6 
Dr. McGarrity asked Dr. Muth to describe exhaust filtration procedures. 
Dr. Muth said most organizations use two filters, either a wet/dry filter or a 
dry only filter and one or the other combinations of these filters. Seme 
organizations use a filter and an incinerator. Meet facilities, however, 
employ a pair of sterilizing grade units in the exhaust system to handle the 
volume of air passed through a 10,000 gallon fermentor. Such filters represent 
a considerable expense. A Demin ic-Hunter filter, for instance, consists of 
sixty 24- inch cartridges set in a large housing. 
Dr. McKinney asked how much air would pass through such a filter. Dr. Muth 
replied that the amount of air depends on the age of the fermentation. Early 
in the process, less air is required; characteristically most fermentation 
plants would operate on a volume of air per volume of medium per minute. For 
a 10,000 gallon tank, that is approximately 1,500 to 2,500 standard cubic feet 
per minute (SCFM) . Dr. McKinney observed that under such conditions, a large 
number of organisms could be discharged to the atmosphere. Dr. Muth said other 
processes are used in the exhaust line in addition to filters. At Eli Lilly 
the first process consists of a a mechanical coalescer vhich throws the air 
into a cyclonic flow. Large particles are dropped out and returned, sterilely, 
to the fermentor. Behind the coalescer, in series, is a coalescing filter 
which removes droplets. Together, the coalescer and coalescing filter remove 
99.9% of the organisms. Dr. McKinney noted that this procedure is equivalent 
to the use of a HEPA filter. 
Mr. Barbeito asked Dr. Muth whether Eli Lilly facilities are built for a specific 
purpose or whether they are multipurpose. Dr. Muth replied that each product 
is anticipated to have a specific plant. Mr. Barbeito said such use contrasts 
with use of a research laboratory; a laboratory will have multiple uses over a 
number of years. Mr. Barbeito asked if the data produced by Eli Lilly (Attach- 
ment II) were generated in an actual test run. Dr. Marsh replied the data 
were generated in a "dunmy run," i.e., the host vector system without the 
reccrribinant gene was tested. 
Dr. McGarrity said industry's primary concerns appear to be with the language 
of Appendix K-II-B and K-II-D. He asked if the problems encountered with this 
language were unique to Eli Lilly manufacturing procedures. Dr. Muth thought 
these issues were germaine to most production companies. 
Dr . McKinney noted that some data on organism survival in aerosols and on sur- 
faces had been sent to the working group for their information (Attachment III), 
and that Schering Corporation had supplied seme additional information (Attach- 
ment IV). Dr. McKinney said Ms. Hunt would present additional data^ on survival 
of recombinant DNA containing organisms in aerosols to the working group. 
Ms. Hunt said that she had compared the aerosol survival of one recombinant 
E. coli host-vector system and a wild type E^ coli at two relative humidities 
7*45% and 70%) and room temperature. She had also compared one HV1 B. subtilis 
host-vector system. 
^Thesis presented as part of the requirements for Dr.FH at Lhiversity of North 
Carolina School of Public Health, Chapel Hill, NC, May 1984, Debra Long Hunt. 
[ 342 ] 
