JAN 2 135,

ANNUAL REPORT 194950

Project No. 742. The nature and action of the cene in bacteria. Lederberg

(Mrs. Esther Lederberg, Project Associate, Chemical Corps;
Phyllis Fried, Ethelyn Lively, Nerton Zinder, Research
Assistants.) State funds, W.A.R.¥., Rockefeller Founda~
tion, National Institute of Health, and Chemical Corps,

The research undertaken in thie f1eld can be reviewed under the

following heedings:

i.
2.
3.
4,

Formal genetics and cytogenetica of Zecheriehis ool}.
Genetic control of fermentation ensynes in B- ool}.
Lysogenis bacterial viruses.

Recombination in Salmonella.
oo Pw

1. Formal genetics and cytogenetics of Eecherichia coli.

Efforta to unravel the complex life cycle of §. coli strain Kel2 were con-
tinued during 1950, but with no great clarifieation of the underlying mechaniens
to date. The main question remaining wneclved is the basis for the elimination
of genetic material from the sygote which results, for example, in diploid cells
which are deficient for part of one gene set (including genes controlling mal-
tose and galactose fermentation, and streptomycin-resistance.) A number of com
plex hypotheses, were tested critically, but none of them were surported by the
experimental results. This problem, though difficult, is an important one, since
on it hinges the question of how close is the similarity between the genetic or~
ganization of bacteria and of other plants and animals whose breeding systens
have been successfully exploited for both scientific and technical purposes.

The evidence, overall, shows a very close similarity, in particular that the
bacterial genes are organized in chromosomes, but there are certain peculiarities
that cannot yet be understood in terms of simple chromosome behavior, like that
of the more familiar materiale of experimental breeding.

Another approach we have been following is a direct cytological examination
of haploid and diploid clones of EB. golj. ‘These can be distinguished almost
absolutely on the basis of their nuclear atructure, as illustrated in the accom
panying photogrephs, Figures 1 and 2. However, a study of the nuclear consti-
tution of haploid and diploid bacteria in all growth phases must be completed
before too definite conclusions may be drawn. In certain growth phases, haploid
bacteria mey somewhat resemble what we have regarded as the typical diploid cay-
telogical picture, raising the possibility that the morphological differences
are based primarily only on the larger size and more open structure of the diploid
nuglei, which gives the appearance cf a greater number of resolvable granules,
fhe cytological pletures so far do not permit an identification of chromosomes

or anything like a count of the chromatic units, so that this level of correlation
-~3-

of genetic and cytological study remains to be exploited. There are indications
thet the nuclear bodies may be visualisable in living cells by phase microscopy,
and it is hoped thet thie may help to f111 in the pioture of the behavior of the
nuclear bodies during eell diviefion, genetic segregation, and so on.

In conneetion with these studies we have also been studying the mode of
action of various bactericidal agents, much es ultra-violet light, mustard gas,
formaldehyde, and others. These agents have profound effects on the genetic
behavior of diploid cells, correlated with distuybances in the nuclear picture.
Figure 3 shows a emall clone of cells, neer recevered from ultra-violet exposure.
The large, filamentous cell is typical of sublethal responses tc some bectericides;
the amali cells have probably split off from the large cell at an earlier tine,
and are multiplying more repidly.

One of the applisations of diploid clones is to the problem of dominance.
Yor a number of reasons, it has been important to determine whether drug-resia-
tance 19 a recessive or dominant trait. Unfortunately, genes controlling resistance
to streptomycin in &. ¢goli are often eliminated during meiosis, so that hetero-
sygotes are difficult to secure; however, one or two exceptional dipleide have
been isclated which are heterozygous for streptomycin-resistance, and the latter
fe found to be recessive: i.e., a cell carrying one factor for susceptibility
and one for resistance is phenotypically susceptible tc this drug. Resistance
to vacteriophages has aleo been found to be receasive, so that it is likely that
most mutations for resistance to antibacterial agents will be recessive to the
wild-type, dominant, sensitive form. This has some bearing on the mechanians
of development of drug-resistance in the course of chemotherapy, but makes it
the more important thet we understand the cenetic organization of the bacteria
we wich to eradicate,

A vecent finding which may make it possible to rationalize the genetic
behavior of ZB. goli K~l2 is the isolation of other strains which can be eressed
-he

with this strein. Until now, the crossability of B, golf was nearly unique for
a single strain, K-12, attempte to use other strains having given inconclusive
or methodologically useless results. However, by courtesy of the State Public
Health Laboratory, we have received « considerable aumber of fresh isolates of
this bacterium from hamen sources, and almost 10% of these straing heve beon
found to cross with Kel2, and probably with eash other, so that the phenomencn of
recombination is no longer a unique one, There are indicetions thet some of
these strains dc not show the aberrant genetic behavior reported for K-12. In
addition, the natural differences between the various intercrosesble etreins
provide abundant material for studying the basis of serological and cultural
eharacteristies which sre used in the claseification of these bacteria ae isolated
from their natural hebditate.

2. Genetic contra) of Sermeutetion anzrmee in B col}.

The primary fermentetion enzymes (in particular the oligesaccharases) of
bacteria are usually recognized ae "adaptive", that is, they are prodused by
the bacteria only after contact for some time with the substrate (e.¢., lactose).
A previous report mentioned that the distinction between “adaptive® and non=-
adaptive or constitutive ensymes was rather blurred, because small amounts of
lactase can be detected, by special techniques, within unadapted cells. It
has now been found that "adaptivity" is under genetic control, for in addition
te previously described mutations which prevent adaptation, and limit enzyme
formation, a new matation has been found whose effect is to cause the cells to
preduse lactase abudantly, regardless of previous exposure to lactose or other
eubatrates. Thies further illustrates the concept that the enzymes are produced
as a result of a complex cell machinery in which the genes play a master-controll-
ing rather than a direct rele. (The alternative hypothesis, that each gene
playe a direct role in the production of a single enzyme hee had wide currency

as the socalled "one gene-one enzyme theory".)
Anothar findings rrevides tha first evidence for 2 cocaible *rosition effect"
of bacterial genes, and shows how complex these may be. Two lactece-nerstive
mutations had baen studied here vraviously, and thought to be "allelie"™, i.e.,
affecting the game cane, for when crossed with esch other, these mutenta had
not given rise to non~autant offspring. Tests on a lerger scole showed, however,
thet these two mutations wers not identical, because non=-mitant offsrring occurred
in a ratio of about 1: 27000, The mutetions may therefore be thonght of 26
affecting two genes extremely close torether or ndjacent, Studies of hetero-
zygous Giploids heve now shown thet a cell which carrier one non-mitant rene
en one chromorome, and the non-mutent form of the edjacent cene on the other
chromosome doer not ferment lactose, whereas these cener in the normal rosition,
togethcr on the same chromescme, show normal activity. Several examples cf this
kind ere Imown in Dronophila: we uay speculste thet the two cenes must lie
adjacent becauee they interact through a non-diffusible primary gene product.
Although this materfial may be more suitable in some respects than Drosophila,
4t ie rather doubtful thet any immediate pregreas on identifieation of the
hypothetical gene products will follew, oving to the insuperable technical diffi-e
culties. Further studies in the purification and kinetics of lectese are being —
pursued in sollaberation with Professor H. A. Lardy and his students at the
Rnzyme Institute.

3+ isogenic bacterial viruses.

(Since this section must be cleared by Chemical Corps before release, no
detailed information will be given here prior to publication.) fhis investi-
gation deals with the transmission of a latent virus in crosses of carrisr,
resistant, and susceptible atrains. Methods of “disinfection” or removal of

the virus from viable, infected bacteria are aleo under study.
4. Recombination ip Salmonelja.

For some time, we have attempted to find a system of genetic recombination

in a pathogenic bacterium, Ealgonelia typhimurium, comparable to that we have
~6-=

been ueine in BH. coli. or nearly three yeers, concerted efforts in this direo=
tion vere wisuccessful, elthovgh aboub thirty different strains of &8. typhimurium
and related species had been laboriously and exhauetively tested in various com
binati-ns. We d{d not have even the leaser satisfaction of a definitely nezative
result, for certain acmbinstiona gave new tynee whick strongl, suggested. = ve
combination precess, but whitch for incidentel vexscne were technicaliy wie:it-ble
far further atudy, However, = rair of atrains has now been found which give
anite elesreuk avidence ef genetic recombination in Selmonelle typhimerius,
opening up the roasthility of an enalysis in thie evester compar2zbie to thet in
Be gold K-12. The epeetel interest tn Salmonella ia the opportunity afforded

to study ceretic amects of nathagenicity, for which B. coli is unmiltebdle.
Upvever, 2 detailed analyela cf the formal genatica of this material will be
necessary before any serious work on the infecticus disease assvects can hore to

be auecesafnl., (Note: Because of the recenty of these results, please eons 1t
Professor Lederberg before publication.)