ANNUAL REPORT 19hg-49

Project No, 742. he nature and action of the gene in bacteria, Lederberg
(Mere, Bether Lederberg, National Cancer Inetitute, Junior
Fellow, and University Fellow; Horton Zinder, Ethelyn
Lively, Research Assistants Donald A. Gordon, Project
Asaoc.) State funds, W.A.B.F., Rockefeller Foundation,
National Institute of Health,

fhe research underteken in thie field can be reviewed under the
following headings:

1, Genetic aspects of life cycle in Escherichia ooli; cytological

studies. (Mise Lively)

2, Genetic control of fermentation enaymes in %. coli.

3. Genetic control of mutability, Concluded. (re, Lederberg)
4. Lysogenic bacterial virus in B. coli. (lire, Lederberg)
5

. Genetic recombination in Salmonella and E. coli. (Hr. Zinder)
a
1. Ge 1 1 % .
Zeushis sopeats of ifs avails in Esshertahia s0hs: Sytological studies

The study of the genetic behavior of heterozygous diploid stocks of EF. coli.
obtained in croases of atrain K-12 of thie organism, has been continued, without
marked clarification, ae yet, of the derrant segregation behavior of these diploide,
namely that various segregant types are thrown off at frequencies markedly different
from the theoretical 1:1 for each character pair. Direct evidence that this aber-
ration is due, in part, to a chromosome deficiency heterozygous in the diploid, has
been obtained by examination of diploids which were pure for maltose fermentation,
although one parent was Mal+, the other Hal-. If the failure of such diploids to
segregate for maltose fermentation were due to the presence of a deficiency in-
cluding the Mal+ gene, then reverse mutations for maltose fermentation in such
stocks would lead to pure (not segregating) Mal+ from pure Mal-. Thie was found
to be the case, On the other hand, comparable experinents with pure lactose-
negative diploids, obtained when both parents were Iac-, resulted in reverse mu-
tations which segregated for lactose fermentation. The experiment can be 11llus-
trated: Mal-~/deficiency re Wal+/ deficiency (pure Mal+), but Lac-/Lao-
(aiprota) 2:45 tac-/tact (1.0., segregating for Lac), Additional evidence bear-
{ng on the structure of the aberrant diploids has come out of single cell pedigree
studies with the collaberation of Dr, M. B. Zelle of Cornell Univeraity. In
these pedigrees, inviable cella occur rather frequently, and apparently correlated
with reduction from diploid to haploid. We may conclude, therefore, that the
aberrant segregation ratios are a result of the inviability of approximately half
the segregants, Were it not for crossing-over, all of the viable segregants
would be of the same type, but to the extent that a gene can crose-over with the
deficient region, a haploid segregants bearing the alternative allels can be
recovered. This conclusion is supported by a gradual) variation in segregation
ratio which parallels the linkage distance of a given factor from the deficient

region on the chromosome wap,
-+

These considerations are probably important for crosses in which haploid re-
combinants are isolated ae well, Thie is best illustrated by the fact that the
genetic factors controlling maaltose-fernentation and atreptoaycin-resietance
which show aberrant behavior (1.e., deficiency) in the diploids, also prove to de
unmappable in haploids. We may infer, therefore, that the selotic mechanian is
generally unatable, and that seguental lessees are occurring, partiqularly for the
region containing the Mal ani atreptomyain-resistance genes, in all sygotes,
whether they undergo immediate segregation and give only haploid progeny, or are
delayed in their segregation so that heterosygous diploids are recovered. The
detailed chromosomal mechanian which underlies: thie meiotic inatability is not
yet understood, but is the object of current research.

in hopes of throwing further light on the genetic ayole in EB. geli, a ayto-
logionl study has been initiated, particularly with the object of comparing the
haploid and diploid oultures, Definite differences have been found, so that
the genetic statua of a ouiture can be predicted from its cytological appearance,
but we have not yet been able to interpret the slides, Attached is an enlarged
photomicrograph illustrating the nuclei of typical haploid cultures; comparable
printe of diploid cultures are not presently available.

2. Genetic control of Lexmentation ensymes in 3. ooli.

The study of 3%. coli lactase, or B-gnliactosidase, ae a model aystem for
gene ensyme relationships, has been contimed. In the course of kinetic and
adaptation stadies on intact cells, considerable variation was found from day to
day, which wae ultimately traced to the fact that washed cells of K-12, although
stable in distilled water, are autolysed upon standing in M/15 or higher phosphate
wuffer and other salte, The autolysis can be detected ae an appreciable reduction
in the optical deneity of a cell suspension. Concomitant with the autolysie is
& atriking increase in the apparent ensymatic activity of the suspension, which
may reach as high as thirty~fold. In addition, a emall fraction, about 20%, of
of the activity is released from the cells to the buffer. This result indicated
ah.

that extreme caution would be needed in studies on intact cells, and further work
on adaptation rates and genetic effects would have to wait apon the development
and verification of methods of agesying celle so that the values obtained would
have some meaning.

Upon further examination, it was found that a variety of substances were even
more effective than buffer in induced this “autolytic activation" of the enzyxe.
Thymol, benzene, and caprylic alcohel and desiccation over F205 were especially
effective, and fairly reproducible and optimum amounts of activation could be
odtained with these agents, although entirely satisfactory conditions have nct yet
been worked out, In addition, soaps, gramicidin, ani other detergents have con-
siderable activity, but they erratically inactivate the released ensyne, The
simplest interpretation of the activation is that the cell contains a diffusion
barrier, which impedes the transport of the substrate to the enzyme, Whether this
barrier ia simply the ceil wall, or a more subtle system comparable to the
“eyclophorase” unite of other cells, ie not yet clear, As mentioned in the pre-
vious report, the ensyme in intact celis has different kinetic properties from
that in soluble extracts, which may be a reflection of the kinetics of the trans-
port mechanierm rather than of the ensyme in situ,

With the help of the amplification of activity induced by these autolytic
agents, a clear demonstration that lactase was not entirely adaptive became
possible, That is, cells grown on glucose give extracts whose lactase activity
was as much as 1% of fully adapted celle grown on lactose. A reinvestigation of
this property showed that very considerable amounts of lactose (ca 10-15% of full
adaptation) were present in cells grown on other sugars, @.4., maltose, or on
non- carbohydrate carbon sources, e@.g., sodium succinate or peptone, Thus it
would appear that glucose has the special property of largely suppreasing the
elaboration of lactase, while on other earbon sources, fairly large amounts of

nonadaptive lactase are formed. The distinction between adaptive and constitu-

tive ensymes is further blurred by these experiments.
-5-

As far ag conld be told, the extent of autolytic activation ie the same in
celia containing varying anounte of lactase, That is, the maximum ratio of activity
in matolysed compared to intact cells was uniformly about 30. This observation
clears up the very puzzling and widely cited observations of Deere who, in 1939, re-
ported that enzymatic adaptation to lactose was entirely spurious inasmuch as un-
adapted cells could be “atapted" merely by drying or desiccating then. He was,
however, owing to technical limitations, unable to test these effects on cells
grown on lactose. ‘The activation of adapted cells to the same extent, 50x, in the
current experiments, leaves Little doubt that cells grown on lactose do adapt by
synthesising more enzyme. Fqually, the differences plactose-positives and lactose-
negative mutants, some of which synthesize small amounts of lactase, do not depend
on permeability differences, but on differences in their capacity to synthesize
lactase,

3. Genetic control of mutability. Gongiuded. (Ha. Lederberg)

Daring 1947-48 « number of lactose-negative stocks of B. goli were isolated
which were unusually stable ae far ae their revertibility to Lact was concerned.
Some of these stocks were of particular interest because, in orosees with wild
type, they gave some mutable lactose-negative progeny. This result implies that
some genetic factors, which can be separated from the Lac- locus by recombination,
may be able to influence the true or apparent mutability of the Tao genes. Ina
number of cases which have now been analysed, it appears likely that the modifying
genes are not influencing the true mutability of the Lac loci, but are influencing
the mnifestation of the mutations which do occur, so that few or none of them are
observed. The most characteristic modification is that exerted by a number of
matations which interfere with galactose-fermentation (Qal-), The effect of the
Gal- modifier is two-fold, Firstly, GQal- interferes to some extent with the
fermentation of lactose, so that on a lactose medium, Lact mutations from Lac-
de not have as decisive a selective advantage, Secondly, sany of the Gal-
mutations have an inverse effect on the lac-~ stock, making the latter better
6.
able to fermant lactose. The biochemical basia of these effects is not well
understood, but their overall résult is to make the difference in selective
advantage between Gal-Lac- and Gal~Lact mutants insufficient to ensure the
uniform manifestation of such reverse mitants on the lactose selective mediun,
4, Ignogenic bacterial virus in 2. coli. (Mre. Lederberg)

The sexually fertile strain, K-12, of EZ. coli that hae been used in many of
the reported experiments has been found to carry a latent bacterial virus, opening
up an investigation of the genetic effects and properties of latent virus, As
mentioned in last year's report, lysogenicity has been a serious problem in
Salmonella genetic work, so that 4t has been especially important to learn what
we could about it.

As in plants (e.g., potatoes), latent bacterial viruses can only be detected
with the help of sensitive "indicator" oultures, for they usually have no obvious
effect on their carrier host. In K-12, smach an indicator culture appeared as a
“mutans” in a population treated vith ultraviolet light. Attempts to disinfect
K-12 with ultraviolet under experimentally controlled conditions have however
vyean disappointing.

fhe likeness between bacterial lysogenicity and latent plant viruses is
emphasized by the behavior of the virus in crosses. In plants, latent virus is
ordinarily not tranamitted to seed, In B. coli X-12, a considerable proportion
of sexual progeny are uninfected with and sensitive to the latent virus, provided
one parent is uninfected. However, when both parents are infected, all or alnost
all of the sexual progeny are infected also. The results speak for some sort
of taterruption of virus muktiplication or trananiasion at the sexual stage, but
the details have yet to be uncovered.

5. Genetic recambination in Salmonella and E. gol. (Mr, Zinder)

In hopes of learning the scope and extent of gene recombination among dac-

teria, a systematic survey is being made of coliforms bacteria and of the related

Salmonella group to determine whether recombination cocure in these forme.
~7=

A distinct strain of B. soli, isolated from chicken fecee by Dre, Shapiro and
Sarles, has deen found which crosses with BE. golf F-12, but the yields of recom
binants have been extrenely low, possibly because the chicken strain secretes an
antibiotic active against K-12, However, resictant mitants K-le have just been
feclated, and these should permit of more effective crossing of the two strains,
It ie of interest that one of these straine is aucroset, the other sucrose-,
showing that the varietal differentiation of BE. ooli into commnis and communior
is of very little genetic significance, and likely reste merely on a single gene
matation, A number of Salmonella strains have given results strongly suggestive
of recombination, but this conclusion has not yet been adequately confirmed,

For such a survey it aight be desirable to have a technique of isolating
recombinants accessory to that which depends on nutritional requirements, Drug
resistmce was fount to be a useful basie for recombination studies, Two mutante
of B. gold K-12, one resistant to sodium aside, the other two streptomycin, wera
obtained, Recombinants between the sutants were selected by plating mized cul-
tures into agar containing both azide and atreptomycin. That the doubly resistant
organiame were recombinants was verified by checks on other markers (fernentative,
nutritional and phage-resistance), which frequently had recombined aleo, although
thay were not constrained to do eo by the selective medium iteelf,
~~

1, Lederberg, J. mad A, L, Potter. Aberrant heterozygotes in Escherichia soli.
Proc, Hat'l, Acad, Sot. 35 (4): 178-184. (1949)

2. Yederderg, J., Doudoroff, 4., Haesid, W. Z., and Putnam, ©. ¥. "Direct"
utilisation of maltose by Escherichia goli. Jour. Biol, Chem. 179 (2):
921-934, (1949)

3. Lederberg, J. Bacterial Variation, Am. Rev, Microbiology 3: 1-27 (1949)

4, Iederberg, J. The selection of genetic recombinants with bacterial growth
inhibitors.

December 19119