2 
understand, there are no certified PU facilities presently available to 
U.S. researchers. Under the revised guidelines it will be possible to 
conduct this work with P2 + EK2 containment. Thus, progress in studying 
the molecular basis of thalassaemia and other human genetic disorders 
will be promoted substantially by the proposed revision of the guidelines. 
, 
There is one containment requirement which I feel is unjustified 
and inconsistent. Based on the increased confidence in biological contain- 
ment, vertebrate shotgun experiments have been classified P2 + EK2 without 
the option of performing these experiments under P3 + EK1 containment. In 
your comments on Section III-A-l-a (page 33060) you indicate "However, the 
RAC felt that in view of their increased confidence in the biological 
containment offered by the EK2 system, P2 + EK2 offers more containment 
than P3 + EK1, and that P2 + EK2 without the option of P3 + EK1 should be 
the containment level for certain specified classes of experiments." I agree 
that P2 + EK2 is more effective than P3 + EK1 containment, but I don't 
understand how this should then lead to the conclusion that P3 + EK1 should 
not be an option. Is there evidence that P3 + EK1 does not provide sufficient 
containment? In light of the conclusions of the Falmouth conference, the 
general confidence in E. coli K12 as a safe vector, and years of experience 
with P3 containment of known pathogens , I see no compelling argument for 
the exclusion of the P3 + EKL option to these experiments. Moreover, the 
philosophy of most effective containment is not consistently followed throughout 
the revised guidelines. For example, the containment requirement for cloning 
total transforming virus or subgenomic fragments containing transforming 
genes (lII-A-2-a-(l)-(b)-(l)-(a)) is P2 + EK2 or P3 + EK1. Thus, the requirement 
for cloning DNA known to transform human cells in culture is more flexible 
than the requirement for cloning total human DNA. It is my understanding 
that the primary concern in vertebrate shotgun experiments is the potential 
biohazard of cloned endogenous viruses. If this is the case, why should 
the requirements for cloning endogenous viruses and intact transforming 
viruses differ? 
An important question is whether or not significant experimental 
advantages are conferred by the P3 + EK1 option. I can think of two. 
First, in most cases bacteriophage X EK2 cloning vectors are employed to 
clone specific eukaryotic genes and their flanking sequences. The total 
size of the vector plus eukaryotic DNA is approximately 50,000 base pairs. 
In order to determine the nucleotide sequence of a small portion of the 
cloned eukaryotic DNA, for example 500-1000 base pairs it is necessary to 
isolate large amounts of DNA because the DNA of interest comprises only 
1-2$ of the total recombinant bacteriophage DNA. To circumvent this difficulty 
the small fragment can be subcloned onto a small plasmid using the EK2 
certified host X1776. Unfortunately, very little is gained by this proce- 
dure because of difficulties in working with X1776. In my laboratory and 
many others the yield of DNA from a liter of X1776 is from 5-10 times less 
than that obtained when plasmids are grown in EK1 hosts. Thus, the advan- 
tage of EK2 containment should be weighed against the necessity of handling 
much larger volumes of cells and in the process compromising physical containment 
[A-84] 
