very many

January 20,1950.

Dr. H. P. Trefferes

Dept. Bacteriology,

Yale University School of Medicine,
310 Cedar Street,

New Haven, Conn.

Dear Pete:

Thanks for your note of information on the mutability of 58-278

(which was isolated at Stanford, although I did some work on it at

OBL). Here is one instance, at least, where recombination has been
of somes use.

The augmentation of the mutation rate is certainly remarkable. It
raises the question whether the S™ mtation generally is a single step
from S°, There seem to me to be a mumber of possibilities, which you
have probably considered already. As you may know, Esther hag been working
on the very similar problem of the genetic control of mtability, but
at the Lac locus. Judging from her results, 58-278 may differ from K-12
in any of the following sigh&ficant respects:

i. It may carry an allele of 5S, say s", which is itself sensitive, .
but with a higher mtation rate to S$. This can be verified if 478 x
SF gives sensitive prototrophs all of which are mutable.

2. It may carry the standard S*° alkel, but also a modifier "Ms",
which may act either a) to increase the intrinsic mutability of the
S locus, or b) interacts with mutations other than S™ lat S or other
loci] so that they are phenotypically resistant, which they otherwise
would not be. On this hypothesis, 478 x S™ will give some stable
sensitive prototrophs. Also, on this hypothesis, one might expect
that some S* stocks, if crossed to S$, will yleld some mutable sensitive
recombinants. This would be true, in particular, of 478 $°.

vail 2% prototrophs tested from your cross were S" tends

THS Lact tha
to support (1)>~for-titts is precisety-ths-behavior of the 3 locus:
faa prototrophs like~the-?hBy—parent. ~~This is not write: ighore 1
I have in press with J. Bact. a note on using inhibitors to
select for recombinants in bacteria (K-12 in particular). Strepto-
mycin and azide were uséd in the experiments which consisted of
Az¥ SS x AzS S*, and recovering Az’ S recombinants by plating
the mixtures into a medium with both compounds. Streptomycin worked
very well indeed, because of the typically low mutation rate,
single-step resgistamce, and low background, but azida was at
best a poor expedient, although most of the dual resistants re-
covered were demonstrably recombinants. If you can suggest another
inhibitor comparable to streptomycin in its desirable characteristics,
which will work well with entérobacteriaceae, and does not "cross-resist"
with streptomyein, I would be grateful for the suggestion.

Some months from now, we.will be stazzting some work on the mechanism
of spontaneous mutation. I had planned to use streptomycin resistance,

but 162° 4s an inconveniently low rate. Would it be intruding on your
plans if we used 58-278 for these experiments?

Sincerely,

Joshua Lederberg