33124 
NOTICES 
either directly or through transfer of 
a foreign DNA fragment to another 
bacterial cell. These properties will 
depend on the ability of the K-12 to 
survive, multiply, and infect other 
living organisms. As already described, 
K-12 is poorly equipped to survive in 
natural environments; but if it should 
survive and multiply, it is still unlikely 
to infect living things. E. coli are 
seldom spread by aerosols; they are 
primarily spread by ingestion of con- 
taminated food and water. Between 
10* and 10* (1 million to 1 billion) cells 
of pathogenic E. coli are required to 
cause disease <12. 25). In other words, 
at least a million bacteria would be re- 
quired to cause disease in a single 
person If some K-12 did become 
pathogenic. 
The guidelines emphasize protection 
of laboratory workers because they 
are the persons most at risk. They are 
also the most likely means by which 
recombinant DNA might be spread. 
Should a worker carry such agents out 
of the laboratory, however, the prob- 
ability that others would be affected is 
still very low, and the risk of a result- 
ing epidemic is virtually nonexistent. 
There is abundant evidence for this as- 
sertion. It has long been known that 
the separation of sewage from food 
and water supplies prevents epidemcis 
of enteric bacteria such as E. coli. 
The following excerpt from a letter 
by Roy Curtiss III to Donald S., Fre- 
drickson discusses the K-12 strain of 
E. coli in relation to infectivity.(fj) 
In terms of communicability of E. coli K- 
12. we know that enteric diseases caused by 
enteropathosenic E. colt and various strains 
of Shigella. Salmonella and Vibrio are trans- 
mitted by contaminated food and water and 
that manifestation of disease symptoms re- 
quires consumption of approximately 1 mil- 
lion bacteria. Such enteric diseases are 
seldom spread by aerosols. Indeed. It Is well 
known, for example, that cages of mice In- 
fected with Samonella can be housed in the 
same room with uninfected mice which 
remain uninfected. The finding that E. coli 
cells can be recovered from the nasophar- 
ynx of approximately 5 percent of those 
humans tested might suggest that aerosol 
spread could occur. Such E. coli cells, how- 
ever. are only intermittently present In the 
nasopharynx and are usually found at con- 
centrations too low to Initiate an Infection 
even If they were representative of a patho- 
genic strain. They most likely get Into the 
nasopharynx due to poor personal hygiene. 
After learning of these observations quite 
some years ago. I monitored my nostrils and 
skin for the presence of those E. coli K 12 
strains I was working with. I was successful 
In detecting these strains about 10 percent 
of the time when the monitoring was done 
at the end of the work day. but never ob- 
tained positive results when the monitoring 
was done the next morning. I should hasten 
to add that my research with E. coli K-12 at 
that time Involved mouth pipetting and 
other aerosol-generating procedures on an 
open lab bench: procedures and conditions 
which are not permitted by the NIH Guide- 
lines. These results, preliminary as they are. 
nevertheless suggest that E. coli K-12 does 
not colonize the nasopharynx. Based on 
these observations, the fact that E. colt's 
normal ecological niche is the colon and the 
fact that transmission of enteric diseases is 
by ingestion of contaminated water and 
food. I doubt that E. coli K-12 could be con- 
verted to an air borne “infectious" agent by 
Introduction of recombinant DNA In terms 
of the more usual means for spread of en- 
teric pathogens, it Is evident tha enteric dis- 
eases are bery well controlled In the United 
States by sanitary engineering, even though 
there have been reports of poor water qual- 
ity In some parts of the country and higher- 
than-desired levels of pollution of rivers, 
streams, etc. There is however, a concerted 
effort to Improve biological waste water 
treatment and thus lessen pollution and im- 
prove water quality. Even If there were a 
natural catastrophe such as caused by an 
earthquake, tornado, hurricane, etc., it is 
unlikely that E. coli K-12 containing recom- 
binant DNA could initiate or sustain an epi- 
demic in view of K-12's inability to colonize 
and overcome host defense mechanisms. 
Seeking a consensus on the matter 
of risk assessment in recombinant 
DNA research, with particular refer- 
ence to the use of E. coli. the National 
Institutes of Health sponsored a work- 
shop in Falmouth. Mass., on June 20- 
21, 1977. In attendance were approxi- 
mately 50 invited participants and ob- 
servers. from the United States and 
abroad, including experts on all as- 
pects of infectious disease. The follow- 
ing excerpt from a letter by the work- 
shop chariman. Sherwood L. Gorbach. 
to Donald S. Fredrickson summarizes 
the principal conclusion: 
Consensus Agreement 
An Important consensus was arrived at by 
the assembled group which I felt was of suf- 
ficient interest to be brought directly to 
your attention. The participants arrived at 
unanimous agreement that E. coli K-12 
cannot be converted into an epidemic patho- 
gen by laboratory manipulations with DNA 
Inserts. On the basis of extensive studies al- 
ready completed. It appears that E. coli K- 
12 does not Implant In the intestinal tract of 
man. There is no evidence that non-trans- 
missible plasmids can be spread from E. coli 
K-12 to other host bacteria within the gut. 
Finally, extensive studies in the laboratory 
to Induce virulence In E. coli K-12 by Inser- 
tion of known plasmids and chromosomal 
segments coding for virulence factors, using 
standard bacterial genetic techniques, have 
proven unsuccessful in producing a fully 
pathogenic strain. As a result of these dis- 
cussions. it was believed that the proposed 
hazards concerning E. coli K-12 as an epi- 
demic pathogen have been overstated. Such 
concerns are not compatible with the exten- 
sive scientific evidence that has already 
been accumulated, all of which provides as- 
surance that E. coli K 12 Is Inherently en- 
feebled and not capable of pathogenic trans- 
formation by DNA insertions. 
The entire letter from Gorbach is 
quoted in the NIH environmental 
impact statement, part II. appendix 
M.(27) The proceedings of the Fal- 
mouth workshop on risk-assessment 
have been published in the May 1978 
Journal of Infectious Diseases. 
There remains the question whether 
the insertion of a foreign DNA frag- 
ment into K-12 will significantly alter 
the properties of the latter with 
regard to survival and multiplication, 
or the ability of the plasmid and bac- 
teriophage vectors to be spread. The 
improbability of converting K-12 to a 
pathogen has already been discussed. 
Changes in ability to survive and mul- 
tiply would be expected to involve not 
only the changes in the K-12 itself, or 
the plasmid or bacteriophage, but also 
the nature of the environment in 
which it finds itself. The subject is dis- 
cussed in the section of this document 
entitled ’Risk and Benefits of Recom- 
binant DNA Research." 
Attenuated K-12 Systems. Theoreti- 
cally. the most desirable bacterial re- 
cipient of recombinant DNA would be 
a species uniquely adapted to carefully 
controlled laboratory conditions and 
unable to survive or transmit DNA to 
other organisms in any natural envi- 
ronment. This means that it should be 
unable to establish itself as a long- 
lived and multiplying resident in or on 
living things, or in soil or water. In ad- 
dition. these properties should not be 
significantly altered by recombination 
of the bacterium's DNA. The organism 
should also, of course, lend itself to 
manipulation for successful execution 
of experiments. 
No bacterium meeting all these re- 
quirements is known. It is possible 
that no such creature exists in nature. 
Available bacterial systems must be 
evaluated for relative safety and util- 
ity. depending on the extent to which 
they approach the ideal. The forego- 
ing summary of knowledge concerning 
K-12 and its known plasmids and bac- 
teriophages indicates that these sys- 
tems measure up well with the ideal 
criteria, and can therefore be recom- 
mended for use in recombinant DNA 
research. 
The K-12 systems, extant and pro- 
jected. are known as EK1, EK2. and 
EK3, referring to increasing degrees of 
attenuation. The guidelines permit the 
use of EK1 for those experiments 
whose potential for hazard is regarded 
as nil. low, or minimal. For experi- 
ments Judged to have a somewhat 
higher (though still conjectural) po- 
tential for hazard, the guidelines re- 
quire the further attenuated system 
EK2. Here, properties of the K-12 and 
the vectors must be so modified as to 
minimize the chance of the vector sur- 
viving in its host outside the labora- 
tory and migrating to other hosts. 
EK3 systems are even stricter, requir- 
ing. for example, the use of vectors 
that cannot propagate outside the 
host. So far, no EK3 systems have 
been certified. [In the proposed re- 
vised Guidelines (PRG-NIH), the EK 
FEDERAL RECISTER, VOL 43. NO. 146— FRIDAY, JULY 28, 1978 
[ 85 ] 
