620 
Gorbacli 
tlier debilitated K.12 strain, ^]776, failed to im- 
plant even in the germ-free mouse. This has been 
ascribed to the sensitivity of this strain to bile 
salts that are present in the intestinal lumen. 
A number of variables are known to influence 
the colonization of organisms in the intestinal 
tract. Implantation can be altered by antibiotic 
administration, starvation, the t\pe of diet, re- 
duction in gastric acid, and antimotility drugs. It 
is clear that more implantation experiments need 
to be performed with attention to these vari- 
ables, many of which may be found in laboratory 
workers exposed to these organisms. It is impor- 
tant to note that once £. coli K.12 has been estab- 
lished in the germ-free animal, the addition of 
wild-type strains to the flora fails to dislodge it. 
This finding may have implications for feeding 
experiments, particularly those involving sub- 
jects receiving antibiotics to which the specific 
K12 may be resistant. 
Attempts to Augment Virulence in £. coli K12 
A number of virulence genes have been iden- 
tified in wild-type strains of £. coli that are as- 
sociated with either enteric or nonenteric dis- 
ease. Using a model of intestinal infection, Smith 
transferred the plasmids for K99 and the entero- 
toxin into E. coli K12 [44]. These strains were 
fed to four colostrum-deprived calves, animals 
that are highly susceptible to the wild-type toxi- 
genic strains. One animal developed slight diar- 
rhea, but all of the calves remained well. When 
they were sacrificed, at 30 hr there was no pro- 
liferation of the K12 strains in the small or large 
intestine. The fully virulent, wild-type strain 
produced severe diarrhea in these animals and 
proliferated in large numbers within the small 
intestine. 
Minshew et al. transferred two virulence 
properties, hemolysin and colicin V, into E. coli 
K12 strains [33]. When tested in their 13-day-old 
chick embryo model, the recipient K12 strains 
still failed to show virulence. Smith performed 
a similar experiment but used a different animal 
test system, inoculation into chickens [44]. The 
K12 strain that had received ColV plasmids from 
wild-type strains of E. coli did demonstrate in- 
creased lethality in chickens. However, the dose 
required to produce death with the K12 strain 
containing ColV was very large, as compared with 
a relatively small dose needed for the wild-type 
ColV strains. 
Several attempts have been made by Formal 
and his coworkers to transfer genetic material 
from virulent shigella strains into E. coli K12 
[47]. The primary focus of these studies was 
to produce vaccines with the K12 strain that 
could colonize the bowel and possess enough shi- 
gella antigen to evoke a local immune response. 
Despite transfer of all of the recogni/ed virulence 
genes identified with Shigella into E. coli K12, 
there has been no success in producing a recipi- 
ent strain that could colonize the bowel or pene- 
trate the intestinal mucosa. When these strains 
were administered to volunteers, there were no 
clinical signs of diarrhea. The recipient K12 
strains were eliminated within six days or less, 
results similar to those reported by Anderson 
with other E. coli feeding experiments. Similar 
studies have been performed with Salmonella. 
Formal has concluded that the invasive proper- 
ty is determined by a multiplicity of genes; he 
has stated, “We consider it unlikely that the ran- 
dom insertion of foreign DNA into the E. coli K12 
genome coidd supply all of the genetic informa- 
tion necessary to convert this organism into an 
invasive enteric pathogen" [20]. 
The Lack of Epidemic Potential of £. coli K12 
On the basis of the available evidence, the parti- 
cipants in the Workshop agreed to the consensus 
statement that £. coli K12 could not be con- 
verted into an epidemic pathogen. In the first 
instance, colonization of the intestine by £. coli 
K12 has proven difficult, if not impossible, in con- 
ventional animals and in human volunteers. 
Transfer of virulence genes into £. coli K12 has 
not produced a fully pathogenic strain, although 
the ColV experiments of Smith suggest that some 
increase in pathogenic potential could be 
achieved. Finally, communicability and spread 
of this organism from person to person is con- 
sidered extremely unlikely because of its fragil- 
ity in nature and the high degree of sanitation 
and public health procedures in this country. 
Thus, the three tenets for pathogenicity — coloniz- 
ation, the capacity to produce disease, and com- 
municability — are blocked, and it is not felt 
[ 158 ] 
