- 2 - 
The rest of what I have to say bears on a rational re-consideration of 
potential hazards of recombinant DNA research. By "rational", I mean related 
to prior experience. I think we have had available relevant experience with 
microbes that was not well utilized in the development of the current guidelines, 
but is reflected in the proposed revisions. For example, the experimental 
evidence that non-pathogenic bacteria like £. coli K12 are not converted into 
contagious pathogens by acquiring small clusters of additional genes, even from 
donor pathogenic strains, has been emphasized by medical microbiologists meeting 
at Falmouth. We have had long experience in the safe handling of pathogenic 
microbes. In most cases simple microbiological methods suffice. For instance in 
our medical microbiology course, students work with cultures of Neisseria 
meningitidis , a cause of meningitis; Shigella dysenteriae , the cause of dysentery; 
Salmonella typhi , the cause of typhoid fever; and many other real pathogens. 
Over a 16 year period I have not been aware of a single case of disease related to 
these laboratory exercises, in spite of periodic spills and obvious infection. 
This experience also illustrates the well-known fact that so-called "escape" of 
microbes from the laboratory or infection of a laboratory worker is far from 
equivalent to disease production, let alone epidemic disease, which is almost 
unheard of as a result of laboratory infection. I would argue that a century of 
experience with safe handling of real pathogens should influence the requirements 
for handling non-pathogenic organisms containing foreign DNA, whose potential 
pathogenicity is entirely conjectural. 
In regard to viruses, the proposed revisions are based on a similar principle, 
namely that as a rule conditions used to work with the virus itself are adequate 
to work with recombinants containing segments of the viral genome, or with 
defective virus vectors. It is difficult to imagine how E_. col i K12 containing 
a segment of SV40 DNA could be more hazardous than the SV40 virion itself. 
Likewise, defective SV40 with a foreign segment of DNA is much less capable of 
spreading in a population than is intact SV40. Moreover, it is clear that each of 
the mammalian viruses proposed as a vector for recombinant DNA (polyoma, SV40, 
murine and human adenoviruses) readily recombines with host cell DNA to form a 
large variety of recombinant viruses. In other words, we appear to be exposed to 
shot-gun type viral recombinants all the time in our natural environment. 
Finally, I would like to comment on the proposed role of local biosafety 
committees in monitoring recombinant DNA research. It has always made sense to me 
to make the institution responsible for its own recombinant DNA research activities, 
as it is for other microbiological research and for use of radioactive materials. 
At Johns Hopkins the biosafety committee is serious in purpose and effective; its 
authority is recognized by everyone. Therefore, I support the proposed increase 
in local responsibility. Since revised guidelines specify the composition of the 
biosafety committees (including non-scientists and at least one member from outside 
the institution) and require prompt coranuni cation to NIH of new proposals, 
infractions, etc., I believe there is sufficient monitoring of institutional 
compliance with the guidelines. 
[ 382 ] 
