread acceptance. They permit to answer the question: how can a potentially harmful infection by an unknown virus be excluded. Knowledge of the efficacy of purification steps to clear infectious model virus is fundamental to devise biotechnological manufacturing processes yielding a purified antibody for use in man.

Optimization of vaccine production for animal health.

Vaccines on the basis of mammalian cell cultures are of major importance for human and animal health. Therefore efforts are undertaken for the improved production of more effective vaccines. Of course, the main purpose of all these approaches is to save lives and improve the quality of life for human beings. However, there is also some remarkable effort in the food industry and the associated animal production, especially in the case of some Flaviviridal viruses (BVD), where > 80% of all cattle herds are found to be infected. These viruses can cause tremendous economic losses of calfs and embryos (Ames, 1990). Because of these facts, there is a continuous endeavour for improving the manufacturing of therapeutics or preventing agents such as vaccines for the treatment of cattle. The competitive economic situation and the specific market demands still require effective and high yield production methods, especially in the case of one of the most widespread viral diseases in cattle like BVD (Ames, 1990). We have succeeded in establishing an improved method for the production of BVD on the basis of a continuous fermentation mode, that consist of modifications of the corresponding process and media improvements.

Process scale considerations in evaluation studies and scale-up.

This paper focuses on the scale of validation studies that are performed in order to demonstrate viral clearance in downstream processing of biopharmaceutical products. A serious concern for rDNA-derived proteins from recombinant mammalian cell cultures and monoclonal antibodies (MAbs) derived from hybridoma cultures is the potential risk from contaminating retroviral particles or adventitious viruses. Accordingly the downstream process has to be designed to reduce a potential virus load significantly: by virus removal and by virus inactivation methods. Such viral clearance is essential for drug safety- and has to be validated. This means that the design of downstream processing has to take into consideration both the capability for viral clearance and the capability for validation respectively.

Pentapeptide identified as a monoclonal antibody binding site in the serine-protease domain of t-PA.

The first defined sequential epitope of the tissue plasminogen activator (t-PA) was determined by a monoclonal antibody against a synthetic peptide segment corresponding to peptide sequence 341-354 of t-PA. This segment was selected by computer assisted epitope prediction. Balb/c mice were immunized with catalase-peptide and tripalmitoyl-S-glyceryl-cys-teinyl-seryl-peptide conjugates. A monoclonal antibody derived from this immunization was reactive with native recombinant t-PA (rt-PA) and reduced carboxymethylated recombinant t-PA (RCM rt-PA). The sequential epitope was detected by Pepscan method using overlapping octa- and nonapeptides. By fine epitope mapping with tetra-, penta-, hexa- and heptapeptides the epitope was minimized to the pentapeptide EEEQK (347-351). Replacement set analysis confirmed the importance of this amino acid sequence, especially of the amino acid E348, for antibody binding. Functional assays of rt-PA were not affected by this antibody indicating that the epitope has no influence on the enzymatic center and the binding site of the inhibitor. The analysis demonstrates that the predicted recognition site of the monoclonal antibody 17-134/11 is exposed on the surface of the native rt-PA molecule.