2677 EHenda 1 . c Place 
Los Angeles, California 9CC07 
(213) 733-9307 
October 17, 1977 
Donald S. Fredrickson, M.D. , Director 
National Institutes of Health 
9000 Rockville Pike 
Pethesda, Marvland 2001k 
Dear Dr. Fredrickson: 
My comments on the proposed revised Recombinant ENA Research Guidelines 
(Federal Register h2 . #187, September 27, 1977, pp. 1 i9596-Ii 96C9) are these: 
1) I note with favor that the list of unfundable "forbicdrn" experiments 
(p. U96C1) is virtually identical with the 1976 list (7R, Jni;. 7, 1976, pp. 
2791U-279155 . Full retention of this set of proscriptions realistically rec- 
ognises the continuing imperfectness of both physical and biological containment. 
?) However, I must advise once again that funding guidelines are inherently 
inadeouate to cope with the potential biohazards of recombinant INA research. 
The Guidelines consist in essence of a series of recommendations for containment 
levels. r hey reflect the assumption that relatively harmful genes can ce accur- 
ately classified by their organismic source, and thus assigned to confinement 
sufficient to prevent escape and establishment beyond the laboratory. Recent 
research on the process o f transcription has vitiated the assumption of basically 
linear relationship between gene and gene product. The Guidelines admit that the 
term "free of harmful genes'* is "unavoidably vague" (p. U9605). Research since 
promulgation of earlier versions of the Guidelines has changed "unavoidably" to 
"enormously". An extremely brief summary of this work would include these points: 
% It has been found that bacterial genes may sometimes jump from place to place 
on ENA molecules ( Science. 30 July 1976, p. 39?). Moreover, gene sequences some- 
times overlap and occasionally may be completely nested within a different gene's 
ENA sequence ( science . 10 June 1977, p. U87). Furthermore, "spliced" messenger 
RNA .has been found which is transcribed from three separated areas of a viral ENA 
molecule ( Science Nev3. July 30, 1977, p. 70). In addition, "action at a distance" 
has been discovered— a protein bound to UTA at one spot can affect gene expression 
at a position some distance away ( Science. 7 October 1977, p. Ul). Cumulatively, 
these findings Indicate that genes cannot be conceptualized as simple uninterrupted 
and stable nucleotide sequences. It follows then, since unknown idiosyncrasies in 
transcription are crucial, that we cannot confidently prevent the formation of harm- 
ful gene products, or their escape, by matching various source (donor) UJAs with 
various containment procedures. 
3) Nor can assurance be drawn from the apparent absence of adverse health 
effects in recombination experimentation to date, for these reasons: (A) Recombin- 
ants potentially harmful to important target species may have already been created, 
but may simply not have been viably carried to those species, (B) Recombinants 
potentially harmful to man may have gone unnoticed thus far due to the fortunate 
good health, irmuno logical history, or inherited resistance of the laboratory 
personnel exposed to them, (C) Harmful recombinant genes transferred to viable 
[Appendix A — 17] 
