Proceedings of the Production of Monoclonal Antibodies Workshop

August 29, 1999 | Bologna, Italy

This document has been adapted with permission from a publication created by the Alternatives Research & Development Foundation (ARDF)

Comments on Scientific Justifications Offered to Support the Use of the Ascites Method to Produce Monoclonal Antibodies

Uwe Marx, PhD
Institute of Clinical Immunology & Transfusion Medicine
Department of Medical Biotechnology, University of Leipzig, Germany

Introduction

As the availability, relative cost and ease of use for in vitro methods of producing monoclonal antibodies continues to improve (see Jackson, et al., this volume), the continued routine use of ascites methods becomes increasingly difficult to defend. Several recent publications (Halder, et al.¹, NRC²) address the scientific justifications commonly offered in support of using ascites to produce monoclonal antibodies. The aim of this contribution is to provide further consideration of such examples. This discussion may be of particular interest to Institutional Animal Care and Use Committees (IACUCs), who must increasingly decide if requests by investigators to use in vivo approaches to MAb production are valid, or if the use of in vitro methods should be required.

The justifications that are offered in support of using ascites are usually characterized by a few general categories:

Some Hybridomas Do Not Adapt Well to In Vitro Conditions
It is often claimed that there is a finite and significant failure rate for in vitro MAb production of between 3% to 4%. In our experience this argument applies equally well to ascites. Until 1990, we produced a number of different MAbs in ascites fluid. Annually between 2 and 4 ascites production runs failed, providing ascites fluids without or with a very low or high but non-specific antibody titre.

Underwood and Bean⁹ provide one early indication that a finite and significant failure rate is also characteristic for ascites production of monoclonals. In scientific literature this aspect cannot be found frequently, due to the very low value of this technical information for scientists. But this observation has a very high value for contract MAb manufacturers. Thus Harlan Bioproducts for Science, Inc., a provider of professional in vitro and in vivo production services, states in its ascites brochure that the company "cannot guarantee that all cell lines received will secrete the desired antibody or produce the antibody in the desired concentration." If both in vivo and in vitro methods have the same problem with a significant failure rate, this characteristic cannot be used to endorse the in vivo method.

Since all hybridomas are initially created in vitro, proper attention to clone selection procedures and characteristics should produce hybridoma cells that will adapt well to additional in vitro conditions. The selection process for a hybridoma cell line is, however, relatively short (several weeks). As known from human hybridoma technology, some cell lines need more than 3 months to establish a stable genotype.
Downstream Purification of In Vitro MAbs Can Lead to Protein Denaturation and Decreased Antibody Activity
Both in vivo and in vitro produced monoclonal antibodies have identical problems with purification methods. For those few cases where ascites can be used instead of purified antibodies, the crude supernatant of in vitro cultures would also be acceptable. Underwood and Bean⁹ described the influence and immunological behavior of 38 different murine monoclonal antibodies. The principal problems they identified were associated with ascites fluids. Changes in cross-reactivity patterns were identified in three of the MAbs after ascites passage. The authors noted that "compared with the changes observed after physical treatments, the changes resulting from ascites passage were very dramatic."
Ascites Mouse Protein Contamination of MAbs Is Not Important
To prove the biological activity of any antibody, we perform all in vivo tests in mice, rats or other animals with purified MAbs. This avoids problems caused by contamination with foreign proteins, whether from fetal calf serum (FCS) or mouse ascitic fluids. Prior to 1990, when we switched to in vitro methods, all ascites-derived antibodies were purified to prevent the side effects of cytokines that were usually present. Production of control ascitic fluid, derived from propagation of a hybridoma, producing a non-specific MAb of the same Ig-class might have different peptide/protein/cytokine compositions that would bias results obtained from its use. As a general principle, potential contamination of MAbs should be an important consideration in all research projects.
Rat Hybridomas and Unstable Heterohybridomas Produce Better in Immunocompromised Mice
Our experiences with rat hybridomas include in vitro production of MAbs from nine different pre-existing cell lines. We preformed 12 bioreactor runs in the membrane-based culture system, Technomouse, and succeeded in all cases without any technical problems or difficulties.

Heterohybridomas are less stable cell lines. Those that produce an appropriate titre in immunocompromised mice have a good chance to have the same titre in membrane-based culture systems. We successfully produced human monoclonal antibodies from at least 15 different heterohybridomas in hollow-fiber bioreactors of both small and medium size.
Serum-Free Cultures Produce Low Yields of MAbs
Most hybridomas can be successfully adapted to serum-free conditions. In those examples where this is difficult, the appropriate alternative is not the use of ascites, but rather the utilization of low-serum media, with or without additional supplements. New, high-yield in vitro methods are now available, which are often designed to work under low-serum or serum-free conditions. High yields of MAbs are possible using such techniques.

Both IgM and IgA are often cited as special cases of antibodies that are difficult to produce in serum-free cultures or to purify if serum-supplemented media are used. In fact, they are equally difficult to purify, whether derived from serum containing supernatant or ascites fluids. There are now several publications available that discuss in vitro solutions to these problems^6,7,8,9.
Ascites Use Avoids Problems With In Vitro Differences in Glycosylation
Differences in glycosylation can result from both in vivo and in vitro production of monoclonal antibodies. They are not a unique characteristic of either method and rarely occur at the active site of the antibody⁵. In vitro methods, unlike ascites, have the advantage of allowing for deliberate modifications of glycosylation patterns by altering various parameters of the culture conditions. Such manipulations can produce MAbs with specific, useful and functional properties.

For example, human monoclonal anti-Rhesus D antibody (BRAD-3) was produced using serum-free media in both low cell density cultures and high cell density hollow-fiber bioreactors. Differences in culture conditions lead to a four-fold enhancement of the disialated molecules⁴.

Halder et al.¹ and Marx et al.² concluded from examination of all of the available evidence in the biomedical literature that "there are no reasonable arguments based on antibody glycosylation which support the use of in vivo methods" to produce monoclonal antibodies.
Purification of In Vitro Cultures
Using ascites to purify contaminated hybridoma cultures is usually done to remove mycoplasma. There are several treatment protocols and in vitro commercial kits (e.g., Boehringer Mannheim: BM Cycline I and II) that can replace such uses of mice. In general, cultures infected with bacteria, fungi or yeast should be discarded. Under exceptional circumstances passage through mice is allowed in Germany to rescue contaminated hybridoma cultures.

Conclusion

The majority of the problems encountered with the in vitro production of monoclonal antibodies and cited in defense of the continued use of ascites could be solved by providing more comprehensive training in in vitro methods for researchers or establishing more specialized cell culture facilities for the production of MAbs.

Those exceptional circumstances that justify the use of ascites should be limited to^1,3:

emergency therapeutic applications;
existing regulatory approvals for diagnostic or therapeutic products (until the approval expires);
circumstances in which verifiable in vitro attempts fail to produce MAbs or protect hybridoma cultures.

References

Halder M., Embleton MJ, Fischer R, de Geus B, Hendriksen C, de Leeuw WA, Marx U and Balls M. 1998. Comments in Appendix C of the National Institutes of Health Response to the Petition of the American Anti-Vivisection Society to Prohibit the Use of Animals in the Production of Monoclonal Antibodies. ATLA 26: 551-554.
National Research Council. 1999. Monoclonal Antibody Production. National Academy Press. Washington, DC.
Marx U, Embleton MJ, Fischer R, Gruber FP, Hansson U, Heier J, de Leeuw WA, Logtenberg T, Merz W, Portelle D, Romette J-I and Straughan DW. 1997. Monoclonal Antibody Production: Report and Recommendations of ECVAM Workshop 23. ATLA 25: 121-137.
Kumpel BM, Rademacher TW, Rook GAW, Williams PJ and Wilson IBH. 1994. Galactosylation of human IgG monoclonal anti-D produced by EBV-transformed B-lymphoblastoid cell lines is dependent on culture method and affects Fc receptor-mediated functional activity. Hum. Antibod. Hybridomas 5: 143-151.
Leiberger H, Hansen A, Schoenherr G, Seifert M, Wuster D, Stigler R and Marx U. 1995. Glycosylation analysis of a polyreactive human monoclonal IgG antibody derived from a human heterohybridoma. Molecular Immunology 32: 595-602.
Stoll T, Chappaz A, von Stockar U and Marison IW. 1997. Effects of culture conditions on the pro-duction and quality of monoclonal IgA. Enzyme and Microbial Technology 21: 203-211.
Roggenbuck D, Marx U, Kiessig ST, Schoenherr G, Jahn S and Portsmann T. 1994. Purification and immunochemical characterization of a natural human polyreactive monoclonal IgM antibody. J. Immunological Methods 167: 207-218.
Lullau E, Heyse S, Vogel H, Marison IW, von Stockar U, Kraehenbuhl J-P and Corthesy B. 1996. Antigen binding properties of purified IgA and reconstituted secretory IgA antibodies. Journal of Biological Chemistry 271: 16,300-316.
Underwood PA and Bean PA. 1985. The influence of methods of production, purification and storage of monoclonal antibodies upon their observed specificities. J. Immunological Methods 80: 189-197.