85 
backup system. I think maybe when we have real data on what the survival 
and transmission is in vivo of these particular strains, maybe we would not 
consider that as necessary. 
DR. GINSBERG: I guess what I am really saying is I would be happier if 
you would have the data before you would approve an HV3 system, and then 
permit antibiotic markers in it. 
DR. GOTTESMAN: We have not had, for instance, any HV3 systems sub- 
mitted yet. I think if we had data on HV3 systems before us, and in vivo 
testing, maybe we wouldn't have put that in. But at the moment we wanted to 
make it the best possible, and two backups didn't seem like-- It is achiev- 
able, that is basically it. 
DR. FREDRICKSON: Dr. Sinsheimer. 
DR. SINSHEIMER: I just wondered if you have some experience now which 
would permit you to comment either on the stability of the EK2 strains, 
whether they are likely to mutate to less stringent, and secondly, on any 
laboratory experience on the frequency with which cultures of such crippled 
strains may become contaminated or overgrown, or things of this kind. 
DR. GOTTESMAN: I haven't worked with any of the strains directly my- 
self, and I would be happy to have comments from other people who have. I 
know that, for instance, in the use of the phage systems, okay, that we have 
approved use there of a host which is partially crippled. That is, it is 
not nearly as sick as Chi-1776, but it has a number of mutations, and in 
that case a couple of tests have been done looking at survival in passage 
through mice, and looking at what comes out in terms of has it reverted for 
the markers which we have considered significant, and it hasn't. So at 
least under those simple experiments it is not reverting at a high rate, 
those particular markers aren't. I am sure people have difficulties working 
with 1776. Maybe some of you people have more — 
DR. FREDRICKSON: Are there any other members of the Committee? Dr. 
Helinski? 
DR. HELINSKI: I would like to comment briefly on the point that Bob 
raised. Several laboratories have deliberately introduced robust E. coli 
K-12 in the same culture with Chi-1776 and then asked if plasmids trans- 
ferred out of Chi-1776 into the E. coli K-12, and the answer in each case is 
no. So these kinds of worst-case situations have been set up. 
DR. FREDRICKSON: Dr. Szybalski. 
DR. SZYBALSKI: I have a paper which I would like to give you to in- 
clude from England, from Petrocheilou and Richmond, which is in press, 
which closes the final gap of safety. It shows that in vivo there is no 
transmission of transmissible plasmid during almost two and half years' of 
testing of all feces of all laboratory personnel twice a week. 
[ 289 ] 
