NOTICES 
33055 
mlttee which has certified th, facilities, 
procedures, and the training and expertise 
of the personnel Involved are adequate; 
3. An approve MUA with a certification is 
on file with the funding agency of the re- 
questing laboratory; 
4. A copy of this letter is on file with the 
requesting laboratory's Institutional Bloha- 
aards Committee. 
B. Prior to shipment or transfer of recom- 
binant DNA materials to non-Federally 
funded investigators or institutions within 
the United States, the sending laboratory 
shall obtain a letter from the requesting 
laboratory stating items 1, 2, and 4 under A 
above. 
C. Prior to international shipment of re- 
combinant DNA materials, the sending labo- 
ratory shall obtain a statement from the re- 
questing laboratory stating that research in- 
volving recombinart DNA molecules shall 
be conducted in a .-ordance with the con- 
tainment levels spe -ified by the NIH Guide- 
lines, or applicable national guidelines if 
such have been adopted by the country In 
which the research is to be conducted, and 
that the requesting laboratory shall not 
transfer the recombinant DNA material to 
other laboratories. 
D. The sending laboratory shall maintain 
a record of all shipments of recombinant 
DNA materials and shall provide NIH with 
a complete list of such shipments in the 
annual progess report for NIH grants and 
contracts. 
Mounth-pipetting at the PI level. 
Both the 1976 guidelines and the 
PRG-RAC prohibit mouth-pipetting 
at the P2, P3, and P4 levels. For the 
PI level, however, they state, “Al- 
though pipetting by mouth is permit- 
ted, it is preferable that mechanical 
pipetting devices be used. When pipet- 
ting by mouth, cotton-plugged pipettes 
shall be employed." A number of com- 
mentators have urged that mouth-pi- 
petting be prohibited at the PI level of 
physical containment. This is strongly 
endorsed by NIH safety experts, who 
point out that this is an important 
safety feature, and that efficient new 
mechanical pipetting aids should not 
greatly hamper researchers. Also, the 
EMBO Standing Advisory Committee 
on Recombinant DNA Research “be- 
lieves that mouth pipetting should be 
prohibited in the PI laboratory, as it is 
prohibited in P2-P4 laboratories.” In 
addition, the Working Group of 
American virologists which met on 
April 6-7, 1978, to review the report of 
the U.S.-EMBO Workshop to Assess 
Risks for Recombinant Experiments 
Involving the Genomes of Animal, 
Plant, and Insect Viruses 1 wrote the 
following in their report: 
In its deliberations, the Working Group 
was impressed with the safeguards afforded 
by a ban on mouth pipetting for recombin- 
’The history of the U.S.-EMBO Work- 
shop and the April 6-07, 1978, working 
group is discussed in detail in Pt. HI of this 
document under the heading "Recombinant 
DNA Experiments Involving Viral DNA” 
and the report of the working group ap- 
pears as App. E to the accompanying envi- 
ronmental impact assessment. 
ant DNA experiments involving E. coli K-12 
host-vectors. The group felt that the only 
plausible way E. coli K-12 could gain entry 
into laboratory workers was by oral inges- 
tion. The analysis contained in the U.S.- 
EMBO Report was predicated on the 
remote possibility that E. coli K-12, con- 
taining eukaryotic viral DNA would be 
swallowed and the viral DNA insert would 
be delivered to a tissue in the body which 
ordinarily would be inaccessible to the virus. 
A prohibition of mouth pipetting would 
clearly prevent this sequence of events from 
even beginning. The Working Group there- 
fore recommended that no mouth pipetting 
be allowed at any level of physical contain- 
ment (including PI) when working with E. 
coli K-12. 
On the other hand, when I request- 
ed that the RAC, at their April 27-28, 
1978, meeting reconsider whether 
mouth pipetting should not be banned 
at the PI level, it was their consensus 
that many experiments classified as 
PI need not include a ban on mouth- 
pipetting, and that therefore PI in 
general should not be redefined. In- 
stead, they recommended that only 
certain classes of PI experiments be 
designated as requiring no mouth-pi- 
petting. 
In resolving this issue, I have decid- 
ed to adopt the conservative position 
and ban mouth-pipetting. Accordingly, 
language has been inserted in the 
PRG-NIH saying that at the PI level, 
"Mechanical pipetting devices shall be 
used; pipetting by mouth is prohibit- 
ed.” Since mouth-pipetting had al- 
ready been banned at the P2-P-4 
levels, this means that it is now 
banned for all experiments covered by 
these guidelines. 
BIOLOGICAL CONTAINMENT 
Review of RAC-proposed guidelines 
Experiments on recombinant DNA’s 
by their very nature lend themselves 
to applications of highly specific bio- 
logical barriers as a means of contain- 
ment. In fact, there are natural' bar- 
riers that limit either the infectivity of 
a vector or vehicle (plasmid or virus) 
to specific hosts, or its dissemination 
and survival in the environment. Both 
the vectors whereby DNA is trans- 
ferred to the recipient host and the 
host cells wherein it replicates can be 
designed genetically to decrease by 
many orders of magnitude the prob- 
ability of dissemination of recom- 
binant DNA outside the laboratory. 
The proposed revised guidelines de- 
scribe the categories of hosts and vec- 
tors to be used in minimizing the 
spread of organisms containing recom- 
binant DNA The PRG-RAC differs in 
some respects from the 1976 guidelines 
as a result of certain changes in defini- 
tions of HV systems and in the re- 
quirements at specific HV levels (nota- 
bly HV3). A new section has been 
added on certification of host-vector 
systems. 
Definitions of host-vector systems. A 
new nomenclature— HV1, HV2, and 
HV3— has been developed to incorpo- 
rate a variety of hosts and vectors into 
the framework initially established for 
E. coli K-12. In particular, the PRG- 
RAC provides criteria for HV1 systems 
other than E. coli K-12. In the 1976 
guidelines, cloning systems other than 
E. coli K-12 were to be considered 
only if superior to E. coli K-12 in con- 
tainment properties; but it is now rec- 
ognized that many useful experiments 
can only be conducted using HV sys- 
tems other than those based on E. coli 
K-12, and that such experiments 
should be permitted so long as the 
proposed HV system provides equiva- 
lent biological containment. The new 
HV1 criteria provide a structure for 
approval of systems that meet these 
requirements. 5 
HV2 systems. AT the HV2 level of 
containment, there are no substantive 
changes comparing the 1976 guidelines 
with the PRG RAC. However, the 
RAC, On June 23, 1977- the same day 
it approved the PRG-RAC— also 
adopted unanimously “Instructions to 
Investigators Concerning Data To Be 
Submitted on Host-Plasmid Systems 
Proposed for EK2 Certification.” Al- 
though not officially part of the PRG- 
RAC, these instructions set forth cri- 
teria that any putative EK2 host- 
vector systems must meet before rec- 
ommendation by the RAC for certifi- 
cation. The RAC applied these criteria 
in reviewing new systems (pBR322 and 
PBR313 in X1776) at the June 23, 1977, 
meeting, and will do so for all future 
submissions. It was made clear at the 
meeting that these criteria are defi- 
nitely more stringent than previous 
ones, and this greater stringency 
means that EK2 host-vector systems 
approved now and to be approved in 
the future are even safer than those 
approved previously. 
Requirements for HV3 systems. 
These have been made more stringent 
in the PRG-RAC than the corre- 
sponding requirements for EK3 in the 
1976 guidelines. The PRG-RAC re- 
quires that the vector be dependent on 
its propagation host or be highly de- 
fective in mobilizability. “Reversion to 
host-independence must be less than 
Vio* per vector genome per genera- 
tion.” Also, the vector may carry no 
resistance to antibiotics used clinically 
or in agriculture. The provision that 
antibiotic resistance markers of medi- 
cal or agricultural importance are not 
to be used in the vector should pre- 
vent any inadvertent advantage for re- 
combinant DNA-bearing vectors that 
encounter antibiotics in the environ- 
ment. 
5 Under the proposed revisions, HVls 
other than E. coli K-12 need not offer a dis- 
tinct advantage over E. coli K-12 host-vec- 
tors, need not be capable of modification to 
HV2 and HV3, and need not be class I etlolo- 
gic agents. 
FEDERAL REGISTER, VOL 43, NO. 146— FRIDAY, JULY 28, 1978 
[ 16 ] 
