The plan describe above would require to clone the Polymerases and transcription factors of the virus (whose genome is now fully sequenced) and then use one of these proteins to challenge the immune system of a volunteer and cause an immune response. After antibodies to the polymerase have been produced, the MTS sequences are attached and the antibody is produced by recombinant DNA technology and is added to the cocktail mix. This could be done, one by one for all the proteins that are coded by the virus and that are not displayed on its surface.,

Here is the Technology to clone antibodies directly from B_ cells:

http://www.ncbi.nlm.nih.gov/pmc/articles/PMC2243222/

http://www.ncbi.nlm.nih.gov/pubmed/19716372

One limitations of the antibodies found in the blood of survivors is that most of them would be directed against epitopes found on the outside of the virus. One of the reasons we should make an effort to genetically engineer antibodies is that we could build them to attack viral targets inside the infected cells, which are typically hidden from antibodies circulating in the blood stream. We could accomplish this by attaching membrane translocation sequences (MTS) to the engineered antibodies.

the result would be something like this:

http://www.newscientist.com/data/images/ns/cms/dn4881/dn4881-1_554.jpg

This technology would improve on Zmapp by increasing the number of targets and the numbers of targeted epitopes, and it would also allow the antibodies to disrupt the viral replication inside cells, as they could be targeted to viral polymerases.