Purpose: Normally drug selection used for identifying genetically modified cell, but there are problems (cell growth inhibition, morphological changes). Therefore instead of negative selection, positive selection is an alternative (whereby a genetically modified cell expressed a receptor that causes cell to recognize a growth signal so that only THAT cell may grow and form colonies). This is called AMEGA (antigen-mediated genetically modified cell amplification)
Remember: They fused antibody to a joint protein of EpoR and gp130. This cell normally grows only in presence of IL-3. Antigen used was flourescin (Fl). It was dimerized by linking two monomers with oligo-DNA sequences.
- An antibody (ScFv) was chosen via the phage display-based selection method
- Phage ELISA done to confirm specific binding (using BSA-Fl and only BSA)
- A retroviral DNA for expression of chimeric receptor was constructed as follows:
- ScFv sequence was amplified with PCR and a mouse IgH signal sequence inserted.
- This inserted into a whole bunch of stuff (Reference 10), resulting in construction of pMX-ScFvgIGFP, which included an IRES-EGFP coding sequence to facilitate sorting of cells producing this protein. The pMX-ScFvgIGFP looks like (click on pic to see caption):
- Transfection of packaging cells
- Ba/F3 cells (target cells) then infected/transduced to make Ba/ScFvgIGFP cells, which are sorted by FACS
- Western blotting done against these cells (using anti-gp130) to confirm expression of chimeric protein.
- Test1: the cells cultured with BSA-Fl but without their normal growth ligand (IL-3) --> results in growth (because BSA-Fl will have several different Fl-Fl distances. remember, BSA crowds proteins, hence BSA-Fl is polyvalent)
- Test2:BSA only --> no growth
- Test3: Fl only --> no growth
- Now that we know that a poly-Fl will work, several Fl dimers designed with different distances in the middle (by self-annealing of each 5'-Fl-labeled oligo-DNA with a palindromic sequence. GC rich-seqs were incorporated to attain higher melting temperature). All were incubated with the cells; the best dimer was taken. This dimer caused best growth of cells (since it was right size to dimerize receptors).
- When an enzyme was introduced that broke the linker dimerizing Fl-Fl, no growth was induced anymore.
- ScFv chosen was 31IJ3 from phage antibody library Tomlinson J (with single human framework by 3 rounds of selection)
- Flourescin (Fl) was used as antigen.
- Retroviral DNA vector: pMX-ScFvgIGFP
- Packaging cells transfected: Plat-E
- Transduced cells: IL-3-dependent Ba/F3 cells
- Western blot band for chimera at: 70 kDa
- Distance between two EpoR D2 domains: 35.9 Angstroms (X-ray crystallography)
- Therefore Fl dimers designed with linker lengths: 8 mer (27.2 angstroms) to 14 mer (47.6 angstroms)
- Best linker: 12/13 mer (though all induced growth)
- Reference 10: M. Kawahara, H. Ueda, S. Morita, K. Tsumoto, I. Kumagai and T. Nagamune, Bypassing antibiotic selection: positive screening of genetically modified cells with an antigen-dependent proliferation switch. Nucleic Acids Res. 31 (2003), p. e32.