STANFORD UNIVERSITY MEDICAL CENTER

DEPARTMENT OF GENETICS

August 12, 1977

Dr. G. Bernardi
Institute de Recherche en Biologie Moleculaire
Universite Paris VII
Tour 43, 2 Place Jussieu
75221 Paris
France-

Dear Dr. Bernardi:                                                . ,

  I was glad to have your letter of July 28th, which asks certain questions
about the hazard assessment of cloning in Bacillus subtilis 168.  There are,
of course, three variables in such experiments: the source of the DNA insert,
the plasmid or cloning vector, and the organism -- in this case g. subtilis ;-
used to propagate the hybrid plasmids.

  In view of the many criticisms that have been offered against the use of
E. eoli, attributed to its close ecological connection with the human species,
it is indeed fortunate that a practical cloning system is now available in
another bacterial species like B. .subtilis.  While it is impossible to
describe any bacterium as totally harmless under all conceivable conditions,
g. subtilis does not depend on its association with animal hosts in its
normal life cycle, has no specific adaptations for such association, and
has no evident specific pathogenic potential. It,can, therefore, be argued
to be as nearly harmless as any microorganism that will ever be discovered
-- including wine yeast, Penicillium Roquefortii, Lactobacillus acidophilds
or many other organisms that are part of a natural diet, but thereby are
even more closely associated with man.  According to this argument, to the
fact that the plasmids grown in &. subtilis are not biologically transmissible.
and to the observed instability of the plasmids in the absence of selective
pressure, the system that Dr. Ehrlich has developed should be an ideal one
from every perspective.  These are compelling arguments to support a sub-
stantial reduction of hazard in the use of this organism relative to E. coli,
                                                                           -
particularly if non-sporing strains are used.

  I would not agree with a blanket elevation of containment levels for
"non-patbog enic soil organism DNA" in g. subtilis, compared to E. coli.  This
view unzerastimates the extent to which enteric bacteria, genetically co-
mingled ;;Lth E. coli, are equally common in soil.  Aerobacter and Pseudomonas
are particularly in point here.  The likelihood of hazard is so low that I


Dr. G. Bernardi                   -2-             August 12, 1977

would advocate general standards of the same 'stringency as for E. coli;
but of course one should take account of specific circumstances-that might
elicit more specific concerns.  If you are discussing DNA from pathogenic
soil organisms, with respect to traits that Could plausibly be expected to
influence  the pathogenicity of B. subtilis,; you would of course be dealing
with a different situation.

  My fear in making these recommendations is that they may give unduly high
credit to the hazard that is supposa)to be associated with "E. coli," a
designation which is in many respects a semantic artifact. yt is just an
accident that E. coli K-12 has the same taxonomic designation as freshly
isolated strains of E. coli that are demonstrably adapted to sustain life
in the.gut.  It is just too bad that mere historical accident requires us
to use the common epithet!

  In summary: whatever merit one attributes to restrictions on the in-
sertion of DNA from non-pathogenic sources into plasmids grown in E. coli,.
one should haLe substantially fewer concerns about the use of the B. subtilis
                                                                          -
system.

  Whatever actions your committee takes, I trust that it will be receptive
to important new scientific information that is being developed at a very
rapid rate.  I believe this already tends to show that the global fears about
the biohazard of recombin\ant DNA experiments have been grossly exaggerated
from the inception of this controversy.  The most pertinent information is
that genetic exchange is a common natural process; and there is already sub-
stantial evidence that the experiments in the laboratory do little to add
to the natural flow of genetic information in the biosphere on a large scale.
The principal distinction between the natural and the laboratory environment
is not the production of novel DNA sequences, but rather the selective
pressures to which they are exposed.               4

Yours sincerely,

Joshua Lederberg
Professor of Genetics

JL:ek-f