Artificial eliminatfon of F factor in Bact.  coli K-12,

A few years ago, Lederberg and his co-workers (1952; 1953)

fsund that in Bact. cgli K-12 there was a difference in
sex-compstibility between the two mating cells, one having a
transmissible agent, F, and the other lacking this agent.
Hayes (1953h) demonstrated that these mating types were very
stable heritable characters, but, the accidental disappearance

of the F factor was f&d by Lederberd 6. al. (1952), and by
Hayes (1953a).
      The study`rep'orted here reveals that it is
possible to obtain F' cells at will from F+ cells under certain

environmental conditions.  This phenomenon may be called "F

eli:nination".  The environmental factor responsible for F
elimination is the cobalt or nicks1 concentration in the medium

( Co(N4 1 t CoCl
    32     2'  etc.; Ni(N33)2, NiSJ4, etc.). The first
method of artificial eliminatin of F factor is called the
"&v&44-"
trBTV method. A culture of F' cells was inoculated into
yeptone medium ( peptone, log. and glucose, 2g. in 1 liter
distilled wate:'.j , and incubated overnight at 37'C.  Bfter

24hrs., c>,15ml. of &4X. cobalt sAution was added to 3 ml.

of the culture ( the final concentration was 20mX.) which was
Incubated for 24 hrs. m$eI'  This treated culture was then spread
on nutrient ager containing 40mM. of sodium citrate, which was
used for detoxification of cobalt, and single colonies grown on
the medium were isolated at random.  By this procedure, it was
found that many colonies wsre converted Into Fr For example, in

the mutant K-12 strain 58-161 and the derived strains from 58-161,
10-30 ; of the colonies were converted into Fr However, all of
the F- were sensitive to cobalt.  'i'he frequency of conversion
varies according to the strain of Bact. coli K-12 used.  The

second method is called "resistant isolatiorP, in which cobalt-
resistant colonies are selected by serial transfer in peptone


2

medium containing cobalt in increasing concentrations.  A slow
and smooth increie of resistance was observed.  Starting from

a concentration of 0,5mM.,  in which strains derived from the

original Bact. coli K-12 are able to grow, resistant strain8
tolerating even 2OmM. were obtained.  These resfstant strains

wsre plated on nutrient agar containing 20mM. of cobalt, and

10 well grown colonies were isolated at random,   In mutant
strains S-161 and Y-40, for example, all 10 colonies were F;
whereas in the W-1485 strain, only 4 colonies were F'.  However,
drtg resistance and the F" character of these strains could

be separated; cobalt ~sensitive strains derived from the resistant
were F' as before, &nd cobalt resistants received the F factor
from F+ bacteria to become F+ themselves. The mating behavior
of the F- strains correspondsd to that af the F' strain described
by Hayes (1952a;bI and by Lederberg et, &., (1952).  By the

cobalt treatment, other characteristics, e.g. nutritional requlre-

ments, sugar fermentation, phage-resistance, lysogenicity, indol

formation, shape, gram staining, and antibiotic resistance, were

                                                                           '

not changed,  The F' trait is very stable in successive serial

passages in ordinary media such as broth, peptone, and minimal

medium,  Therefore, in the Ft clone, the replication of the F

factor may be closely concernd with bacterial reproduction.

This hereditary conversion of mating type within a clone gives

rise to the question of the basis of the change and the

mechanism by which the F' strains occured.  There are two well-

recognized possibilities:  (a) F elimination may be due to the
selection of spontaneous gene mutation4 (b) it may be induced by

the nickel or cobalt In the medium.  From the experimental data,

the following may be concluded;  (1) F elimination occurs sven in

the absence of cell division ( in the first method); (2) Ths


                                              3
occurrence of F elimination is about 100,000 times more frequent than
that of other bacterial mutants; (3)F+ and F- strains are equally
susceptible to the killing action of cobalt or nickel as judged
  by viable counts, and F- trait is n& linked with cobalt or

  nickel resistance, These conclusions seem to excluderiossibility
  (a).  The conjection used to explain the contradiction in the
stability and instability of the F factor is that it is a non-

lytic symbiotic agent, as reported Cavalli &. a&(1953), and
Hayes (1953a 1.  It may behave as a cytoplamic factor JgpJ? w fa4r-
J-& a#& dt 44+cGfdlc F+LLU dlc,
certain stage of its life cycle,  In this state, the sensitivity
of the F factor to cobalt or nickel may be greater than that of
other cytoplasmic factors or of nuclear genes.

                  Acknowledgement
   The author wishes to express his appreciation of the sugges-
tions and encouragement given by professor H, Kikkawa,  The

writer is also indebted to Mr. T. Iijima who gave his sound
advice and criticism during the course of this study.
                  Y.'Hirota

                     Department of Genetics,

                    Faculty of Science,

                    %aka University, Osaka,
Cavalli, L.L., Lederberg, J., and Lederberg, E.M., J. Gen, Microbial,
   8, 89 (1953).

Hayes, W., Nature, Lond, 169, 116 (1952a).

Hayes, We, Nature, Lond. I&, 1017 (1952b).

  Hayes, W,, J. Gen. Microbial. 6, 72 (1953a).

Hayes, W., Cold Spr. Harb. Symp. Quant. Biol. 18, 75 (1953b).
Lederberg, J., Cavalli, L. L., and Lederberg, E. M., Genetics,
   22, 720 (1952).