NOTICES 
60089 
this has been done. For Sections III- 
A-l-a-(l) and III-A-l-a-(Z) (DNA 
from mammals), the RAC recommend- 
ed addition of the phrase “or P3 + 
EK1 with a non-mobilizable plasmid.” 
Because the term "non-mobilizable 
plasmid” is not clearly defined, I am 
not accepting this recommendation at 
present, pending further review at the 
next RAC meeting. 
Section III-A-2-a. DNA From Virsus of 
Eukaryotes Into E. Coli K-12 
There was considerable comment 
about various aspects of Section IU-A- 
2-a, extending from general concerns 
about the reduction of containment 
levels from viral inserts to detailed 
and specific recommendations for 
clarification of certain phrases. 
Several correspondents objected to 
the general relaxation of containment 
levels for the cloning of viruses. One 
stated that although the Ascot confer- 
ence concluded that "cloning of the 
whole or any part of a viral genome 
must logically be less dangerous than 
working with the virus,” his view is 
just the opposite. The arguments pre- 
sented by this correspondent are (1) 
whole virus can elicit the production 
of antibodies, (2) whole virus with its 
protein ooat is subject to a biological 
barrier which limits infection across 
species lines, and (3) whole virus Is 
eliminated from the body after infec- 
tion. In the view of this correspon- 
dent, all of these protective devices are 
subverted by cloning viral genes in a 
microorganism that may become es- 
tablished in the intestinal tract. He is 
also concerned that the Guidelines are 
not sufficiently stringent for cloning 
subgenomic DNA fragments of viruses. 
Another respondent indicated that 
belief that containment experiments 
involving animal-cell transforming vir- 
uses were substantially reduced, based 
on the supposed inefficiency of infec- 
tion by naked DNA. She continued by 
stating that the decision to lower con- 
tainment did not take into account the 
results of the Rowe-Martin risk -assess- 
ment experiments with polyoma DNA, 
which showed that naked polyoma 
DNA when injected into the blood- 
stream of mice caused a low-level in- 
fection. 
The recommendations of the Ascot 
Workshop (Appendix E to the Envi- 
ronmental Impact Assessment), the 
subsequent Working Group review 
(Appendix F to the ELA), and the sub- 
sequent RAC review took into account 
the concerns of correspondents about 
viral containment, but nevertheless 
recommended lowered containment 
levels based on the conclusion that 
cloning of viruses or their fragments 
in E. coli could be no more dangerous 
than working with the intact virus. 
The containment levels were set ac- 
cordingly. The three reasons cited 
FEDERAL 
above as to why the cloned virus DNA 
might be more dangerous than the 
whole virus apply only to the first in- 
fected cell. The DNA would produce 
disease only if it became virus and 
spread; thus, the situation becomes 
the same as the usual virus Infection. 
This possibility was thoroughly con- 
sidered at Ascot, and is in the sum- 
mary report of that meeting. Also, the 
existence of a subgenomic fragment 
precludes virus production and so ef- 
fectively prevents the spread of infec- 
tion. 
The Rowe-Martin risk-assessment 
experiments (still in progress) are de- 
signed to compare the infectivity of a 
recombinant molecule containing po- 
lyoma DNA with that of nonrecombin- 
ant polyoma DNA and of whole po- 
lyoma virus. To date, there are no data 
available from these experiments 
which would lead me in any way to 
revise the Judgements of the meetings 
cited above. 
In response to the concern expressed 
about naked DNA, there is a signifi- 
cant difference in level of infectivity 
between naked DNA and whole virus 
particles; naked DNA is much less in- 
fectious than the virus. For additional 
discussion of naked DNA, see part I of 
this document. 
A correspondent stated that there 
should be some definition as to wheth- 
er a DNA virus is a transforming or 
nontransforming virus. To clarify this, 
a new footnote (37 A) has been added 
to Section III-A-2-a-( 1 Mb). 
A correspondent pointed out that 
there are places in Section III-A-2-a 
where the word “purified” appears 
without referencing to Footnote 38. 
This has now been corrected by insert- 
ing reference to Footnote 38 where 
necessary, and inserting the phrase 
“subgenomic segments that have not 
been purified to the extent required in 
Footnote 38” at other places. 
A correspondent discussed 
"EK1CV.” defined in Footnote 40 as 
the "the use of an EK1 host and a 
vector certified for use in an EK2 
system.” He points out that certain 
EK1CV systems provided containment 
comparable or almost comparable to 
EK2, while others are only slightly 
better than EK1. This is true. Howev- 
er, in all cases the level of contain- 
ment is higher than EK1. The cases in 
the Guidelines where EK1CV contain- 
ment is specified as an option were so 
recommended by the RAC on the basis 
of the recommendation of the April 6- 
7, 1978, Virus Working Group (Appen- 
dix F to July 28, 1978, Environmental 
Impact Assessment). 
Section III-A-3. Lowering of Contain- 
ment for Characterized or Purified 
DNA Preparations and Clones 
As in many other sections of the 
Guidelines, commentators have ex- 
REGISTER, VOL 43, NO. 247— FRIDAY, DECEMBER 
[ 12 ] 
pressed their opinions in both general 
and specific terms. A commentators 
states that there is no information at 
present to “show any hazard deriving 
from recombinant DNA or organisms 
harboring such DNA * * *. In particu- 
lar, it would seem quite clear that 
characterized cloned eukaryotic DNA 
elements in E. coli pose no conceivable 
hazard.” Pursuing this line of reason- 
ing, the commentator argues for elimi- 
nation of regulation and, in particular, 
suggests that “consideration be given 
to a change in the Guidelines so that 
institutional biohazard committees 
could be empowered to exempt charac- 
terization clones from further regula- 
tion.” 
The proposed revisions in the Guide- 
lines empower the IBCs to give ap- 
proval for a single-step reduction in 
physical or biological containment 
upon receipt of evidence of character- 
ized of a clone and its freedom from 
harmful genes. Further reductions, or 
cases involving primate DNA or lower- 
ing of containment levels below PI + 
EK1, require prior approval by NIH. I 
do not believe it prudent to accept the 
commentator’s recommendation at 
this time. 
Two oomments were received on the 
criteria for the terms "purity” and 
“free from harmful genes.” Once com- 
mentator expressed concern over the 
inadequate definition -of these terms. 
He believes they are not defined with 
sufficient precision to guarantee uni- 
form decisions, particularly when the 
authority to lower containment by one 
step is left with the IBC. The com- 
mentator suggests more rigorous crite- 
ria for purity (other than 99 percent) 
and a broader definition of "harmful 
gene” to include the concept "that 
genes which might not be harmful 
when expressed in their original orga- 
nism could indeed be harmful if ex- 
pressed out of context in an unrelated 
organism.” 
The other correspondent discussed 
Footnote 41 which requires that the 
"desired DNA represents at least 99 
percent (w/w) of the total DNA in the 
preparation” before it may be consid- 
ered “purified.” He stated: “Its adop- 
tion reflected the supercautious mood 
prevalent at that time rather than a 
clear-cut scientific judgement. In my 
view, requiring that a DNA fragment 
should be 90-95% pure (as judged by 
at least two different analytical proce- 
dures) to remove it from the shotgun 
classification is more realistic and no 
less safe. Requiring that a DNA frag- 
ment be >99 % pure prior to cloning 
asks for the most stringent and de- 
tailed documentation without any real 
advantage.” 
The term "purity” is very explicitly 
defined in Footnote 41. The require- 
ment is that the desired DNA must 
represent at least 99 percent (w/w) of 
22, 1978 
