69244 
Federal Register / Vol. 44, No. 232 / Friday, November 30, 1979 / Notices 
Exempt All Experiments in E. coli K-12 
From the Guidelines 
Some called for the complete 
exemption from the NIH Guidelines of 
all recombinant DNA experiments using 
E. coli K-12 as the host. 
The recommendation of the RAC was 
not, in fact, for complete exemption of 
such experiments. Three important 
safety features for these experiments 
that will not be exempt, but will 
according to the proposed decision form 
a special class under the Guidelines, 
are: 
1. Pi Containment — Including the ban 
on mouth pipetting and the requirement 
that all biological wastes shall be 
decontaminated. Proper employment of 
Pi conditions eliminates the primary 
means of E. coli escape from the 
laboratory. 
2. EKl — Allowing only E. coli K-12 
strains and not allowing the use of 
conjugation proficient plasmids or 
generalized transducing phages. This 
greatly reduces the probability that any 
escaping E. coli K-12 would survive and 
transfer their recombinant DNA to other 
organisms. 
3. IBC Oversight — Continuing local 
surveillance and registration of these 
experiments. 
In addition, keeping these 
experiments under the Guidelines rather 
than exempting them means that any 
scale-up of the experiments beyond 10 
liters will require prior NIH approval. 
Treat Experiments Equally in which 
There Is or Is Not a Deliberate Attempt 
To Achieve Gene Expression 
Some argued that recombinant DNA 
experiments should be treated no 
differently whether there is, or is not, a 
deliberate attempt to achieve gene 
expression. 
In my Decision Document of July 1978, 
I wrote: 
Although clearly the time has come to 
revise the original NIH Guidelines for 
Recombinant DNA Research, it is not the 
time to conclude that they are being altered 
in preparation for their early abandonment. 
Understanding of gene regulation and 
expression is increasing inexorably and at an 
awesome pace. We may predict that ways 
will be found to achieve and control the 
translation of foreign genes by a variety of 
hosts. As the barriers to translation are 
dropped, some of the larger promise of 
recombinant technology will be realized. In 
some proportion to the harvest of positive 
results, a capability must be maintained for 
observing any capacity of these experiments 
to yield harmful products, and for 
communicating this to all who have an 
interest in similar experiments. 
It is now one year later, and more 
ways to assure gene expression are 
being found. 
Some of the recent papers reporting 
on gene expression in E. coli of inserted 
recombinant DNA include: 
L. Villa-Komaroff et al., A Bacterial 
Clone Synthesizing Proinsulin, Proc. 
Natl. Acad. Sci. USA 75, 3727, 1978. 
T. H. Fraser and B. J. Bruce, Chicken 
Ovalbumin Is Snythesized and Excreted 
by Escherichia coli, Proc. Natl. Acad. 
Sci. USA 75, 5936, 1978. 
D. V. Goeddel et al. Expression in 
Escherichia coli of Chemically 
Synthesized Genes for Human Insulin, 
Proc. Natl. Acad. Sci. USA 76, 106, 1979. 
T. M. Roberts et al., A General 
Method for Maximizing the Expression 
of a Cloned Gene, Proc. Natl. Acad. Sci. 
USA 76, 760, 1979. 
C. J. Burrell et al., Expression in 
Escherichia coli of Hepatitis B Virus 
DNA Sequences Cloned in Plasmid 
pBR322, Nature 279, 43, 1979. 
D. V. Goeddel et al., Direct Expression 
in Escherichia coli of a DNA Sequence 
Coding for Human Growth Hormone, 
Nature 281, 544, 1979. 
The RAC, at their September 6-7 
meeting in discussing the “E. coli K-12/ 
Pi Recommendation,” grappled with the 
issue of “biologically active 
polypeptides” produced by E. coli K-12 
carrying recombinant DNA. As 
discussed in Section HI— B above, there 
is only a remote possibility that E. coli 
K-12 carrying recombinant DNA would 
escape, survive, compete, and implant in 
an environmental niche, etc. Were such 
an unlikely event to occur, however, an 
E. coli K-12 deliberately programmed to 
achieve eukaryotic gene expression 
could conceivably be a greater hazard 
than an E. coli K-12 not so endowed. 
Therefore, experiments in which there is 
a deliberate attempt to achieve gene 
expression continue to merit special 
attention. For these reasons, as 
discussed below in Section HI— E, my 
proposed decision is to place a special 
requirement on any experiment in which 
there is a deliberate attempt to have the 
E. coli K-12 efficiently express a gene 
coding for a eukaryotic protein. For such 
an experiment, prior review and 
approval by the IBC would be required. 
This will allow the IBC to judge whether 
it wishes to require any added 
restrictions to be placed on the 
experiment, and to remain fully 
informed of its progress. 
Include Ff Bacteriophages (Filamentous 
Single Strand Male Specific 
Bacteriophages Such As M13 and fd) 
With Lambda or Lambdoid 
Bacteriophages To Be Permissible 
Under the "E coli K-12/P1 
Recommendation " 
The 1978 Guidelines define EKl as 
follows: “The host is always E. coli K-12 
or a derivative thereof, and the vectors 
include nonconjugative plasmids (e.g., 
pSClOl, ColEl, or derivatives thereof) 
and variants of bacteriphage, such as 
lambda. The E. coli K-12 hosts shall not 
contain conjugation-proficient plasmids, 
whether autonomous or integrated, or 
generalized transducing phages.” 
A memorandum of December 26, 1978, 
from the NIH Office of Recombinant 
DNA Activities to Institutional Biosafety 
Committees said, “The M13, fd, and 
other related single-strand 
bacteriophages * * * in conjunction 
with E. coli K-12 strains that do not 
contain conjugation-proficient plasmids 
(i.e., F") are acceptable for experiments 
that require EKl biological containment. 
Since the host strains are not the natural 
hosts, the means of infection would 
involve transfection.” 
At the RAC meeting of February 15- 
16, 1979, a motion passed by a vote of 20 
to 1, with 1 abstention, that 
"Conjugation-deficient mutants, such as 
the fcraD and tral mutants of the F factor 
may be used with the Ff bacteriophages 
if they have been shown to exhibit low 
levels of transfer (of the order of 10'5 or 
less) and also have low reversion rates 
(such as found for deletion or double- 
mutants).” This recommendation was 
accepted by NIH and transmitted by 
ORDA to IBCs on April 23. 
(At the RAC meeting of May 21-23, 
1979, a proposal to allow the use of 
bacteriophage vectors in E. coli K-12 
hosts containing conjugation-proficient 
plasmids was rejected by a vote of 10 to 
4, with 5 abstentions.) 
Thus today EKl experiments are being 
conducted under the NIH Guidelines 
using Ff bacteriophages. 
The “E. coli K-12/P1 
Recommendation” as passed by the 
RAC on September 6 includes die words 
"lambda or lambdoid bacteriophages,” 
without mention of Ff bacteriophages. 
One alternative which I considerd was 
to add the words “or Ff ’ before 
"bacteriophages.” However, in the 
absence of a specific recommendation 
from the RAC on this point, and because 
Ff bacteriophages have a broader host 
range than lambda or lambdoid 
bacteriophages, I am not adopting this 
alternative. 
What shall be the status then of 
ongoing EKl experiments involving Ff 
bacteriophages when the new 
Guidelines are promulgated? At the time 
of impending transition from the 1976 to 
the 1978 Guidelines, a memorandum 
was sent from ORDA to IBCs and 
Principal Investigators (‘Transition to 
Revised Guidelines for Recombinant 
DNA Research”) which allowed certain 
on-going experiments to continue at the 
[ 2 30] 
