NOTICES 
33061 
P3-EK1 in every case where it ap- 
peared in the PRG-RAC. 
The section of this document on 
Recombinant DNA Experiments In- 
volving Viral DNA" discussed the 
"Ascot" workshop report, and the 
April 6-7. 1978, working group report 
which endorsed the "Ascot" report. 
The RAC at its April 27-28. 1978. 
meeting unanimously endorsed the 
working group report recommending 
lower containment levels for deliber- 
ate cloning of viral DNA into E. coli 
K-12 (see below for discussion of sec- 
tion III-A-2). One of the reasons given 
originally for the higher containment 
level for shotgun experiments involv- 
ing primate DNA into E. coli K-12 was 
the possible inadvertent cloning of 
viral DNA. In view of their recommen- 
dation of lower containment for delib- 
erate cloning of viral DNA into E. coli 
K-12. the RAC on April 27-28. 1978. 
reconsidered primate shotgun levels, 
and voted unanimously for new lan- 
guage as follows: " Primates . P2 physi- 
cal containment + an EK2 host- 
vector. Any lowering of containment 
below these levels (i.e., for purified 
DNA or characterized clones) cannot 
be made solely by an institutional bio- 
safety committee but requires NIH ap- 
proval." I have accepted this new lan- 
guage and inserted it in the PRG- 
NIE. as well as a similar lowering of 
conta inm ent for shotgun cloning of 
cold-blooded vertebrate DNA into E. 
coli K-12. 
One commentator noted that section 
III-B-l-a-awg) of the PRG-RAC en- 
titled “Cloning of Viral Genomes 
From Eukaryotic Ceil DNA” * * * “fo- 
cuses on cloning integrated retrovirus 
nucleotide sequences from mammalian 
cell DNA but says nothing about nu- 
cleotide sequences of integrated DNA 
viruses." This entire section has been 
eliminated from the PRG-NTH and in- 
stead a new subsection III-A-2-a-<3) 
entitled "Intracellular Viral DNA" has 
been added to the PRG-NIH which 
covers both integrated retroviruses 
and DNA virus sequences: it says. 
"Physical and biological contaminant 
specified for shotgun experiments 
with eukaryotic cellular DNA (See sec- 
tion III-A-la) shall be used for DNA 
recombinants produced with integrat- 
ed viral DNA or viral genomes present 
in infected cells.” 
Section III-A-l-b. Shotgun Experi- 
ments Into E. Coli K-12 With In- 
serted Prokaryotic DNA 
In the 1976 guidelines, the section 
(III-B-2-(a)-<ii)) dealing with shotgun 
experiments into E. coli K-12 with in- 
serted prokaryotic DNA was subdi- 
vided into two sections, i.e., "Prokar- 
yotes That Exchange Genetic Infor- 
mation With E. coli" and Prokar- 
yotes That Do Not Exchange Genetic 
Information With E. colt" In the 
PRG-RAC it was assumed that all pro- 
karyotes that exchange genetic infor- 
mation with E. coli would be exempt 
from the guidelines by appearing on 
the “list of nonnovel exchangers." 
Therefore, in the PRG-RAC the sec- 
tion (III-B-l-a X2» dealing with shot- 
gun experiments into E. coli K-12 
with inserted prokaryotic DNA actual- 
ly considered only prokaryotes that 
did not exchange genetic information 
with E. colt The problem with this ap- 
proach was discussed by commenta- 
tors, focusing especially on the case of 
Agrobacterium tume facie ns. It meant 
that a prokaryote which exchanges ge- 
netic information with E. coli, and was 
therefore properly assigned a low con- 
tainment level under the 1976 Guide- 
lines, would under the PRG RAC 
either appear on the "list” and there- 
fore be exempt from the guidelines, or 
if for some reason it did not appear on 
the list, the containment level would 
actually in some cases be raised. This 
was not the intent of the RAC. There- 
fore, I proposed to the RAC, and they 
accepted at their April 27-28, 1978. 
meeting, that language be reinserted 
in the the PRG-NIH covering prokar- 
yotes that exchange genetic informa- 
tion with E. coli but which do not 
appear on the list. This section in the 
PRG-NIH (in-A-l-b-U)) reads: 
Prokaryotes That Exchange Genetic Infor- 
mation [351 with E. coli. It is expected that 
many of the prokaryotes that exchange ge- 
netic information with E. coli by known 
physiological processes will be exempted 
from these guidelines by appearing on the 
"list of exchangers” (see sec. I E-4). 
For those not on the list, the containment 
levels are PI physical containment - an 
EK1 host-vector. In fact, experiments in 
this category can be performed with E. coli 
K-12 vectors exhibiting a lesser contain- 
ment (e.g.. conjugative plasmids) than EK1 
vectors. However, for prokaryotes that are 
classified 111 as Class 2 the containment 
levels are P2-EK1. 
For prokaryotes that do not ex- 
change genetic information with E. 
coli, the PRG-RAC proposed that 
P1-EK2 or P2-EK1 conditions apply 
only in cases of extensive characteriza- 
tion and RAC approval. "Experiments 
with DNA's from bacteria that are not 
extensively characterized require P2 
physical containment + an EK2 host- 
vector or P3-EK1. Experiments with 
DNA's from pathogenic species (class 2 
and plant pathogens, see App. B) must 
use P3-EK2.” A number of commen- 
tators objected to two different as- 
pects of this subsection of the PRG- 
RAC: (1) Many felt that experiments 
involving nonpathogenic prokaryotes 
should be conducted at P1-EK2 or 
P2-EK1 without extensive character- 
ization or RAC approval; (2) It was 
argued that plant pathogens should 
not be included with CDC class 2 
agents as requiring P3 -EK2 contain- 
ment. Both of these comments were 
referred to the RAC at their April 27- 
28. 1978, meeting and they agreed with 
the commentators. Therefore, this 
Section of the PRG NIH (III-A-l-b- 
(2): reads: 
(2' Prokaryotes that Do Not Exchange Ge- 
netic Information with E. coli P2 physical 
containment + an EK1 host-vector, or 
PI - EK2. except for DNA from class 2 
agents. Cl) which require P3 + EK2. 
The EMBO Standing Advisory Com- 
mittee on Recombinant DNA Re- 
search recommends that the contain- 
ment level for all novel non pathogen- 
ic prokaryotic DNA into E. coli K-12 
be P1 + EK1. It is my opinion that it is 
prudent to retain the levels of 
P2 + EK1 or PI ->-EK2 for nonpatho- 
genic prokaryotes that do not ex- 
change gentic information with E. 
coli 
The PRG RAC received substantial 
criticisms for identifying all agents 
classified as class 2 in the CDC’s publi- 
cation "Classification of Etiologic 
Agents on the Basis of Hazard” 
(Fourth edition. July 1974) as being 
pathogenic for the purpose of assign- 
ing containment levels. Many com- 
mentators stated that many of the or- 
ganisms so classified were harmless 
and others were of such low pathogen- 
icity that severe safety precautions 
were unwarranted. It was also pointed 
out that the pathogenicity of an intact 
micro-organism and the conjectural 
hazard of a piece of DNA from such 
an organism within E. coli K-12 were 
quite different matters. It should be 
noted that the difficulties in applica- 
tion of the CDC classification for the 
purposes of these guidelines was recog- 
nized in the original guidelines. For 
example, all species of Salmonella are 
classified as class 2 organisms by CDC. 
The original guidelines, however, dis- 
tinguish between the pathogenicity of 
S. typhimurium and S. typhi for the 
assignment of containment levels. I 
have therefore accepted the sugges- 
tion of these commentators and have 
added footnote 1 to the PRG-NIH. 
This gives NIH the authority, upon 
the recommendation of the RAC, to 
designate certain agents which are 
listed as class 2 by CDC as class 1 
agents for the purpose of these guide- 
lines. 
Section III-A-2- a. DNA from viruses of 
eukaryotes into E. coli K-12 
Discussed earlier within part III of 
this document under the heading "Re- 
combinant DNA Experiments Involv- 
ing Viral DNA” was the history of the 
"Ascot” workshop report (App. E to 
the accompanying environmental 
impact assessment) and the report of 
the working group which met on April 
6-7 1978 (App. F to the accompanying 
environmental impact assessment). 
Section III-A-2 of the PRG-NIH 
adopts the recommendations of the 
working group with minor modifica- 
FEDERAl REGISTER, VOL 43, NO. 146— FRIDAY, JULY 28, 1978 
[ 22 ] 
