_ 0 _ 
The low incidence of laboratory infections with LT2 is to be con- 
trasted to the case of natural isolates of Salmonella which can 
cause gastrointestinal disease with inocula of as few as 1000 
organisms. Consistant with this view are experimental data on mice 
obtained in the Stocker and the Botstein laboratories; in both 
cases particular laboratory LT2 strains were tested directly and 
found to be 1,000 times less virulent than Salmonella typhimurium 
strains isolated independently from nature. An additional important 
point is that LT2 infections, when they occur, appear to be self- 
limiting. We know of no case of secondary infection by LT2 . Our 
point is that, based on danger of laboratory infection, strain LT2 
stands in relation to natural isolates of Salmonella typhimurium 
much as E. coli K12 stands to natural isolates of E. coli , some of 
which cause serious disease. 
The final point which we would like to bring to your attention 
is that the NIH Guidelines incorporated, as a convenience, the CDC 
classification of bacteria according to pathogenicity. We do not 
believe that this convenience should cause NIH to prevent (abso- 
lutely in the present case) experiments with particular strains 
clearly less virulent than many CDC "class I" organisms. We do 
not believe that the scientists who framed the Guidelines and who 
made use of this convenience intended to prevent the use of Sal - 
monella typhimurium LT2 or other strains of formal pathogens whose 
virulence has been shown by experience and experiment to be of low 
virulence . 
We do not know of any scenario by which introduction of LT2 
DNA back into LT2 increases the virulence of LT2 since no new DNA 
combinations are being created. Even transfer of E. coli DNA into 
Salmonella generates no DNA combinations which could not occur in 
nature and which cannot now be made by genetic means. Both plasmid 
and chromosomal DNA pass easily between Salmonella and E. coli 
(10,11,12). Many Salmonella diploids and interstrain hybrids 
between Salmonellae (including highly virulent strains) have been 
made over the past fifteen years with no ill effects. 
What we would like you to do is to concur with your advisory 
committee and permit the introduction of in vitro recombinant DNA 
from Salmonella and/or E. coli into Salmonella typhimurium LT2 
and its derivatives. Thus, we request permission to use in vitro 
cloning techniques to construct strains which can already be made 
by strictly genetic means. Use of new technology merely makes the 
task orders of magnitude easier. If permission to transfer cloned 
Salmonella DNA to Salmonella is granted, we are in a position to 
promise that interesting science will result (see Appendix) , and 
that this involves no risks beyond those encountered in current 
laboratory work. These risks can be easily managed by good labora- 
tory practice. 
[Appendix A — 130] 
