Federal Register / Vol. 46, No. 233 / Friday, December 4, 1981 / Notices 
59409 
pla-smids or phages (See Section III-A- 
1-b.) 
III-A-3. Lowering of Containment 
Levels for Characterized or Purified 
DNA Preparations and Clones. Many of 
the risks which might conceivably arise 
from some types of recombinant DNA 
experiments, particularly shotgun 
experiments, would result from the 
inadvertent cloning of a harmful 
sequence. Therefore, in cases where the 
risk of inadvertently cloning the 
“wrong” DNA is reduced by prior 
enrichment for the desired piece, or in 
which a clone made from a random 
assortment of DNAs has been purified 
and the absence of harmful sequences 
established, the containment conditions 
for further work may be reduced. The 
following section outlines the 
mechanisms for such reductions. 
UI-A-3-a. Purified DNA Other than 
Plasmids, Bacteriophages, and Other 
Viruses. The formation of DNA 
recombinants from cellular DNAs that 
have been purified[41] and in which the 
absence of harmful sequences has been 
established[3] can be carried out under 
lower containment conditions than used 
for the corresponding shotgun 
experiment.[42] The containment may 
be decreased one step in physical 
containment (P4 P3; P3 P2; P2 PI) while 
maintaining the biological containment 
specified for the shotgun experiment, or 
one step in biological containment (HV3 
HV2; HV2 HVl) while maintaining the 
specified physical containment. The 
Institutional Biosafety Committee (IBC) 
must review such a reduction and the 
approval of the IBC and of the NIH must 
be secured before such a reduction may 
be put into effect. IBC approval is 
sufficient for such a reduction except for 
any lowering of containment under 
Section III-A-3-a to levels below 
Pl+HVl, which requires prior NIH 
approval. (See Section IV-E-l-b-(3)- 
(e).) 
III-A-3-b. Characterized Clones of 
DNA Recombinants. When a cloned 
DNA recombinant has been rigorously 
characterized and the absence of 
harmful sequences has been 
established[3], experiments involving 
this recombinant DNA may be carried 
out xmder lower containment conditions. 
Institutional Biosafety Committees 
(IBCs) may give approval for a single- 
step reduction in physical or biological 
containment on receipt of evidence of 
characterization of a clone derived from 
a shotgun experiment and its probable 
freedom from harmful genes. IBC 
approval is sufficient for such a 
reduction except for any lowering of 
containment under Section III-A-3-b to 
levels below Pi +HV1, or reduction of 
containment levels by more than one 
step, which also requires prior NIH 
approval.(See Section IV-E-l-b-3-(e).j 
III-B. Experiments with Prokaryotic 
Host-Vectors Other Than E. coliK-12. 
III-B-1. HVl and HV2 Systems. 
Certain certified HVl and HV2 
hostvector systems appear in Appendix 
D. The containment levels for these 
systems are given in the subsections of 
Section III-A. Other systems in the 
future may be certified as HVl and HV2. 
At the time of certification, the 
classification of containment levels for 
experiments using them will be assigned 
by NIH. 
III-B-2. Return of DNA Segments to 
Prokaryotic Non-HVl Host of Origin, 
Certain experiments involving these 
prokaryotes that exchange genetic 
information with E. coli by known 
physiological processes will be exempt 
from these Guidelines if they appear on 
the "list of exchangers” set forth in 
Appendix A (see Section I-E-4). For a 
prokaryote which can exchange genetic 
information[35] with E. coli under 
laboratory conditions but which is not 
on the list (Host A), the following type of 
experiment may be carried out under PI 
conditions without Host A having been 
approved as an HVl host: DNA from 
Host A may be inserted into a vector 
and propagated in E. coli K-12. 
Subsequently, this recombinant DNA 
may be returned to Host A by 
mobilization, transformation, or 
transduction and may then be 
propagated in Host A in any desired 
vector under Pi conditions. 
For a prokaryote which does not 
exchange genetic information with E. 
coli (Host B), the following type of 
experiment may be carried out without 
Host B having been approved as an HVl 
host; DNA from Host B may be inserted 
into a vector and propagated in E. coli 
K-12. Subsequently, this recombinant 
DNA may be returned to Host B and 
propagated in Host B under Pi 
conditions.(43] 
III-B-3. Non-HVl Systems. 
Containment levels for other classes of 
experiments involving non-HVl systems 
may be approved by the Director, NIH. 
(See Sections IV-E-l-b-(l)-(b), FV-E-l- 
b-(2)-(c), and IV-E-l-tH3)-(b).) 
III-C. Experiments with Eukaryotic 
Host-Vectors. 
III-C-1. Vertebrate Host- Vector 
Systems.[44] The subsections of 
Sections Ill-C-l-a, -b, -c and -d involve 
the use of specific viral vectors, namely 
polyoma, SV40, human adenoviruses 2 
and 5, and mouse adenovirus strain FL, 
respectively. The subsections of Section 
Ill-C-l-e involve the use of viral vectors 
including the specific viral vectors 
considered in the subsections of 
Sections Ill-C-l-a, -b, -c and -d, as 
well as any other viral vector. When the 
reader finds that the containment level 
given for a specific experiment in a 
subsection of Section Ill-C-l-e is 
different from the containment level 
given in a subsection of Section III-C-1- 
a, -b, -c or -d, he may choose which of 
the two containment levels he wishes to 
use for the experiment. 
Ill-C-l-a. Polyoma Virus. 
Ill-C-l-a-(l). Productive Virus-Cell 
Interactions. 
III-C-l-a-(l)-(a). Defective or whole 
polyoma virus genomes, with 
appropriate helper, if necessary, can be 
used in P2 conditions to propagate DNA 
sequences: 
III-C-l-a-(l)-(a)-(i). from bacteria of 
Class 1 or Class 2(1] or their phages or 
plasmids, except for those that produce 
potent polypeptide toxins;[34] 
III-C-l-a-(l)-(a)-(2). from mice; 
III-C-l-a-{l)-(a)-{5). from eukaryotic 
organisms that do not produce potent 
polypeptide toxins, [34] provided that the 
DNA segment is >99% pure. 
III-C-l-a-(l)-(b). Defective polyoma 
genomes, with appropriate helper, if 
necessary, can be used in P2 conditions 
for shotgun experiments to propagate 
DNA sequences from eukaryotic 
organisms that do not produce potent 
polypeptide toxins.[34] 
III-C-l-a-(l)-(c). Whole virus 
genomes with appropriate helper, if 
necessary, can be used in P3 conditions 
for shotgun experiments to propagate 
DNA sequences from eukaryotic 
organisms that do not produce potent 
polypeptide toxins.[34] 
III-C-l-a-(l)-(d). Experiments 
involving the use of defective polyoma 
virus genomes to propagate DNA 
sequences from eukaryotic viruses will 
be evaluated by NIH on a case-by-case 
basis[45] and will be conducted under 
the prescribed physical and biological 
containment conditions. (See Section 
IV-E-l-b-(3)-(c).) 
ni-C-l-a-(2). Nonproductive Virus- 
Cell Interactions. Defective or whole 
polyoma virus genomes can be used as 
vectors in P2 conditions when 
production of viral particles cannot 
occur (e.g., transformation of 
nonpermissive cells or propagation of an 
unconditionally defective recombinant 
genome in the absence of helper): 
Provided, The inserted DNA sequences 
are not derived from eukaryotic viruses. 
In the latter case, such experiments will 
be evaluated by NIH on a case-by-case 
basis [45] and will be conducted under 
the prescribed physical and biological 
containment conditions. (See Section 
IV-E-l-b-(3)-(c).) 
[ 298 ] 
