viral proteins necessary to package RNA into intact 
infectious virus. 
The LNL6 and GlNa.40 retroviral vectors contain additional 
modifications designed to decrease the likelihood of 
recombination events between the retroviral vector and the 
packaging cell line helper virus genome that could generate 
infectious helper virus. 53,54 These modifications include 
deletion of 5' and 3' regions homologous to parts of the 
packaging genome, and a mutation of the gag region start 
codon. More detailed information on these marking vectors is 
given in Appendix F. 
Producer cell lines: Since the LNL6 and GIN. 40 vectors are 
replication-defective and lack viral genes, their genome 
must be packaged into a virion with gag, pol and env gene 
products supplied in trans by a MoMLV helper genome. This 
is accomplished in a packaging cell line that contains the 
necessary gag, pol and env genes in a modified MoMLV 
proviral genome, but with. deletions that make this helper 
genome incapable of packaging itself. Fibroblast cell lines 
producing the LNL6 or the GIN. 40 vector were generated by 
introducing the LNL6 or GIN. 40 proviral plasmids via calcium 
phosphate co-precipitation into the PA317 packaging cell 
line, and selecting high-titer (1-5 X 10 6 viral particles/ml 
supernatant), helper-virus-free clones. Clinical grade, FDA- 
reviewed vector-containing supernatants from these producer 
cell lines are produced at Genetics Therapy Incorporated, 
and extensively tested for contaminating microorganisms, 
replication-competent virus, and biologic titer of 
retroviral particles prior to use. (See Appendix G for a 
list of safety tests performed on retroviral supernates by 
GTI prior to release) Over 200 lots of clinical-grade LNL6 
and GIN. 40 supernatants have been produced at GTI, and none 
have shown evidence of helper-virus contamination. 
Theoretical risks: There are a number of theoretical 
concerns regarding the use of retroviral producer lines for 
the transduction of long-lived marrow precursor cells. 
There is the possibility that replication-competent virus 
may be generated via recombination events within the 
producer cell line, or within the target cell. The risk is 
exceedingly low, due to the safety modifications made in the 
vector as noted above, the extensive testing of the producer 
cell line for helper-virus production, and the lack of known 
endogenous human retroviral sequences with homology to the 
vector that would allow recombination within target cells. 
A second concern is insertional mutagenesis. Proviral DNA 
could insert near or within cellular genes and either turn 
off or turn on gene expression, or lead to the production of 
an aberrant gene product. If such a gene were involved in 
Recombinant DNA Research, Volume 16 
[15] 
