Abstract

Vesicular stomatitis virus (VSV) is a candidate for development for cancer therapy. It is an oncolytic virus that is safe in humans. Recombinant virus can be made directly from plasmid components. We attempted to create a virus that targeted specifically to breast cancer cells. Nonreplicating and replicating pseudotype VSV were created whose only surface glycoprotein (gp) was a Sindbis gp, called Sindbis-ZZ, modified to severely reduce its native binding function and to contain the Fc-binding domain of Staphylococcus aureus protein A. When titered on Her2/neu overexpressing SKBR3 human breast cancer cells, pseudotype VSV coated with Sindbis-ZZ had <1% the titer of pseudotype VSV coated with wild-type Sindbis gp. Titer was increased 50-fold when the Sindbis-ZZ pseudotype was conjugated with 4D5, a mouse monoclonal antibody directed against the Her2/neu receptor. Titers of antibody-conjugated virus were increased 36-fold on a second human breast cancer cell line, MCF7/H2, which expressed lower concentrations of Her2/neu receptor on the cell surface. At multiple concentrations of antibody, titers on SKBR3 cells were significantly greater when the virus was incubated with Herceptin, an antibody with a human Fc, than with 4D5, a mouse antibody, reflecting the known higher affinity of the protein A Fc-binding domain for human Fc. Analysis of the protein composition of the pseudotype VSV found low expression of the modified Sindbis gp on the virus accounting, in part, for a viral titer that did not exceed 1.2 x 10(5)/ml. This work demonstrates the ability to easily create, directly from plasmid components, an oncolytic replicating VSV with a restricted host cell range.