8 
Dr. Landy said sane of the cells in the GI tract may have receptors for growth 
factors. Could the presence of large anounts of growth factor in the GI tract 
result in tumor formation? Are growth factors highly biologically active 
molecules? Would exposure to snail amounts have an effect? 
Dr. Martin con lectured that an animal exposed to large amounts of a potent 
growth factor secreted by an E^_ coli host-vector system could develop a tumor 
of the GI tract. He did not know if the tumor would recede when the growth 
factor was no lonqer present. He did not think the tumor would continue to 
grow in the absence of growth factor. He pointed out, however, that the effect 
of growth factor on cells in culture cannot be compared with the _in vivo situa- 
tion; the normal intestine continually grows but growth factors are assayed by 
their stimulatory effect on stationary phase tissue culture cells. 
Dr. Anderson agreed. He said fibroblasts in tissue culture may possess receptors 
for specific growth factors; however, the intestinal mucosa is designed to 
absorb nutrients and may not possess receptors for growth factors. 
Dr. Simpson did not think enteral exposure to an oncogene product would be 
of great concern. He would be more concerned with parenteral exposure. He 
cited the case of Shiga toxin which is very potent when introduced parenterally 
in small amounts but has no effect on monkeys when large amounts are introduced 
into intestinal pouches. 
[RaDporteur 's Note: In 1953, Branham, Dack, and Rigqs showed that large amounts 
of Shiga toxin instilled directly into monkey intestinal pouches has no effect. 
These investigators administered to monkey intestinal pouches with no observed 
effect the equivalent of approximately 20,000 lethal doses administered paren- 
terally. In a worst case scenario, 10^ engineered coli with the Shiga 
toxin gene on a high expression, high copy number plasmid might produce one 
milligram of toxin in the human gut. This anount is roughly equivalent to 
approximately 14,000 lethal doses for humans when the toxin is administered 
parenteral ly.l 
Dr. Gottesman asked whether any mechanisn exists by which a retro/irus cloned 
in E. coli host-vector systems could become an infectious particle. She asked 
whether the retroviral transformation systen could function in prokaryotic 
systems and whether prokaryotic systems could enhance retroviral replication. 
[Rapporteur's Note: Che of the initial concerns regarding the use of E_. coli 
host-vector systems to molecularly clone animal virus qencmes was that the 
bacteria might spread epidemically throuqh the ecosysten and deliver potentially 
infectious virus particles/virus gencmes to tissues not ordinarily exposed to 
such agents. TO address some of these concerns, Chan et al. , and Israel et al. , 
conducted experiments to determine whether coli K-12 harboring recombinant 
phaqe or plasmids (such as pBR322) containing the polyoma virus genane (PY) 
could cause a spreading infection in sensitive test animals (mice). In the 
experiments, genetically disabled coli K-12 containing PY-pBR322 reccmbinant 
plasmids was fed or inoculated into mice. None of the animals developed a PY 
[ 447 ] 
