Hybridomas are immortal cell lines that secrete an antibody of only one defined specificity into the culture medium. These antibodies are called “monoclonal antibodies” because the cell lines descend from only one ancestor cell by clonal expansion. The majority of newly approved therapeutics consists of humanised monoclonal antibodies that originated from hybridoma cells. Most applications of these therapeutic antibodies are in the field of cancer therapy. In addition, monoclonal antibodies are indispensable tools for research and diagnostics . The long term objective behind the development of our hybridoma library is to speed up and simplify the search for novel monoclonal antibodies . Alternatively, this screening for novel monoclonal antibodies is done in bacterial systems where antibodies are presented on the surface of bacteriophages, a procedure that is called “phage display” [10, 12]. However, phage display has two decisive drawbacks when compared to our hybridoma library: (1) a bacteriophage displays only a few antibodies on its surface, i.e. the signal strength is very low when screening such libraries “blind” for antibody binders  (while we “see” every binding event in the FACS sorter when the many antibodies on the surface of the hybridoma cell bind to their antigen), and (2) usually bacterial systems yield only a few µg of antibodies for functional tests (while our hybridoma cells easily produce 100mg of a monoclonal antibody). Due to these technical advantages, we believe that our hybridoma library could be used to reliably find subtle differences when comparing complex mixtures of antigens (e.g. differentiated cell vs. stem cell) once it is assembled.
To achieve these goals, we started a close cooperation with the DKFZ-department Translational Immunology. We first chose an especially fast growing hybridoma cell line that produced a monoclonal antibody that specifically binds to the protein L1CAM . Starting with this cell line, we then selected with a FACS sorter again and again those cell variants that displayed an especially large number of the anti-L1CAM antibody on their surface (antibody display is due to a naturally occurring splice variant that codes for an antibody with a membrane anchor). These experiments resulted into a stable cell line that displays an estimated 50.000 to 100.000 anti-L1CAM antibodies through the natural splice variant in their membrane. In a sense, we thus created an antibody that carries its own gene in a rucksack. Thus, selecting the antibody protein with a binder at the same time selects the antibody’s gene.
Next, we exchanged the variable vH- and vL-genes of the cell line’s anti-L1CAM antibody against other vH- and vL-chains. This was done by homologous recombination, a very rare event that happens approx. in 1 out of 108 cells. Although technically challenging, we were able to pick this one cell from a background of one hundred million unchanged cells by specifically staining the newly introduced variable chain and sorting the antibody displaying cells with a modern high-speed FACS sorter. In doing so, we “smuggled in” short DNA stretches that now flanked the vH- and vL-genes of our re-engineered cell line that was grown to a clone. These short DNA stretches are specifically recognized by a “recombinase” that now would exchange vH- and vL-genes in a much higher frequency against the diversity of different antibody genes (Fig. 7).
This assembly of the hybridoma library is currently done. Late in 2011 we expect to have a library of ~105 different hybridoma cells, each cell displaying 50.000 to 100.000 of “its” human IgG1 antibody on their surface . These experiments were headed by Dr. Sandra Lüttgau. We hope that this hybridoma library will yield novel monoclonal antibodies in unprecedented speed and quality, and we hope that we then can improve the affinity of these antibodies with the help of our cooperation partners Prof. Jean-Claude Weill and Prof. Claude-Agnes Reynaud (INSERM, Paris) that explored a procedure to induce the endogenous apparatus for somatic hypermutation in our hybridoma cell line . We think, that the technical characteristics of our hybridoma library should allow for applications (more) that are not possible with other antibody libraries. Especially, we hope that our hybridoma library will give us the opportunity to compare related antigen mixtures, i.e. to find antibodies that specifically bind to the differences between two mixtures. Thereby, we might be able to find an antibody that differentially stains a stem cell antigen.
Fig. 7; “Re-engineering” a cell line to generate of a library of hybridoma cells.
 Breitling F and Dübel S. (1999) Recombinant Antibodies. 161 Seiten. John Wiley & Sons, Inc., New York N.Y. ISBN 0-471-17847-0
 Breitling F, Moldenhauer G, Lüttgau S, Kühlwein T, and Poustka A. (2001) Methods of producing protein libraries and selection of proteins from them. Family of patent applications EP1298207B1 (issued 2010), WO2003029458A2. Currently maintained by DKFZ.
 Breitling F, Dübel S, Seehaus T, Klewinghaus I, and Little M. (1991) A surface expression vector for antibody screening. Gene 104, 147-153.
 Rondot S, Koch J, Breitling F und Dübel S. (2001) A helper phage to improve single-chain antibody presentation in phage display. NatureBiotechnology 19, 75-78
 Huszar M, Moldenhauer G, Gschwend V, Ben-Arie A, Altevogt P und Fogel M. (2006) Expression profile analysis in multiple human tumors identifies L1 (CD171) as a molecular marker for differential diagnosis and targeted therapy. HUMAN PATHOLOGY 37, 1000-1008
 Reynaud CA, Delbos F, Faili A, Gueranger Q, Aoufouchi S und Weill JC. (2009) Competitive repair pathways in immunoglobulin gene hypermutation. Phil. Trans. R. Soc. B 364, 613–619.