5 
HVI systems may be used under P2 containment conditions for shotgun experiments 
with phages, plasmids, and DNA from nonpathogenic prokaryotes which do not pro- 
duce polypeptide toxins. Therein is the quandary! Both of the proposed organisms 
for recombinant DNA research are pathogenic, but only to target insects, and one 
of them, B^. thuringiensis produces a polypeptide toxin which is the basis of 
its pathogenicity. The toxin, however, is only specific for certain insects and 
has been deomonstrated ineffective against other life forms - including man. 
I would respectfully request that the NIH define more concisely the nature of 
the pathogenicity and the types of polypeptide toxins meant in the revised guide- 
lines. That is to say, pathogenic and/or toxic to what life forms? Further, 
the investigator is hard pressed to analyze the biological containment require- 
ments for such experiments. Can an organism that grows in an insect or is 
found as a common soil organism, but only affects certain insects, be developed 
for pest control use and be considered an HV-1 system for shotgun experiments 
using DNA from entomopathogenic bacteria. The revised guidelines apparently do 
not adequately address such types of bacterial recombinant DNA experiments. 
We further would ask the' NIH to consider that the entomopathogenic bacteria or 
the baculoviruses be only required to meet the criteria of EPA for a temporary 
exemption from a requirement of tolerance which verifies that the organism has 
been rigorously tested for safety to mammals and other non-target organisms in 
the environment rather than have EPA registration for a specific containment 
level. Registration. of the organism is granted only afterwards provided the 
organism is shown to be efficacious. 
In section III-C-3 (Plant Viral Host-Vector Systems) dealing with DNA 
plant viruses serving as vectors for cloning genes in plants, we question the 
statement that "the plants should be grown under PI conditions - that is in 
[ 310 ] 
