7 
SLT clone is sinply cultured in the conditions usually used to culture the 
host-vector system, toxin production is generally a factor of 10^ lover than 
the highest level of expression obtained under cptirral conditions with that 
clone. 
Dr. Newland said he had observed that the recombinant plasmid carrying the SLT 
gene appears to be rapidly lost from the E. coli host-vector system in the 
absence of selective antibiotic pressure. The plasmid without the SLT gene is 
not so rapidly lost in the absence of selective pressure. He and Dr. O'Brien 
hypothesized that high levels of toxin may be toxic to bacteria. The rapid 
loss of the recombinant plasmid in the absence of selective pressure may be an 
additional safety factor. 
Dr. Gottesman ashed whether the working group would consider modifying Appendix 
F-IV-H by raising the upper level of toxin expression to 10® CDs per mg protein 
in cell lysates in exchange for an assurance that the plasmid will be rapidly 
lost in the absence of antibiotic pressure. Alternatively, an upper level of 
expression need not be fixed if the plasmid will be rapidly lost since rapid 
plasmid loss will provide a measure of safety. 
Dr. Levine said one problem with Dr. Gottesman 's proposal is that language 
generic for experiments involving EL_ ooli may not apply to experiments involving 
Vibrio or Shigella . A second concern is the variability of the SLT assay; with- 
out an internal standard it is difficult to know hew much toxin is produced. 
He asked Dr. O'Brien how great a difference is observed using the same procedure 
from experiment to experiment. Dr. O'Brien replied that a 100 fold difference 
in the amount of CDs produced may be observed from experiment to experiment. 
Dr. Gill moved that Appendix F-IV-H be amended to read in part: 
" E. coli host-vector systems expressing the Shiga-like toxin gene product 
may be moved f ran BL3 + EX1 to BL2 + EK1 containment conditions provided 
that: (1) the amount of toxin produced by the modified host vector systems 
be no greater than that produced by the positive control strain 933 E. 
coli 0157:H7, grown and measured under cptirral conditions, or ten times 
this level if the maintenance of the plasmid carrying the gene is dependent 
upon growth in the presence of an antibiotic...." 
Dr. Levine suggested the motion be amended to read in part: 
" E. coli host-vector systems expressing the Shiga-like toxin gene 
product may be moved from BL3 + EX1 to BL2 + EK1 containment conditions 
provided that: (1) the amount of toxin produced by the modified host 
vector strain be no greater than 10® CDs per mg protein. ..." 
Dr. Formal suggested the motion should name a toxin producing strain as the 
standard rather than citing a specific nurrber. He suggested Shigella dysenter - 
iae 1 strain BOR be used for this purpose. Shigella dysenteriae 1 strain 60R, 
a rough strain not capable of colonizing the human gut, produces about ten 
[ 311 ] 
