o 2. 
The P3-LS specifications should be equal to or less stringent 
than the existing laboratory P3 Guidelines. 
Discussion: The P3-LS Level proposed Guidelines require many containment 
specifications and are accordingly significantly more demanding than 
existing laboratory P3 Guidelines. If anything, the P3 — LS specifications 
should be equal or less stringent than laboratory P3 Guidelines as the 
written exception request to NIH to exceed 10 liters of culture requires 
that the DNA recombinants be "rigorously characterized and shown to be 
free of harmful genes". Typical P3 laboratory research with a wide 
variety of new uncharacterized plasmids has significantly higher 
potential risk without this rigorous characterization and establishment 
of the absence of harmful sequences. In addition, most P2 and P3 large 
scale fermentations are not experiments but are repetitive production 
operations using the same characterized culture in a standardized routine 
protocol with specific equipment, validated testing procedures and 
operating protocols which have been reviewed by the RAC and approved by 
the Director of NIH. 
o 3. The large scale Guidelines should describe the specific 
confinement results to be achieved and not the means to this 
result . 
Discussion: In several sections, the proposed large scale specifications 
describe the means to achieve a certain result rather than specifying the 
desired result. For example, section VII-C-3-k would require that all 
exhaust air from production equipment be HEPA filtered prior to discharge 
to the outdoors. The specification should rather require that exhaust 
air released to the outdoors from such aerosol generating equipment not 
contain viable recombinant organisms. The result could be achieved by a 
number of means that may be more practical for a particular installation, 
e.g., air incineration. 
o 4. Large scale containment should be accomplished by the steel 
fermentor itself and not b y a separate room around the 
fermentor(s) . 
Discussion: Insignificant attention has been directed to the excellent 
containment characteristics of modern stainless steel fermentors. With a 
well designed and tested stainless steel fermentor having contained 
sampling and treated exhaust air systems there is no need for a special 
fermentor room or even for curbing to be built around the fermentor (to 
contain the total volume of broth from a fermentor). Such requirements 
are extravagant and impractical for fermentors containing large amounts 
of culture medium (such as the greater than 10,000-L production batches) 
which will be required in the near future to bring the products of 
recombinant DNA technology to the public. In my ten years of extensive 
fermentation experience with both pilot plant and four large fermentor 
production facilities, I have never seen or heard of a fermentor 
rupturing or losing a significant amount of broth containing 
microorganisms. Indeed, in the case of any significant but highly 
improbable leak, the contents of steel fermentors can be rapidly 
disinfected by in-place chemical sterilization, e.g., the immediate 
addition of on-line formaldehyde solution. 
- 2 - 
[ 282 ] 
