- 2 - 
It is of the utmost importance that the Guidelines discuss explicity 
the dangers of introducing foreign sequences into strains other than K12, 
for example into known epidemic strains of coli. Otherwise an environment 
will be created in which wild strains will be substituted for disarmed 
strains. No one present at Falmouth, for example, maintained that the 
introduction of mammalian DNA sequences linked to antibiotic resistant 
plasmids, into wild strains of IS. coli would not be a serious potential 
public health threat. Since there are presently no laws, regulations, or 
statues prohibiting such experiments, the RAC has primary responsibility 
for presenting this clearly and cogently. It has so far failed in this 
responsibility. 
Elaboration: 
A) The documents on numerous occasions quote Dr. Gorbach's personal 
letter to the effect that K12 is unlikely to be converted to a 
pathogen. There is no mention of the more serious concern, namely 
the transfer of foreign DNA to strains of IS. coli which are already 
pathogens, and some of which are epidemic pathogens. In this case 
the concern is the conferring of additional deleterious properties 
on a strain that is already a problem. I present below data on 
the existing public health problem from IS. coli infections. The 
absence of such material in the PRG leads me to doubt the credibility 
of the revised guidelines. 
1) In the United States, strains of IS. coli are the number one cause 
of community acquired infections which are serious enough to require 
hospitalization (Gangerosa, 1978; National Nosocomial Infections Study). 
2) IS. coli strains are the number one cause of hospital-acquired 
(nosocomial) infections in the United States. These infections are rarely 
gastroenteritis, but more often acute infections associated with the urinary 
tract, surgical wounds, pneumonia, and bacteremia (bloodstream infection). 
Stamm, Martin and Bennet (1978) have recently reviewed the incidence and 
character of gram negative nosocomial infections. To give a sense of the 
magnitude of the problem, I calculate from their data the incidence of some 
IS. coli infections; 
per year/U.S. 
90.000 JS. coli surgical wound infections, with 2,700 associated deaths. 
40.000 IS. coli pneumonia infections, with 10,000 associated deaths. 
17.000 JS. coli bloodstream infections, with 4,000 associated deaths. 
Of course all of these patients were compromised in some way or another, 
for example by being subject to surgery, but then it is just such individuals 
who are the normal target of bacterial infections, and are the relevant 
population of concern. 
[A-296] 
