E. In vitro studies testing whether pleural or peritoneal fluids from malignant effusions support infection 
by retroviral vectors. 
Because our protocol includes infusion of retroviral vector into malignant infusions, it is essential that 
we demonstrate that retroviral vectors can infect cells under these conditions. We obtained pleural and 
peritoneal fluid from 5 patients and incubated a beta-galactosidase virus [37] with these fluid samples for 
1, 4, and 24 hours and then infected 3T3 target cells (the purpose of this experiment was to determine if 
retroviral infection can occur in the presence of malignant fluids). The results are presented in Appendix C; 
Figure 6 and indicate that retroviruses can survive for short periods in pleural and peritoneal fluid. These 
findings are similar to those of Ram which indicate that retrovirus can infect cells in cerebrospinal fluid 
[38]. 
F. Structure of retroviral vector in Producer cells and Target cells 
Southern blot analysis indicates that the retroviral vector and provirus are stable: rearrangement or 
recombination have not been observed. The vector was designed to minimize rearrangement by avoiding 
tandem repeated sequences, GC-rich polylinkers, and using short antisense sequences thereby minimizing 
the chance introduction of unstable sequences. We have employed Southern blotting to analyze stability of 
transferred gene sequences and believe that rearrangement of greater than 5% of transferred DNA could be 
detected by this assay. We have also tested whether retroviruses are stable in these malignant effusions 
and have demonstrated that incubation within these fluids does not destroy infectivity (Appendix C, Figure 
6) 
Preclinical summary: 
We have presented preclinical data which shows that antisense c-fos and c-myc produce inhibition 
of cellular proliferation in a variety of cells through antisense inhibition of their respective target genes. 
Based on this data, breast-targeted retroviral vectors were employed to deliver antisense to MCF-7 breast 
cancer cells in vitro and in a nude mouse cancer model. These studies demonstrated that infection of MCF- 
7 cells with XM6:antimyc retroviral vector prevented these reinfused cells from forming a palpable tumor. 
Infection of MCF-7 cells with XM6:antifos retroviral vector resulted in cells which formed smaller tumors 
that showed less invasiveness and induction of differentiation and programmed cell death (compared with 
MCF-7 cells infected with XM6:sensefos, a control virus). Because only 30-60% of the cells are infected 
under these conditions, these results suggest some sort of "bystander" phenomenon is responsible for the 
observed inhibition of tumor growth. 
II. Clinical Data 
No clinical data has been obtained. 
Recombinant DNA Research, Volume 20 
[169] 
