STANFORD UNIVERSITY MEDICAL CENTER 
STANFORD, CALIFORNIA 94305 
DEPARTMENT OF BIOCHEMISTRY 
PAUL BERG 
Willson Professor of Biochemistry 
August 20, 1979 
Dr. Donald Fredrickson, Director 
National Institutes of Health 
Building 1, Room 124 
Bethesda, Maryland 20014 
Dear Don, 
One of the principal recommendations in The Asilomar Conference Report on Recombi- 
nant DNA was that the guidelines for containment and administrative procedures 
for recombinant DNA techniques be reviewed continuously. We expected that in time 
and with growing experience our views about the nature of the risks would change 
and this would be translated into corresponding changes in the detailed recommen- 
dations in the NIH Guidelines. It is to your credit that you prevailed in incor- 
porating a procedure for orderly changes and this procedure, though time consuming, 
is working. 
I do want to take this opportunity to comment on the proposed exemption for E. coli 
K-12 Host-Vector Systems from the Guidelines. When the group of which I was a mem- 
ber expressed our concern about potential risks associated with recombinant DNA 
research, we had in mind the possibility that such experimentation might employ 
a wide variety of E. coli strains and transmissable vectors as host-vector systems. 
We imagined that some of these would be able to establish themselves in nature 
or in the intestinal tracts of man and animals. Had we known then that E. coli 
strain K-12 and non-transmissable plasmids or phages would be widely adopted as 
the preferred host-vector systems, and that this system would be as secure as all 
the risk-assessment experiments have shown, I doubt that we would have raised the 
issue in the manner we did. I am persuaded by the evidence I have seen that mole- 
cular cloning of any DNA segments in E. coli K-12 using the array of present day 
cloning vectors is no longer of any real concern, certainly not enough to warrant 
the continued inclusion of such experiments within the purview of The Guidelines 
and the IBC’s. Perhaps it would still be useful to maintain a log of recombinant 
DNA experiments in this category so that these experiments can contribute to the 
ongoing assessment of the safety of such research activities. 
With best personal regards, 
Sincerely, 
Paul Berg 
