Vol. 296 No. 21 
EDITORIALS 
1227 
So programmed, bacteria are envisioned as factories 
for the production of hormones, antibiotics, antibod- 
ies and enzymes of potential use in medicine and in- 
dustry, as well as other gene products needed for ba- 
sic research. So engineered, the normal control sys- 
tems regulating bacterial growth and function may be 
circumvented. 
Recombinant DNA technology is now in wide- 
spread use in basic-research laboratories, with the 
number of such experiments constantly increasing. A 
not unreasonable estimate is that 10' 5 to 10 16 recom- 
binant organisms per year will be produced in the lab- 
oratories and that this number will be substantially 
increased by industrial involvement.' Like the bene- 
fits of nuclear-power plants, the benefits of recombi- 
nant DNA research are likely to come with mixed per- 
il. The National Institutes of Health and other pro- 
ponents of the technology are to be lauded for their ef- 
forts in struggling with this dilemma and the result- 
ant development of guidelines for future studies. In 
some instances these guidelines seem particularly ef- 
fective when dealing with the more obvious concerns 
of public-health nature by prohibiting overtly dan- 
gerous classes of experiments — for example, the clon- 
ing of recombinant DNA’s from known oncogenic vi- 
ruses or pathogens. 2 
The real shortcoming of the guidelines is that they 
do not give equal importance to potential problems 
developing from less obvious situations than the 
known pathogens mentioned above. The enormous 
complexity of micro-organisms and ignorance of their 
intricately complicated ecology make it impossible to 
predict with certainty the end effect of any transplant 
of large numbers of genes into the micro-organism of 
choice, Escherichia coli K-12, ora further enfeebled ge- 
netically modified strain derived from K-12. Escli. coli 
is indigenous to human beings and all warm-blooded 
animals, is carried by many other organisms such as 
insects and fish, and can be found throughout the bio- 
sphere in soil and water. This microbe can normally 
exchange genetic information with 40 other species of 
micro-organisms, and though the likelihood of ex- 
change from the genetically enfeebled strain has been 
markedly reduced, it still may occur. The vast sea of 
micro-organisms in the environment makes events 
whose probability is the reciprocal of Avogadro’s 
number (i.e., 1/10 23 ) not impossible.* 
Many experiments being carried out employ “shot- 
gun” technics where complete chromosomes of unde- 
fined gene sequence are split into many segments fol- 
lowed by insertion of these unknown genes into Esch. 
coli with potentially unpredictable results. Every im- 
planted gene becomes a new hypothesis, and there 
will be tens of thousands of such new hypotheses in- 
serted. This is the “Catch-22” dilemma associated 
with the technology. Could we inadvertently, via un- 
expected transfer of genetic information from the en- 
‘Curlis R: Potential biohazards of recombinant DNA molecules (unpub- 
lished memorandum to Paul Berg et al.) August 6, 1974. 
feeblcd host strain to one of the estimated 10" to 10 M 
healthy Esch. coli organisms in the human bowel, cre- 
ate a novel self-replicating pathogen? Do we know fof 
certain the effect in human beings or any other crea- 
ture of the biosphere, of a novel micro-organism pro- 
ducing a powerful insulin-like hormonal activity? 
Cloning of the insulin gene is but one of many experi- 
ments now in progress, any of which could have po- 
tential toxic or autoimmune effects. What effect might 
such a micro-organism have on other varieties of en- 
feebled hosts — the human patients who are on an- 
tibiotics therapy, who are immunosuppressed or who 
are otherwise debilitated by disease. In some cases 
these patients reside in hospitals that also include re- 
combinant DNA research laboratories. We do not as 
yet know the answers to the above questions. These 
questions should be addressed, and their answers and 
solutions framed in a new and badly needed public- 
health policy. 
That the NIH is providing millions of dollars for 
construction of containment facilities for some of this 
research is to its credit. Yet the “Fort Detrick Experi- 
ence” 3 and the records of the Center for Disease Con- 
trol make it painfully clear that absolute containment 
of micro-organisms is impossible — even in laborato- 
ries where workers are exceedingly well trained and 
experienced in working with pathogenic organisms. 
Human frailty will guarantee mistakes. What, then, 
can be expected from numerous new laboratories re- 
cently involved in recombinant DNA research whose 
personnel have at best had a three-day course on con- 
tainment procedures? Here, we have the potential for 
a unique form of pollution, biologic pollution, and it 
will be distinguished by its irreversibility. In contrast 
to nuclear-power plants, which can be closed down, 
micro-organisms will reproduce and perpetuate them- 
selves. They cannot be recalled. Unlike the defined 
pathogens studied at Fort Detrick, their escape can- 
not be monitored for. 
The great deal of time and effort put into self-regu- 
lation, and the development of biologic and physical 
containment systems by the proponents of the tech- 
nology, is unprecedented. They have set an example 
of genuine concern for human well-being and the bio- 
sphere. But they have not faced containment of their 
own exponentially expanding activities, which may be - 
an equally irreversible route to ecologic disaster at the 
rate at which the research activities are expanding. In 
reality, we have all become experimental subjects for 
recombinant DNA experiments. Yet none of us have 
signed an informed-consent statement. 
A risk-assessment panel has recently been con- 
vened by the NIH to determine the risks involved and 
where information is lacking to propose experiments 
to obtain the information. This is a proper and im- 
portant step in terms of public-health considerations. 
The burden of proof or lack of hazards should rest nei- 
ther on the opponents nor on the proponents of the re- 
search. Their role should simply be to distill and de- 
fine the issues for the remainder of the biomedical 
[Appendix C — 92] 
