Document 25 
- 1 - 
COMMENTS ON THE RECOMBINANT DNA RESEARCH GUIDELINES - 
DRAFT ENVIRONMENTAL IMPACT STATEMENT 
The Recombinant DNA Research Guidelines represent a serious effort by many 
responsible scientists to evaluate the possible hazards of recombinant DNA 
research and to provide, before the fact, sensible methods for their control. 
The fact that individuals involved in this research voluntarily, and without 
outside pressure, temporarily stopped their work to develop methods for increas- 
ing its safety, is without precedent, and is a considerable tribute to their 
sense of responsibility and concern for human welfare. 
It is therefore most unfortunate that the Guidelines as constituted seriously 
underestimate, in my opinion, several risks: (1) The "catalytic function" served 
by the scientist may create hybrids previously not tested for survival by natural 
selection; (2) The hazard of vectors such as Escherichia coli and SV AO may be 
much larger than estimated, for reasons to be outlined; (3) The use of physical 
containment, while necessary, will provide only a slight increase in safety. 
Thus, the use of safe vectors must be the main mechanism of hazard control. I 
will try to amplify my concerns in the subsequent paragraphs and then indicate 
some changes in the Guidelines which I believe would increase the safety of 
recombinant DNA research. 
I. The Creation and Survival of Interspecies Recombinants 
It is stated in the Guidelines that "There are various indications that 
bacteria . . . containing . . . foreign DNA are less likely to survive and mul- 
tiply than are the original organisms." Similar statements, both in the Guide- 
lines and in the scientific press, have noted that, because of the long period 
of time allowed for natural selection of organisms uniquely suited to their 
ecological niche, most genetic changes can be expected to be deleterious. How- 
ever, as also noted in the Guidelines, recombinants between related species of 
bacteria probably occur in nature, but recombination between bacteria and animal 
DNA is likely to be extremely rare, or perhaps nonexistent, because of the low 
homology between the DNA's involved. We must therefore re-examine the argument 
that recombinants formed in the laboratory will be unlikely to survive, because 
many of the recombinants of which we are speaking may not have had enough time 
to occur in nature since the evolution of Homo sapiens . Lest it seem foolish to 
state that a reaction that thermodynamically can occur will not occur over such 
a long time span, consider the formation of a peptide bond. It has been calcu- 
lated [Koshland and Neet, Annu. Rev. Blochem. 37 , 359 (1968)] that this reaction 
is 10~20 to 10“24 as likely to occur in the absence of an enzyme as in its 
presence. If an enzymatic reaction takes 10“2 seconds, then we can safely say 
that peptide bond formation would not take place, in the absence of an enzyme, 
during the length of time that Homo sapiens has been on earth (1-3 x 10^ years). 
By analogy, it is quite possible that certain interspecies recombinations have 
not taken place within that time span. We will now provide the catalyst — just 
as the enzyme catalyzes peptide bond formation — and we cannot predict the 
survival value of the untested product. 
Appendix K — 119 
