appendix o 
In summary then, it is not possible to definitely exclude on theoretical 
grounds the possibility that the inadvertent introduction of one or a few 
genes into a wild type (i.e. not "enfeebled") E. coli strain may augment 
those disease-causing properties which some coli strains already possess. 
None of these diseases can, byany stretch of the imagination, be expected 
to result in worldwide epidemics. Current knowledge of the manner in which 
bacteria cause disease allows us to exclude the possibility of an inadver- 
tent creation from coli of a super-pathogen by recombinant DNA technics. 
Even if the deliberate introduction by some terrorist of a large number of 
genes into coli were to succeed in the creation of a pathogen which 
rivals in virulence the ancient scourges of mankind (and it should be 
remembered that some countries have spent large sums of money on such 
projects with little apparent success) , modern medicine may be expected 
to deal swiftly with such an emergency. 
If, for purposes of this discussion, we assume that an coli strain 
carrying a segment of cloned DNA has succeeded in implanting itself into 
the intestinal flora of man, what is the chance that this DNA can enter the 
body and subsequently enter cells of the body in intact form? (I will 
not discuss in this paper the measures which have been taken to reduce the 
possibility that the DNA might be transferred in the gut to another, more 
virulent bacterium.) It is well known that bacteria from the human mouth 
frequently enter the body and may then be found in the blood. This is 
a normal and everyday process which may be triggered by, for example, 
the simple act of chewing (22). It has been shown in man, mice, rats 
and dogs (23-25) that particles such as starch granules or bacteria from 
the intestinal tract regularly Invade the body. The number of bacteria 
involved is very small, rather crude estimates made in my laboratory sug- 
gest tjiat one in a billion bacteria present in the mouse gut may invade 
and reach the mesenteric lymph nodes during a period of several hours. 
Since the number of coli in the intestine rarely exceeds 10^ per 
gram of intestinal contents, the actual number of _E. coli cells reaching 
the body's tissues must be very small (i.e. 1 per 1000 grams). Moreover, 
when bacterial cells of species' like coli invade the healthy body, 
they are quickly phagocytosed by leukocytes. Inside the leukocytes the 
bacteria are sequestered in vacuoles, killed and subsequently broken up 
by the enzymes released into the vacuoles. Since these enzymes include 
DNA'se, it is vnlikely that intact DNA would escape into the general 
circulation of the body, even in the event that the leukocytes would kill 
the bacterium, but subsequently either egest the contents of the vacuole 
or be totally unable to lyse certain structural components of the bacterial 
cell (26). Alternately, Invading bacteria may be killed and lysed by the 
antibacterial substances in the body fluids. In this case, any DNA 
released by the disintegrating bacterium would be quickly destroyed 
by the DNA'se in these fluids, probably long before it has a chance to 
enter a cell. It is apparent therefore that, even if an coli strain 
succeeded in implanting Itself into the gut, the chance that its DNA 
would ever enter a cell of the human body in intact form, is indeed 
exceedingly small. However, on theoretical grounds (the only tool we 
have in this paper), we cannot exclude this possibility with the same 
degree of certainty as we were able to reach in discussing the possibility 
of creating super-pathogens from coli or of implanting into man an 
coli strain which cannot multiply in a germfree animal. 
Appendix O — 5 
