I,
Ii,

Tit,

REPORT OF RESEARCH INVESTIGATIONS
for the
ACADEMIC YRAR 1945-49
and for the
SUMMER OF 1948

Subject: RM8:207 Genetics of bacteri«,
Personnel and Other Support:

J. Lederberg, Ass't Professor of Genetics, Project Leader

N, Zinder Research Assistant (now N.1.H.)

D, Gordon Project Assistant

This and related projects are also being supported by a three-
year grant-in-aid from the Rockefeller Foundation ($7500), which is
primarily for the purchase of major equipment, 1948-1951, and by annu-
al grants from the 3.1.H. (oa. $3500) which cover costs of personnel,
supplies, and some equipment for the Saimonella work,
Sqlentific Progress, 1948-491

A.Salmonella. The enclosed mimeographed report was orepared
for the N,I.H., but summarizes the work nearly to date. Since the
report was prepared, (April-May, 1949), somewhat more definite evi-
dence for genetic recombination in Salmonella typhimurium has been
obtained, but the results are still obscured by the problem of lyso-
genicity as discussed therein.

B, Gene enzyme relationships in E. golf. The genetic deter-
mination of /}-galactosidase in Eegherichia coli has been studied,
A large number of lactose-negetive mtants have been isolated fol-
lowing ultra-violet irradiation, These have then compared geneti-
cally and physiologically. At least seven, and possibly as many ag
twelve genes have been identified which are necessary for the for-
mation of this enzyme. As » background for more detailed studies,
~ 2 «

the enzyme has been extracted from the cells, and its properties
examined in yitro with the aid of a chromogenic substrate for gal-
actosidase! o-nitrophenyl- ~galactoside, This compound was kindly
provided by Professor K. P. Link and Mr, M. Seidman of the Depart~
ment of Biochemistry.

Tae extracted galactosidase shows no remarkable features, It
appears to behave as a single enzyme, with a linear tine course of
action, and a linear response to dilution, It ie not inactivated
by dialysis, nor does it require phosphate or other inorganic fons
for activity. No evidence for a coenazyme was found. However, the
enzyme is stimulated by sodium ion, in contrast to a strong inhibdi-
tion by rodidium fon, and a relative inertness of otassium, except
in competition with the other alkali metals, A number of substi-
tuted ammonium ions are also inhibitory (cometitive with sodium).
‘A kinetie study of alkali metal inhibition suggests that it is
largely competitive with substrate but alse shows some “uncompeti-
tive" inhibition, comparable to the inhibition of Atmungsferment
by aside. The results might be interpreted, however, as indic:-ting
that the metal ions are bound to the enzyme at a site adjacent to,
mt not congruent with, that which binds the substrate.

Kinetic studies, involving measurements of velocity as a function
of substrate concentration, fit very well to the Michaelis-Menten
theory of uni-unimoleoular complexes of enzyme and substrate, It
has also been possible to measure the affinities of various analo-
gous galactosides for the enzyme by measuring the extent to which
they impede the splitting of nitrophenol from its galactoside, Ex-
cept that lactose has a relatively low affinity (high dissociation

constant) compared to other galactosides, it is immaterial to
-3-

reproduce the constants here. The interactions of these galacto~
sides provide an exceptionally clear example of competitive inhibi-~-
tion, with which formulation, the kinetics of the inhibition has
fitted well. |
Glucose and other reducing sugars (fructose, maltose, etc.,

bat not sucrose or trehalose) are apparently non-specific inhibitors
of galactosidase, as they are effective only st high concentrations
(ca H/10) and show essentially non-competitive kinetics. The lack
ef competition of glucose is of some importance, because glucose is
an exceedingly effective inhibitor of adaptation to lactose.

pb Geleotosidase is a strictly adaptive enzyme, produced by E.
gold only after exposure to lactose or other galactosides. Cells
grown on glucose or other sugars contain very small quantities of
the ensyme, less than 1% of the wild type, and the amounts that tha
do ecntain may yet be a residual from previous exposure to lactose,
and to some extent aleo, a reeult of adaptation to the nitrophenyl
galactoside used to estinate its activity. The most certain method
of producing adapted cells is cultivation in a lactose-peptone medium.
However, cells harvested and washed from a glucose mediun may be
adapted by suspending them for three hours in a lectose-buffer so~
lution, Gontrary to published statements, adaptation is not sti-
mulated by the addition of inorganic nitrogen sources, but may be
accelerated 2- or 3-fold by the addition of amino acids. Adaptation
can be demonetrated in resting, non-growing suspensions, but is more
Fegular and vigorous vhen accompanied by multiplication of the cells.
Yor reasons not yet understood, some batches of cells fail to adapt
when exposed to lactose for short periods, which has hindered further

quantitative work pending the clarification of the necessary conditions.
The effects of the gene mutations can only be understood well in
relation to the normal process of anzymatic adaptation, on which much
work must be done. It seems to be apparent already that there is no
direct relationship between a single gene and the specificity of the
enzyme finally produced, as already indicated by the very complex
genetic control involving more than seven already identified genes.
However, a start has been made in our attempts to determine how each
of the various mutations affeots the production of the ensyne.

The mutations which absolutely and completely prevent adaptation
ean be analysed no further at present. This comprises the Lace, lacy
and Lacy lool as now understood. wo of the other loci have been ox-
amined to some extent. Lac)- proves not to be absolutely lactose
negative, but produces, under optimal conditions with lactose as th:
growth substrate, about 5% of the normal galactosidase activity.

With batyl galactoside, however, adaptation proceeds very much further,
and results in about 30% of full activity. fhis difference in galasto-
sidase activity is reZlected in the rate at which leuctose if fermented
end results in this paradox! eells of the mutant grown on lactose fer-
ment lactcee mush more slowly than they do if grown on butyl galacto-
side. The Lac,- matation, therefore, has not modified the "specificity-
determination capacities" of the cell, but much more restrictedly, the
capacity of the enzyme adaptation mechanism to respond te one substrate
ae compared with another.

Another mutation, Lac3-, was found to affect not only lactose
fermentation, but aleo the utilisation of glucose, and maltese, One
such isolate proved to be distinotive in its temperature sensitivity:
viz., it behaved like the normal at 30°, and like the mtant at 40°,

Within this temperature interval, there are distinct thresholds for
-5-

the different effects, so that at 37°, for example, the culture vould
be classified as lactose—positive, glucose and maltose-negative. Pre-
liminary experiments have shown that the effect of temperature ie not
on the enzyme, but on ite formation, as shown by tests on cells grown
and tested at different temperatures. Here again, we have evidence
that the genetic effect 1s not on the integrity of a presumed “Template*
but on some aspect of the comlex adaptation mechaniam. Competition
between different substrates during adaptation shows that there 1¢ sone
element common to adaptation of different enzymes, which is sup vorted
here by the spreading or pleiotropic effect of one mutation on several
enzymes.

Studies on maltose utilization by thie mutant have resulted in
a@ collaboration with Professor M, Doudoroff and his colleagues at
the University of California, As mentioned the Lac3- mutant is un-
able to ferment either glucose or maltose. By selection on maltose
medium, however, it is possible toe derive strains which ferment mal-
tose vigorously, but which are inactive on glucose. Doudoroff st al,
showed that this anomaly is an expression of an amylomaltase mechani sn,
whereby n moles of maltose are polymerized into an amylose molecule
and n glucose are split off. The amylose is then phosphorolysed to
glucose-l-phosphate, thus shunting glucose itself which cannot be
utilized. In dried preparations, this interpretation fits very well,
and glucose is acoumilated in requisite amount. It was observed here,
however, that glucose is not accumleated during the fermentation of
maltose by living cells, although small amounts of glucose added to
the mixture were untouched. Serious questions of internal organi~

sation of enzymes, permeability, etc., are ralsed by these obser-

vations, for which there {s so far no satisfactory explanation,
(It is hoped thet some clarification may come out of Professor
Doudoroff's visit to this laboratory this summer.)

C, “Extracellular transfer of heritable factors." It was
hoped initially to study the phenomenon of type transformation
which Boivin had reported in strain "C" of E. gold. A few pre-
liminary experiments, however, goon indicated that our qultures,
labelled "C", received from Boivin did not accord with his de-
scriptions. Attempts to commmicate with him have been unsuccess—
ful, owing, we believe, to his hospitalization. This project has,
therefore, been held in abayance until suitable material can be
obtained,

In the phenomenon of lysogenicity, however, there are certain
aspects which may bear very strongly on what may be going on in sch
verified transform:-tions as those of the pneumococct.

During several years’ handling, strain X-12 of EB goli guve uo
sign of contamination with bacteriovhace. Fortuitously, a mtunt of
Ke12 was isolated which revesled that K-12, and almost all of the
matant gubstrains, that have been obtained from it, ure tenaciously
infeeted with a bacteriophage. ‘This phage has no apparent effects
on the carrier strain, and can be detected only with the help of the
indicator strain, the vhave-sengitive mtant. The latent, lysogenic
phage was termed "lambda",

When lambda~sensitive celle are exposed to lambds, about 3/4 of
the cells are lysed, and release an augmented titre of the phage, The
remaining 1/4 of the cells are not lysed, but instead become immne
carriers of the phage, indistinguishable from the originsl lysogente
Ke12. The phage is rather tenaciously bound to the lysogenic cells,

as full grown cultures of the bacteria contain only a very smull count
-~ Fe

(ca. 100/ ml)of free phage particles. In addition, attempts to dis-
infect the lysogenic bacteria by artificial means (heavy ultra~
violet radiations, rapid cultere at high temperatures, viristatic
chemicals, etc.) have so far been unsucceseful.

At first sight, the only sign by which sensitive and carrier
bacteria can be distinguished is their sensitivity to lysis by lambda.
Subseouently, another phage, p20, wae isolated, from about a thousand
plaques tested from sewage, which although it does not induce lyso-
genicity, ie blockaded by vre-infsction with lambda. That is, cells
which are uniformly sensitive to p20 can be transformed to resistants
by infceting them with lambda, It should be noted that, although con-
sidereble lysis by lambda can be detected in solid medium by plate
counts in liquid medium there is no perceptible change in turbidity
to signalize the lysis. On theother hand, resistance to p20 can al-
go be realized by a gene amtation, which turns out to be the same
as confers resistance to phage 76. From the point of view of the
alteration of phage-resistance, lambda wight de considered as a
genetic transforming principle, lysogenicity is probably far more
common among bacteria than is generally realized, and should always
be considered in mixed culture experiments. However, there is no
evidence that # has any bearing on genetic recombination as studied
in K-12,

D. Segregation from the diploid heterozygote in B. goli K-12.
Previous investigations had involved the isolation of prototroph
haploid recombinants, which indicated a single linkage system in
this Decterium, More recently, diploid heterozygotes have been
{eolated which undergo frequent segregation and reduction to haploids,

which are restricted to being prototrophic. Preliminary studies
IV.

(as published) have indicated that the segregation ratios for vari-

ous genetic markers are far from 1:1, and may deviate as far as 15:1.
This is probably due to the occurrence of a lethal deficiency in om

or both of the chromosomes of the heterozygote. That lethality accounts
for the devietion from normal segregations is supported by the singte
studies which Dr. M. R. Zelle of Cornell University has been doing with
our collaboration. He has been able to follow single-cell pedigrees
ander the microscope, and has found numerous inviable cells, usually

in association with a haploid segregant. However, the mechanism by
which these lethal factors are produced has not been established, nor
has any technique yet been devised which will cirounvent this pert-
urbation, and permit of the detailed study of the genetics of this
becterium in the absence of the disturbed segregations.

Summary.

fhe work on bacterial genetics may be summarized under the following
headings:

A, Salmonella (now with N.I.H. support). Attempts to demonstrate
the occurrence of gene recombination in Salmonella are being carried
out by Mr. Zinder, Although some encouraging, but equivocal results
have been obtained with 8, typhimurium, the prevalence of bacterio-
phage lysogenicity in this group hae eo far interfered with a clear-
cut demonstration. So far, sexual reproduction among bacteria has
deen demonstrated only in one strain of Eegherichia coli. It is
important to determine the prevalence of this phenomenon.

B, Gene-ensyme relationships. A number of different genes have been
found to influence the formation of lactase. Kew methods have been worked

out for assaying this enzyme, and ite kinetics and inhibition have been

studied, It has become apparent that there is not a "one-to-one" relationship
v.

-9-

between gene and enzyme in this instance, and that if there is any
unitary action of the gene in this system, it must be on components
of the adaptive enzyme mechanism, These components have different
specificity, and some mutations may affect several enzymes, Bilo~
chemical study of a maltose-positive, glucose-negative has brought
forth evidence for a "direct" mechanism of maltose fermentation. in-
volving polymerization of the maltose to starch, with subsequent
phosphorolysis.

G. Lysogenicity in 3. soli K-12. It was accidentally discovered
that strain K-12 which had been used for recombination studies, is al-
so an imoune carrier of a latent bacteriophage. The properties of this
system, which is analogous to kaopa in Paramecium, suggest an inter~
pretation of the transformation of serological type in pneumoocce!
based on infection with a cytoplamic "virus".

D. Aberrant heterozygotes 4n K~12,. The isolation of the diploi!
phase in EB. colf permits of the direct analysis of genic segregations.
These segregations are disturbed by a lethal factor which has not yet
been enalysed. The analysis of the disturbance will enable a judgnent
as to the extent to which the chromosomal mechanisms of this bacterium
are the same as in higher forne,

(a) Publications, appeered in print during 1948-1949.
1, Lederberg, J., Problems in Microbial Genetics, Heredity 2:
145-198 1948 (Sept.)

2. * Detection of fermentative variants with

tetrazolium, J. Bact. 56: 695 1948 (Nov.)

3e ® Gene control of B-golactosidase in Escherichia
goli, (abstract) Genotics 33: 617-618 1948 (Nev.)
4, Lederberg, J., Aberrant heterozygotes in Escherichia goli,

Proc. Nat. Acad. Sei, 35: 178-184 1949 (Apr.)
*S. Lederberg, J. and Zinder, N., Concentration of biochemical
matants of bacteria with penicillin, J. Am.
Chem. Soc. 70: 4267-68 1948 (Dec.)
*6, Doudoroff, N., Hassid, W. Z., Putnam, E. W., Potter, A. L.,

Lederberg, J., "Direct" utilization of maltose
in Esgherichia coli. J. Blol. Chem. 179:
921-934 1949 (June)

*Neprints not yet available, will be forwarded when received,

VI. (b) Manuseripte prepared or in press.

1. Lederberg, J., Becterixl Varietion. Ann. Rev. Microbiol.
3: In press.

2. " Isolation and characterization of bicchemical
mutants. Methods of Medicul Research 3: In
preos.

3. fl Reredity, Variation and Adaptction. Ohapter
dn Werkman Vilson, "Bacterial Physiology", in MS.

4, " The -galactosidase of Escherichia golf K~12.
In HS, to be submitted to J. Biol, Chen.