Interim Report to the Natural Sciences Division
Rockefeller Foundation
May 1, 1949
Project: Genetice of Bacteria,
Jouhua Lederberg, Ass't Professor of Genetics

University of Wisconsin, Madison, Wisconsin

1. Gene-Enszyme Relationships in B. coli.

In order to elucidate gene-enzyme relationships, attention has
deen focussed on a single enzyme, sany mutants obtained in which the
enzyme is deficient or modified, and genetic and physiological differences
Detween the mitante studied. The enzyme chosen for study is beta-d-
galacto-sidase, or lactase, in Escherichia coli K-12. This system was
Selected because (a) lactase works on a readily obtainable substrate,
lactose; (b) 1% proved to be a simple matter to isolate any number of
motante deficient in lactase, by the use of indicator medium which
told whether a given colony on a plate was producing acid or not,
according to whether it could split and ferment lactose; (c) a very
precise speétrophotometric method for the determination of the enzyne
was developed using the analogous substrate, nitrophenyl galactoside.
The compound is colorless, but when nitrophenol is split off under
enzymatic action, 1% gives a yellow color which can be quantitatively
mensured in the spectrophotometer; and (4) {1t proved to be relatively
easy to extract the enzyme from dried cells and to study its vroperties
in vitro.

Several hundred independent lactose-negative mutants have been
{solated from nearly a million colonies examined on indicator medium,

after ultra-violet irradiation, Most of these have been crossed among
themselves in order to define their allelic relationships, At least
seven well-defined groups have been found, such that a lactose-negative
from one group will give some lactose-positive recombinants when crossed
with a lactose-negative from another. Within the same group, lactose-
negative parents give only lactose-negative progeny. The conclusion ¢revn
is that there are at least seven distinct loci interdependently con-
trolling lactose-fermentation, so thet mutation of any one of them
interferes with or modifies the formation of the necessary enzynes,

Studies on extracted galactosidase support the contention that it
is a single ensyme critical in the pathway of lactose metabolism. It
ia completely or nearly absent in extracte and intact cells of each of
the matants.

In addition to the multigenie control of galactosidase, some of
the mutants show alterations of several snayues. For exagple, the mutant
Glass "Lae3z-" fails to ferment glucose, maltose or lactose, These
effects are probably due to distinct enzymes, because (1) the enzymes
occur independently of one another in wild type celle adapted to different
gubstrates, (2) different suppressor mutations have been found which will
reverse the different components of the effects of Lacz-, and (3)
temperature-sensitive allele at the Lacy locus has been found, which is
like the wild type at 30°, like typical Lac3> at 40°, but within the
temperature interval has a different threshold for the different effects.
Temperature affects ensyme formation rather than action: cella adapted
to lmctoue at 30° retain their activity when tested at 40°; celle fail
to adapt to lactose at 40° and later show no activity when tested at
either temperature.

Several findings have suggested that enzymatic adaptation rather
than the specificity of the final enzyme produced met be scrutinized.
Lactase is fairly strictly adaptive, i.e., it is present in very small
amount in cells which have not recently been exposed to lactose. It
seems likely that the site of action of most of the outstions may be
at early stages in adaptation. For example, the *Lac,-" mutant type
which produces practically no lactase in response to lactose, adapts
to a very appreciable extent (abont « third as well as wild type) on
other galactoeide substrates such as butyl galactoside. Thus, Lac,~
cells grown in the presence of lactose ferment lactose very weakly,
while cells grown on butyl galactoaide ferment lactose nearly as well
ag wild tyve. The sume probedly holds for Lacy~. On the other hand,
Lago-, Lae,-, and Lucg- seem to be completely devoid of lactase under
ell conditions tested. Lace~ has a pleiotropic effect similar to that
of Lac3}=-, except that glucose is fermented, and it is gluconate which
is not. Lacg~ ferments lactose, but only very slowly.

These observations suggest thet the genetic control of galactosidase
is quite complex, but they do not yet afford more than leade for further
research in elucidating the mechanism of this control,

This material has not yet been published, except in abstracts, ant
in passing in a review article "Bacterial Variation" which will appear
in Volume 3, Annual Reviews of Microbiology, 1949-50, A description

of the wild type enzyme and the asaay methods is in manuscript.
In collaboretion with a group of biochemists at the University of
California, Berkeley, the mechanisms of maltose fermentation in the
guppressor-muitant combination of Lac3~ which ferments maltose but not
glucose has been investigated. It was found that maltose is polymerigsed
to a ateroh-like polysaccharide under the influence of an enzyme named
“amylomaltase". At a later stage, the polysaccharide is phosphorolysel
to yield glusose-l-phosphate which is then further metabolized, Thus
the splitting of maltose eircumvents glucose, but goes instead via
starch to glucose~l-phosphate. However, this explanation is incomplete,
Amylomaltuse in, 0.¢., dried cells, converts maltose to glucose and
starch in equimolar proportions, and one mol of glucose accumulates

for each mol of maltose utilised.

anylo-
(1) a (G6) —} nc # (8)n
n maltose maltase n glucose plus polysaccharide
of nth degres.
phospho-
(2) (@), # no BPO, Ld a = BP Ld further
é glycolysis
rylase .
starch plus phosphate &lucose-1-phosphate

But intact cells do not accumulate glucose during maltose fermen-
tation, although they will not metabolize exegenous glucose supplied to
them, even during maltose utilization, In some way, the cell must be
able to distinguish between the glucose which is released from the
amylomaltase reaction (1), and that supplied from without. It in not
yet clear whether there might be a chemical difference, or whether it

will be necessary to resort to some explination based on “permeability".

These observetions are reported by Dowdoroff, M., Hassid. W. Z

a ° e . Putnam E. Ww
and Lederberg, J., "Direct" Utiltzation of Maltose by BE. colt, acee:tel for
publication in the Journal of Biological Chemistry. :

oe
- 5 -

2. Gene Recombination in BE. Coli, The isolation and behavior of diploid
heterozygotes.

The usual course of sexual reproduction in B. coli K-12 is shown on
the top line of the enclosed figure. Cells occasionally fuse, forming
the diploid sygote, which undergoes immediate reduction, (See Fig. 1).
In the diagram (Fig. 2) the crossover pattern that would result in a
lnctose-negative prototroph is shown, Of course, in other sygotes, a
erossover between Lac and 3B might rewult in a lactose-positive prototroph,
mt in any event, a single prototroph colony would be either pure Lac-
or pure Lact,

Stocks have now been found which produce sygotes which occasionally
continue to proliferate as diploids for an extended period of tine as
illustrated in Pig. 2. However, in about one division in eight or ten,
@ diploid may undergo segregation, with eressing-over, along the same
patterna as the standard. However, different segregations will be
Lace and Lact, accordingly, the colonies produced will be mosaics
of + and =~ rather than pure ¢ or -, Since the diploid can be isolated
as @ prototroph, and then transferred to complete medium for segregation,
we are not restricted to the recovery of pretotroph recombinunta, but
ean also secure types such as the “sultiple mutant® which is shown as
A~B- in the top line of the diagram. This was not previously possible,
as crosses had to be conducted on mininel medium, (which permits only
prototrophe to develop), in order to suppress the parental celle which
are present in excess, Now, the selection of prototrophs can be used
to secure the intermediate diplofd stage, which can then be plated out

freely on complete mediua.
-~6-

In collaboration with M. R. Zelle of Gornell University, it has been
shown that single cells can be picked individually under the microscope
and that they will later segregate into the yerious genetic types. This
observation further completes the proof for sexual fusion in thie
bacterium. It is hoped to continue with cytological comparisons of
diploid heterosygotes and haploids, with the continued aid of the
present grant, and with assistance from the Research Committee of
the University of Wisconsin,

Some of these experiments have heen published as "Aberrant
Heterosygotes in Escherichia coli", which is due to appesr in the
April 1949 number of the Proseedings of the National Academy of

Selences,
 

 

 

 

 

 

 

 

 

HAPLOID DIPL.OID HAPLOID
PARENTS ZYGOTE PRO FOTROPH
. 7 forms pure Lac-
A if I - *T | colony
Lac Kt |+|-F | 9 a] FR] bb Fy
BI + - +
: cross-! :
fusion gver/ segregation
Qo f
Lg?
proliferation
as diploid
| mosaic colonies:
occasional haploid Lac- &+
segregation

hetererozygous Lact

Fig. a Behbavier of persistent helevenygoles

FACTOR RECOMBINATION

aw ite +e OR
BM: TL Blac V
i x ——____»
teice 27 $
BMiTL Blac V
TYPES
B, Lac V B, Lac V
+ + R - + R
+ - R - = 8
+ + 5 + $s
4+ - 8 - Ss

Fig. |

IN

PROTOTROPHS

7

ae 9
BMTL Blac V

REACTION

Recombmation Co form prototrophs