one-to-one correspondence between the biosafety levels and the 
classification of agents. 
Dr. Stevenson emphasized the need for a list that would not equate 
safe microorganisms with dangerous ones. For example, E. coli K-12 
strain chi-1776, which is a safe recombinant DNA host, should not 
be listed with more dangerous, toxigenic strains of E. coli . He 
stated that such distinctions will become an increasingly important 
problem for shipment and exchange, particularly when up to 30,000 
cosmids are produced from the chromosome material of the human 
genome, each of which could be maintained in an E. coli host. 
Dr. McKinney stated that the Environmental Protection Agency has 
recognized that recombinant DNA molecules express different 
characteristics, some benign and some extremely toxic. For 
shipping purposes, regulators need to look at the combination of 
the host and the vector and what characteristics are being 
expressed. Dr. Langston added that E. coli 1776 could express 
toxicity. She asked: (1) how could shipping be monitored; and (2) 
how should regulators evaluate the infectivity or disease-producing 
properties of agents. If agents are not infectious or disease- 
producing, they should not be labelled "etiologic . " 
Dr. McKinney agreed. The outer container label should accurately 
reflect the contents. Dr. Langston added that, to protect against 
spills or breaches of packaging, further actions in the NIH 
Guidelines should be issued on decontamination procedures. For 
example, postage handlers could be instructed to pour clorox bleach 
over the container. 
Dr. Stevenson suggested several categories for etiologic agents: 
infectious, toxic, and contagious. He believed that if the product 
does not fall into one of these categories, it should not be 
shipped as an "etiologic agent." Such a policy might eliminate the 
labelling problem. For example, blood specimens that contain 
etiologic agents differ from blood cultures with etiologic agents; 
the latter does not contain a high enough concentration to pose a 
health threat and so should not be labelled as "etiologic." Dr. 
Langston argued that handlers should know the contents of a 
package. Dr. McKinney asked if the regulations should be 
rewritten. 
Dr. McVicar believed that it should. He stated that the "etiologic 
agent" label was the least common denominator for shipment in the 
NIH Guidelines and other regulations. He suggested that the 
Subcommittee examine the formal definition of an "etiologic agent." 
He noted that the CDC have been struggling with definitions. For 
example, while the meaning of "diagnostic specimens" is clear to 
many shippers, it causes regulatory problems because it is too 
restrictive. For example, when a blood sample is sent to a 
laboratory, it is treated as a diagnostic specimen subject to 
applicable regulations and guidelines. If the same sample is sent 
instead to a storage facility, it falls outside of all regulation. 
Recombinant DNA Research, Volume 13 
[284] 
