Dr. Fredrickson 
2 
15 May 1978 
Streptomyces lividans (which are non-exchangers) in order to design a 
more effective antibiotic, could not be done in the wild type antibiotic- 
producing strain ar_ any level of physical containment whatsoever once the 
revised guidelines have gone into effect. The anachronism of prohibiting 
cloning of recombinant DNA molecules containing genes from these two non- 
pathogenic species is especially striking in view of recent changes in the 
very same guidelines that allow a wide variety of genes from infectious 
animal cell viruses to be introduced into Eh_ coli K12 at sharply reduced 
levels of containment. 
Surely, the committee could not have intended to prevent the experi- 
ments I have described above from being done at all. Yet, unless the 
wording of the proposed revised guidelines is modified by you, a whole 
class of experiments currently being carried out in entirely safe non-E . 
coli K12 systems will necessarily come to a halt in the United States 
when the new guidelines go into effect. 
The changes in the definition of recombinant DNA in the revised 
guidelines do not appear to solve the problem. Moreover, in addition 
to precluding use of genetically unmodified non-pathogenic organisms as 
hosts, the guidelines' definition of an HV1 host-vector system as one 
that has "limited survival outside of the laboratory" is ambiguous. 
Does this mean all non-laboratory locations ranging from sewage to soil 
and air? What data would be required to demonstrate the limited survival 
outside of the laboratory of organisms such as Streptomyces and Pseudomonas ? 
Would Eh_ coli K12 continue to qualify as an "HV1 host" under the new 
definition? Certainly, limited survival outside the laboratory has not 
been demanded of E. coli K12 (or of yeast, for that matter) and it is not 
clear why this requirement is imposed for other organisms that are at 
least as non-pathogenic as E^_ coli K12. 
The solution I propose is a modification of my earlier suggestion. 
The revised guidelines should simply state that "any bacterial species 
not known to be pathogenic to humans, to domestic animals or to agricul- 
turally important plants may be used as an HV1 host-vector system". HV2 
and HV3 host-fector systems would of course require certification by the 
Recombinant DNA Molecule Advisory Committee. Under this arrangement, 
non-E. coli K12 hosts could be used as recipients for DNA molecules that 
can be cloned at present in an EK1 host (i.e., those molecules which have 
been determined by the committee to present the lowest degree of risk) . 
It seems to me that this proposal establishes a reasonable basis for the 
designation of non-E. coli K12 microorganisms as HV1 hosts. 
Although only a limited number of investigators in the United States 
are presently working with non-E. coli K12 hosts as potential recipients 
[Appendix A — 307] 
