Reprinted from TAR JOURNAL OF BIOLOQICAL CHEWSTRY
      Vol. 165, No. 1, September, 194G

DETECTION OF BIOCHEMICAL MUTANTS
      OF MICROORGANISMS

SiTX

Biochemical or nutritional mutants of microorganisms have many uses in
biochemistry and in chemical gcn&ics, l but their application to many
specific problems has been limited by the effort that must be spent to obtain
the specific mutants required in a given instance.  The methods that have
been described previously2 may be summarized briefly as follows:

Mutations are induced in a culture of the microorganism by any of a
variety of agents, including x-radiation, ultraviolet light, and nitrogen mus-
tard gas3  Even with the most efficient mutating agents, only a small
fraction of the cells in the culture are mutants.  In order to detect the
mutants, numerous single spore or single colony isolations are made, and
cultures are established and maintained on a richly supplemented or "com-
plete" medium.  Each culture is then tested for its ability to grow on the
minimal medium.  The occasional strains which lack this ability are classed
as mutants, and their specific nutritional requirements may then be deter-
mined.  Since most of the cultures represent the original wild type, a great
many test.s must be made for the isolation of relatively few mutants, of
which again, only certain ones may be of particular interest.

   The following procedure has been developed to facilitate the detection of
.    biochemical mutants.  It depends on selecting the unmutated cells by their
  ability to form large colonies on a minimal agar medium, and the detection
  of the residue of mutants by the subsequent addition of a mult,iple supple-
  ment which will then allow the mutants to proliferate.  The method has
  been applied primarily to the detection of mutants in ultraviolet-treated
  Escherichia coli, but there is no apparent reason why it should not be
  equally applicable to any organism which forms compact colonies when
  grown on minimal agar medium.

As applied to Escherichia co&, a minimal agar plate is poured in three
layers, to the middle one of which 50 to 400 of the cells to be screened have
been added while the agar was still liquid.  The purpose of the stratification
is to insure a uniform depth for all the colonies, to prevent spreading growth
which is more liable to occur at a glass or air interface, and to keep the

1 Beadle, G. W., Chem. Rev., 37, 15 (1945).
* Beadle, G. W., and Tatum, E. L., Am. J. Bot., 32, 678 (1945).  Gray, C. H., and
Tatum, E. L., Proc. Nal. Ad. SC., 30, 404 (1944). Roepke, R. R., Libby, R. L.,
Small, H. H., J. Bact., 48,401 (1944).
`Taturn, E. L., in Cold Spring Harbor symposia on quantitative biology, Cold
Spring Harbor, in press (1946).
                        381


382               LETTERS TO THE EDITORS               .-

cells in a colony from contaminating hhe entire plate in the next step.  The
plates are incubated until the unmutated cells have developed into colonies
of a convenient size (as experimentally determined). Then a layer of
"complete" ngnr medium is poured on the surface of the plate.  Growth
factors diffusing through the agar reach the unproliferated mubant cells and
permit t.heir development into visible colonies.  In E. coli, these colonies of
mutant cells may be detected by their uniquely small size 6 to 24 hours
after the "complete" medium has been added. The small colonies are
picked to complete medium, and retested for their ability to grow on mini-
mal medium, to verify whether they are indeed mutants.  In the case of
23. cd, most, but not all, of the small colonies have been demonstrably
mutants, but there are indications that highly unstable mutants occur,
which revert readily to the wild t.ype.4  Such mutants would behave like
the wild type subsequent to their initial isolation.

The method has been applied successfully to mixtures of wild type and
previously isolated mutants.  It has, furthermore, been used to isolate a
variety of biochemical mutants from x-ray and ultraviolet-treat,ed E. CO&
including strains blocked at some point in the synthesis of proline, methio-
nine, histidine, isoleucine, cystine, thiamine, or p-aminobenzoic acid.

Osborn Botanical Laboratory
Yale University
New Haven

JOSHUA LEDERBERG~
E.L. TATUM

Received for publication, July 31, 1946

4 Ryan, F. J., and Lederberg, J., unpublished experiments.
6 Fellow of the Jane Coffin Childs Memorial Fund for Medical Research.