UNIVERSITY OF CALIFORNIA, RERKELEY 
fj.RKEI.EV • DAVIS • IRVINE • LOS ANGELES • RIVERSIDE • SAN DIEGO • SAN FRANCISCO 
SANTA BARBARA •, SANTA GRUZ 
DEPARTMENT OF MOLECULAR BIOLOOY 
WENDELL M. STANLEY HALL 
BERKELEY, CALIFORNIA 94720 
March 9, 1976 
Dr. Donald S. Fredrickson 
Director, National Institutes 
of Health 
Bethesda, Md. 20014 
Dear Dr. Fredrickson: 
Thank you for your letter asking for comments on the proposed NIH guidelines 
on artificially recombinant DNA. Unfortunately, the letter and information 
arrived in Berkeley after the Feb. 9-10 meeting, but I hope that additional comments 
are still in order. 
Generally, I think that the January, 1976 version of the guidelines is a 
reasonable compromise between those of us who are deeply concerned about the poten- 
tial hazard of this technology and consequently are in favor of stringent regula- 
tion and those in favor of pressing on with important experiments with minimal re- 
strictions. However, there are two points that still bother me a great deal. 
First, the drastic reduction of protection for DNA fragments that have been "rigor- 
ously characterized" and are "free of harmful genes" strikes me as highly danger- 
ous (p. 28). I think one can reasonably demand three things for lowered contain- 
ment: (1) the DNA has been rigorously defined (e.g. the hemoglobin gene); (2) the 
DNA can be justified as not dangerous; (3) the cloned DNA does not contain any 
unidentified DNA of size sufficient to code for a small protein (e.g. M.W. about 
5000). The January guidelines cover (1) and (2), but not (3); I believe (3) is 
important and won't stop good experiments (e.g. cDNA from hemoglobin mRNA). Fur- 
thermore, I favor reducing the containment one step in those situations and not 
essentially abolishing it. 
Second, I believe that the health surveillance aspect of this work is of criti- 
cal importance and feel strongly that the January guidelines don't really deal com- 
pletely with some critical issues. On p. 42, it says "Appropriate monitoring of 
personnel .. .should be conducted frequently." How about must for should (at least 
for P3 and/or EK2 situations)? In the next sentence, "evidence has suggested that 
.. .naIR ara~ , or mtl ~ are useful..." Excellent in principle, but where is the im- 
plementing sentence that the bacteria (or better still, the vectors themselves) 
used must carr$de‘dectabl e 1a*g. To me, that requirement seems absolutely essential 
for an EK2 situation in which potentially hazardous genetic material will clearly 
be involved. 
As a final point on a somewhat different topic, it seems to me that now may be 
an appropriate time to extend the safety concerns generated by the recombinant DNA 
work to other biological research, especially animal virus research. I co1 ) ^ 
hear horror stories of careless research in this area (as well as serious i 5 
[ 497 ] 
