ABSTRACT

STUDIES ON THE GALACTOSE LOCI OF
SALMONELLA PULLORUM

by Richard Willis Vaughan

Galactose—negative mutants of Salmonella pullorum

 

were isolated in order to construct a map of the relative
order and positions of mutants within the galactose region.
An attempt was made to study some aspects of regulation of
the enzymes involved in the Leloir pathway-

The mutants were grouped according to the enzyme(s)
of the Leloir pathway in which they were deficient. A
mutation in the kinase locus was found to permit constitutive
synthesis of transferase and epimerase. Six of the 8 trans—
feraseless mutants studied showed some effect upon the enzyme
level of galactokinase. Possible reasons for the effects
are discussed.

The independent probabilities of recombination between

mutants following transduction by Salmonella pullorum phage

 

P38 were used in the ratio test to assign the most probable
relative positions. An unlinked arabinose-negative marker
was utilized as an internal control. The transferase locus

was found to locate between the kinase and epimerase loci.

Richard Willis Vaughan

The ratio test was then used to map the relative order and
relative positions of 7 mutants within the transferase locus.

A modification of the ratio test was used to de—
termine the relative order of 19 mutants within the kinase
locus. In the modified ratio test the kinase locus was ap-
proached by transducing from the linked transferase and
epimerase loci, both of which are located on the same side
of the kinase locus. This method introduces non—linearity
into the kinase locus map, precluding determination of any
relative distances.

One possible 00 operator type mutant was found which
mapped between the kinase and transferase loci or in the
region of the transferase locus closest to the kinase locus.

The theoretical distances employed in the modified
ratio test were transformed into an equation of a straight
line. The slope of the straight line should equal the
distance between the average epimeraseless mutant and the
average transferase mutant. When data obtained from mapping
the kinase locus were used in the formula, the slope of the
resulting straight line approximated the distance found by

the ratio test.

STUDIES ON THE GALACTOSE LOCI OF

SALMONELLA PULLORUM

BY

Richard Willis Vaughan

A THESIS

Submitted to
Michigan State University
in partial fulfillment of the requirements
for the degree of

DOCTOR OF PHILOSOPHY

Department of Microbiology and Public Health

1965

(pf)

This thesis is dedicated to

my family

ii

ACKNOWLEDGMENTS

I wish to express my appreciation to Dr. Delbert
E. Schoenhard for his patience and guidance throughout the
course of my education at Michigan State University.

I should also like to acknowledge the valuable
advice and assistance given by Dr. Harold L. Sadoff.

During the course of this study, I was supported in
part by an Agricultural Experiment Station Research Assistant-

ship and a Graduate Council Tuition Scholarship.

iii

TABLE OF CONTENTS

Page

INTRODUCTION . . . . . . . . . . . . . . . . . . . . . 1
LITERATURE REVIEW 4
Galactose Metabolism . 4
Mapping of the Galactose Genes 9
MATERIALS AND METHODS . . . . . . . . . . . . . . . . 12
Cultures . . . . . . . . . . . . . . . . . . . . . 12
Phage . . . . . . . . . . . . . . . . . . . . . . 12
Media . . . . . . . . . . l3
L—Cysteine HCI stock solution . . . . . . . . l3
Nutrient agar . . . . . . . . . . . . . . . . l3
Nutrient broth . . . . . . . . . . . . . . . . 13
Slant agar . . . . . . . . . . . . . . . . . l3
Nutrient soft agar . . . . . . . . . . 13

40% stock carbohydrate solutions . . . . . . . l3
Eosin-methylene blue agar . . . . . . . . . . l4
Phenol- red agar . . . . . . . . . . . . . l4
Tryptone- tetrazolium agar . . . . . . . . . . l4

M—9 broth . . . . . . . . . . . . . . . . . . 15
Tryptone broth . . . . . . . . . . . . 16
Induction and Selection of Mutants . . . . . . . . l6
2- aminopurine . . . . . . . . . . . . . . . . l6
Diethyl sulfate . . . . . . . . . . . . . . . l6
Nemethyl-N—nitroso-N'-quanidine . . . . . . . l7
Selection of epimeraseless mutants . . . . . . 17
Preparation of Phage Lysates . . . . . . . . . . . 18
Donor phage . . . . . . . . . . . . . . . . . 18
Homologous phage . . . . . . . . . . . . . . . l9
Transduction Procedure . . . . . . . . . . . . . . 19
Preparation of recipients . . . . . . . . . 19
Preparation of transducing lysates . . . . . . l9
Transduction test . . . . . . . . . . . 20
Determination of Galactose Enzymes . . . . . . . . 22
Chemicals . . . . . . . . . . 22
Preparation of cell free extracts . . . . . . 22
Assay of galactokinase . . . . . . . 24

Assay of galactose— 1- phosphate— uridyl

transferase . . . . . . . . . . . . 25

iv

Page

Assay of uridine-diphosphogalactose-4—

epimerase . . . . . . . . . . . . . . . . . 26
RESULTS . . . . . . . . . . . . . . . . . . . . . . . 27
Carbohydrate Utilization of Mutants Isolated . . . 27
Strains of Galactose-Negative Mutants Used . . . . 27
Classification of Galactose Mutants According
to Enzyme Deficiencies . . . . . . . . . . . 33
Constitutivity of the Enzymes in Various
Mutants . . . . . 33
Transduction Frequencies of P38/35W Donor Phage
with Selected Ara Gal recipients . . . . . . . 37
Nomenclature . . . . . . . . . . . . . . . . . . 38

Use of the Comparative Recombination Frequen-

cies of the Kinase, Transferase and Epimerase

Loci . . . . . . . . . . . . . . . . . . 38
Use of the Ratio Test to Order Individual

Mutant Sites Within the UDPbgal Transferase

Locus . . . . . . . . . . . . . . . . 42
Use of the Ratio Test to Order Individual
Mutant Sites Within the Galactokinase Locus . . 43
Theoretical Aspects of the Linearity of the
Galactose Loci . . . . . . . . . . . . 48
Attempts to Transduce Gal Epimeraseless
Mutants to Gal+ . . . . . . . . . . . . . . . . 54
DISCUSSION . . . . . . . . . . . . . . . . . . . . . . 56
SUMMARY . . . . . . . . . . . . . . . . . . . . . . . 64
BIBLIOGRAPHY . . . . . . . . . . . . . . . . . . . . . 66

Table

LIST OF TABLES

Characteristics of gal-ara- mutants

+ - .
Ara gal stra1ns used as sources of donor
phage for transduction

Phenotypic description and inducibility of
galactose enzymes of ara‘gal‘ mutants

Transduction frequency of KI and T_ mutants
by P38/W35

Mean probability of independent integration
(p) is shown for seven T and seven K
recipients involving mutants at K, T and
E loci as donors

Total number of transductants analyzed and
the probability of independent integration
(p) are shown for seyen T recipients in
tests involving 14 K and 2 E donors

Summary of data in Table 6 and their use in
positioning mutant sites within the T
locus of Salmonella pullorum

 

Total number of transductions analyzed and
the probability of independent integration
(p) are shown for 19 K recipients in tests
involving 2 T- and 2 E- donors

Summary of data in Table 8 and its use in

ordering mutant sites within the K
region of Salmonella pullorum

vi

Page

28

32

34

39

41

44

45

49

50

LIST OF FIGURES

Figure Page

1. Map drawn to scale and showing the relative
locations of 7 mutants within the T (UDP-
galactose transferase) locus of Salmonella

 

 

pullorum . . . . . . . . . . . . . . . . . . 46
2. Map of the most probable relative order of

galactose region mutants of Salmonella

pullorum . . . . . . . . . . . . . . . . . . 51
3. Use of data from ordering kinase mutants in

determining the distance from T to E . . . . 53

vii

INTRODUCTI ON

Since the rediscovery of Gregor Mendel's early work,
geneticists have been interested mainly in the genetic
aspects of gross properties. The work of Morgan (1910) and
Sturtevant (1913) with Drosophila led to the theory that
genes are arranged in a linear sequence along a chromosome.
Much effort has been expended in attempting to dissect pheno-
typic "markers" into their simplest components. During this
period some geneticists must have asked themselves the
question, ”How does a gene function?"

In 1941 Beadle and Tatum began investigating not
only the phenotypic aspects of individual genes but also
their function. Results from their work led to the one-gene—
one-enzyme hypothesis. It is now assumed (though not neces—
sarily true) that a particular function, or related series of
functions, is determined by only one gene or by a closely
linked cluster of genes.

The discovery of transduction in the bacterium
Salmonella by Zinder and Lederberg (1952) provided geneti—
cists with a method of high resolution for dissecting a gene
into smaller segments. By simultaneously investigating the
biochemical function and control of formation of the protein

produced by a gene, an overall scheme can be analyzed. High

resolution analysis provided the focal point at which
genetics and molecular biology met.

It was through just such a study of lactose utili-
zation that Jacob et a1. (1960) developed the operon
hypothesis.

Similar work on galactose utilization, undertaken by
Kalckar and his coworkers (1959) and Lederberg (1960) with
Escherichia coli and Fukasawa and Nikaido (1961) with
Salmonella typhimurium, prompted an investigation of galactose

utilization in Salmonella pullorum. Several reasons were re—

 

sponsible for this decision.

‘E. coli and S. typhimurium are closely related, as

 

determined by their chromosomal map, while evidence has ac—
cumulated that S. pullorum is somewhat divergent. Conju-
gation studies (Robinson, 1964) have indicated a galactose
"marker" which maps in a different location than in E'.£Qll

and S. typhimurium. Perhaps the order of the genes within

 

the galactose sequence is also different.

Buttin (1960) reported that phage A.has interference
effects on the operation of the galactose operon in g. 291;.
Therefore the )1 —.§..ggli system may not be the most ideal
situation to study.

A mutation in the kinase gene produces pleiotropic
effects which are not attributed to kinase being an operator.
It was hoped that some insight into this problem could be

gained.

Finally, a transduction system for fine structure

mapping would serve as a model for investigations of other

aspects of_§. pullorum. Some of the natural mutant character—

istics found in "wild type" S. pullorum are of particular

interest.

LITERATURE REVIEW

Galactose Metabolism

The enzymic mechanism of galactose metabolism re-
cently elucidated in yeasts (Kosterlitz, 1943; Leloir, 1951;
Kalckar, Braganca and Munch-Petersen, 1953), bacteria
(Kurahashi, 1957; Kalckar, Kurahashi and Jordan, 1959;

Nikaido, 1961) and man (Isselbacher t al., 1956; Kalckar,

 

Anderson and Isselbacher, 1956) is due primarily to the
pioneering work of Leloir (1955).

Utilization of galactose by_§..ggli K12 has been
shown by Kurahashi (1957), Kalckar, Kurahashi and Jordan
(1959) and Soffer (1961) to proceed virtually exclusively by

the Leloir pathway:

 

'1. Galactose + ATP galactokinase} galactose-l-phosphate +

ADP

2. Galactose—l—phosphate + uridine diphosphate glucose

galactose-l—phosphate uridyl transferase\

 

 

\
uridine diphosphate galactose + glucose—l-phosphate

3. Uridine diphosphate galactose

uridine diphosphate galactose-4—epimeraseX

 

 

Y
uridine diphosphate glucose

 

Sum: Galactose + ATP-————9 glucose—l-phosphate + ADP

Kurahashi (1957) also demonstrated the presence of uridine
diphosphate glucose pyrophosphorylase in E. coli which

catalyses the conversion (Munch-Petersen t al., 1953):

 

Uridine triphosphate + glucose—l-phosphate

uridine diphosphate glucose pyrophosphorylaseA

 

 

\
uridine diphosphate glucose + pyrophosphate

Kurahashi (1957) found that extracts from galactose
mutants of g. 9911 K12 were deficient in kinase or transferase
and demonstrated that 12.11EEQ complementation occurred with
these extracts. Kinaseless and transferaseless mutants were
demonstrated in g. typhimurium by Fukasawa and Nikaido
(1961b). Lack of epimerase was shown in g. 391; C7M by
Kalckar, Kurahashi and Jordan (1959), Fukasawa and Nikaido
(1959b), in E.‘ggli K12 by Soffer (1961), and in S.
typhimurium by Fukasawa and Nikaido (1959b, 1961a). Uridine
diphosphate glucose pyrophosphorylase is required to supply
uridine diphosphate glucose to initiate the conversion of
galactose-l—phosphate to uridine diphosphate galactose. This
is inferred from the fact that .E..ggli K12 mutants deficient

in uridine diphosphate glucose pyrophosphorylase have a

galactose-negative phenotype (Fukasawa, Jokura and Kurahashi,
1962; Sundararajan, Rapin and Kalckar, 1962). Buttin (1963a)
described mutants which appear to be lacking galactose
permease.

Mutants deficient in kinase, transferase, epimerase
or pyrophosphorylase exhibited effects other than a simple
galactose-negative phenotype. The majority of kinase mutants
showed a pleiotropic effect in which transferase and epimerase
are synthesized constitutively (Kalckar, Kurahashi and
Jordan, 1959: Jordan, Yarmolinsky and Kalckar, 1962; Nikaido
and Fukasawa, 1961: Fukasawa and Nikaido, 1961b). Evidence
has been presented indicating that the constitutivity of
transferase and epimerase is due to an internal inducer
(Jordan, Yarmolinsky and Kalckar, 1962; Jordan and Yarmolinsky,
1963).

Transferaseless mutants of g. 2211 undergo bacterio-
stasis in the presence of galactose when grown in synthetic
media with glycerol as the sole source of carbon (Kurahashi
and Wahba, 1958; Yarmolinsky _t_a1 , 1959). Sundararajan
(1963) found that the galactose-l—phosphate accumulated by
transferaseless mutants interferes with induction of glycero—
kinase. The bacteriostatic effect of galactose was almost
completely overcome by the addition of yeast extract or
Camain hydrolysate to the growth medium. Glucose completely

reP’r‘essed bacteriostasis (Kurahashi and Wahba, 1958).

In

(I)

“I

Epimeraseless mutants of E. coli (strains K12 and

C7M),_§. typhimurium (strains LT2 and LT7) and S. enteritidis

 

undergo bacteriolysis when grown in the presence of galactose
(Yarmolinsky gt al., 1959: Fukasawa and Nikaido, 1959a,
1961a, 1961b). Fukasawa and Nikaido (1961a) demonstrated
that cells in the stationary phase of growth are resistant

to galactose-induced sensitivity, and protein synthesis was
necessary for bacteriolysis to occur.

Sundararajan, Rapin and Kalckar (1962) reported that
an E'.£211 K12 mutant deficient in pyrophosphorylase becomes
bacteriostatic when grown on minimal medium supplemented
with galactose, a situation analogous to transferase de—
ficient mutants.

Buttin (1961, 1963a) showed that fucose, a gratuitous
inducer, could induce kinase, transferase and epimerase to
levels equal to or higher than those obtained with galactose.
Since fucose is an inducer not utilized by the cells, control
of galactose enzymes must be a coordinate induction and not
sequential.

Buttin (1961, 1963b) isolated two types of_§._ggli
mutants which constitutively synthesize kinase, transferase
and epimerase. One type of mutant (R_) maps close to a
lYSine gene but away from the cluster of kinase, transferase
and epimerase genes. The second mutant type (OC) maps with-
hfitflde galactose gene cluster at one end of the epimerase

locus. Coordinate induction of the galactose enzymes and

isolation of R— and OC mutants has led Buttin (1961, 1963b)
to the conclusion that the galactose gene sequence is an
operon under the control of a regulator gene, analogous to
the regulator postulated for the lactose system (Jacob §£.él°:
1960; Jacob and Monod, 1961; see also Jacob and Monod, 1962).
Kalckar, Kurahashi and Jordan (1959), Lederberg

(1960) and Soffer (1961) reported that several mutants of g.
£911 K12 can synthesize only traces of all three enzymes in—
volved in galactose metabolism. Nikaido and Fukasawa (1961)

and Fukasawa and Nikaido (1961b) demonstrated similar results

with one mutant of S. typhimurium LT7. These mutants were

 

not capable of complementing with any of the known mutants
lacking singly kinase or transferase and are regarded by
Jacob gt 31, (1960) as examples of a typical 00 mutation.
Buttin (1963b). however, stated that these typical 00 mutants
could be the result of a mutation in the epimerase gene re—
sulting in interrupted transcription of the transferase and
kinase genes. It was proposed that DNA transcription is a
polar process being initiated at the operator locus (Jacob
and Monod, 1961). The m-RNA so synthesized would contain
information for the complete operon. Hence, an interruption
at the proximal end could result in "missense" or ”nonsense"
transcmiption of loci distal to the interruption.

Buttin (1963b) presented evidence indicating the
galaflrbose permease locus is separate from the galactose

Opercnq but under the control of the galactose regulator gene.

Mapping of the Galactose Genes

A majority of the galactose—negative mutations found
in §._ggli K12 affect a short chromosome segment (gal region)
which is selectively transduced by bacteriophage‘A, The
transductants are relatively stable diploid clones for the
gal region (Morse. Lederberg and Lederberg, 1956a,b).

This property has made it possible to classify gal
mutants into four functionally complementary groups
(Lederberg, 1960). Biochemical characterization of the
phenotypes has shown that three of these groups correspond
to an activity defect in kinase, transferase or epimerase
respectively, and the fourth group consists of OO mutants
(Kalckar, Kurahashi and Jordan, 1956; Soffer, 1961). A
fifth group of gal mutants. not transducible by phage )_, are
probably defective in pyrophosphorylase (Morse, 1963).
Echols, Reznichek and Adhya (1963) obtained similar results
using Fgal partial diploids of E. £911 K12.

By transduction, galactose mutants of S. typhimurium
have been classed into three different groups corresponding
to defects in kinase, epimerase and a single 00 mutant
(Hartman, 1956; Fukasawa and Nikaido, 1961b).

Several methods have been employed in attempting to
order the functional groups relative to each other. Fukasawa
andNikaido (1961b) used phage P22 in reciprocal trans-

duction tests between epimeraseless and kinaseless S.

. , o
IQEEgggurium mutants. One 0 mutant was used only as a

10

recipient. Their results indicated the 00 mutant was located
closer to the K locus than to the E locus.

The location of several gal mutants belonging to the
same structural gene locus has been mapped by means of the
probabilities of recombination obtained from the study of
clones of heterogenotic cells. In such clones. the £1999-
position genes, + — / - + (mutant phenotype) undergo recombi—
nation to yield the £19-position ++/-— (wild phenotype).
Certain OO mutants showing noncomplementarity with mutants
from more than one locus allowed ordering of the loci them-
selves. Morse (1962) applied this method to mutants of E.
£911 K12 and determined the loci order as K—OO-T. The order
of 7 kinaseless mutants, 2 OO mutants and 8 transferaseless
mutants was determined within their respective loci.

Echols, Reznichek and Adhya (1963) introduced an
Fgal particle into a gal OO recipient to determine the re—
combination frequency between the Fgal particle and the
bacterial chromosome. Their results indicated that K was
the farthest distance from 00, E was closest to 00, and T
was intermediate. The experiments did not determine on which
side of O0 a given mutation was located.

By transduction with phage P1, Adler and Kaiser (1963)
have genetically mapped_§- £911 K12 mutants. The order of
lS‘gal mutants was studied by means of three-point crosses
usirug prophage 82 as the closely linked reference point.

The rnost probably sequence of the loci was shown to be

11

K—T-OO-E—prophage 82. However, the authors were unable to
establish an unambiguous order for mutant sites within their
respective loci.

Buttin (1963b) employed Hfr x F- crosses to map gal
mutants of E. £911 K12. His results indicated an order
K-T—E—O with OC located at the far extremity of the E locus.
The R gal locus mapped close to lysine and away from the gal
region. Buttin's work is perhaps the most definitive study
to date concerning the ordering and regulation of the loci

involved in galactose metabolism.

MATERIALS AND METHODS

Cultures

Three strains of Salmonella pullorum, 35W, 38W and

 

25W, were employed in this study. Strain 35W was mutated to
an arabinose negative phenotype by ultraviolet light.
Mutant 35ara-2 served as the parent strain for isolation of
galactose-negative mutants (35ara-2,gal-) to be used as
sources of homologous phage or as recipients in transduction
tests. Spontaneous ara+ backmutants from 35ara-2,gal-
mutants were selected to serve as sources of donor phage.
Strain 38W is lysogenic for bacteriophage P38 and was used
as the initial source of P38. Strain 25W, sensitive to P38,
‘was used to titer phage lysates.

All bacterial strains were maintained on slant agar

111 screw cap tubes and kept at room temperature.

Edaaqe
Phage P38 is a temperate phage isolated from 38W
leated on 35W. Initial isolation was done on nutrient agar

plaates according to the soft agar overlay method of Adams
(119559). The phage particles were suspended in T2 buffer and

stcbrned in screw cap tubes at 4 C.

Media

13

L-Cysteine~HCl Stock Solution--Stock L-Cysteine°HCl

solutions were prepared by dissolving 200 mg
L—Cysteine~HCl in 100 ml distilled water and
were sterilized by filtration. Appropriate

amounts of the L-Cysteine°HCl stock solution
were added to media after the media had been

autoclaved and cooled to 45 C.

Difco—nutrient agar and Difco—nutrient broth were

supplemented with 20 mg L-Cysteine°HCl and 5 g

NaCl per liter.

Slant Agar
Dehydrated nutrient broth (Difco) 8 g
NaCl 5 g

L-Cysteine°HCl (10 m1 of stock solution) 20 mg
Bacto—Agar (Difco) 20 g

Distilled water 1000 ml

Nutrient Soft Agar

40%

Dehydrated nutrient broth (Difco) 8 g
NaCl 5 g
L—CysteineeHC1 (10 ml of stock solution) 20 mg
Bacto—Agar (Difco) 7 g
Distilled water 1000 ml
Stock Carbohydrate Solutions--Individual carbo—
hydrate solutions (40 g of carbohydrate made up

to 100 ml final volume with distilled water)

14

were autoclaved separately. In all cases where

media containing carbohydrate were used,

ate amounts of the 40% stock solutions were added

appropri—

after the media had been autoclaved and cooled

to 45 C. Carbohydrates were obtained from

Pfanstiehl Laboratories, Waukegan, Illinois.

Eosin—methylene Blue Agar (EMB agar)

Levine EMB agar without lactose 27.4 g

(Baron, Spilman and Carey formula;
Baltimore Biological Laboratory)

Bacto-Casamino acids (Difco, no.

0230-01) 8 g
Carbohydrate (12.5 ml of 40% solution) 5 g
Distilled water 1000 m1
Phenol~Red Agar
Phenol—Red Broth Base (Difco) 16 g
Bacto-Agar 15 g
Carbohydrate (12.5 ml of 40% solution) 5 g
Distilled water 1000 m1
Tryptone-Tetrazolium Agar (TTC agar) (Arber, 1958)
Bacto-Tryptone (Difco) 10 g
Bacto—Agar 15 g
L-Cysteine»HCl (10 ml of stock solution) 20 mg
2,3,5-Tripheny1 Tetrazolium Chloride 500 mg
(Sigma Chemical Company) (10 m1 of 5%
solution)
Carbohydrate (12.5 ml of 40% solution) 5 g

Distilled water 967.5 ml

15

L—CysteineoHCl, 2,3,5-tripheny1 tetrazolium chloride
and carbohydrate were added after the agar had
been autoclaved and cooled to 45 C.

M—9 Broth (Levine and Borthwick, 1963)
20X M-9 Salts-—Components were dissolved in

order and stored over chloroform.

KH2P04 60 g
Na2HP04 (anhydrous) 120 g
NH4C1 20 g
Distilled water 870 m1

10X Casamino Acids
Casamino acids (Difco no. 0230-01) 150 g
Distilled water 950 ml
Norite Activated Charcoal 20 g
The mixture was allowed to stand over-
night at 4 C, filtered, and stored over
chloroform at 4C.
Preparation of 2X M—9 salts and 2X Casamino Acids—-
Ten ml of 20X M-9 salts were diluted with 90
ml distilled water and autoclaved. Twenty
m1 of 10X Casamino acids were diluted with
80 ml distilled water and autoclaved.
Preparation of M—9 Broth
2X M—9 salts 100 ml

2X Casamino acids 100 m1

16

1M MgSO4°7H20 0.5 ml
25% NaCl 0.4 ml
40% Dextrose 2 m1

Tryptone Broth

Bacto-Tryptone 10 g
NaCl 5 9
L—CysteineaHCl (10 ml of stock

solution) 20 mg
MgSO4-7H20 2.5 g
Distilled water 990 m1

Induction and Selection of Mutants

 

2-Aminopurine (Robinson, 1964)—-Approximate1y 100
cells of a logarithmic culture of strain 35ara-2 were inocu-
lated into 10 ml of nutrient broth containing 200‘bg/ml 2-
aminopurine° The culture was incubated at 37 C with aeration
until maximum turbidity developed. An induced culture was
diluted 1:103 and 1:104, and 0.1 m1 samples were spread on
EMB agar supplemented with 0.5% (W/V) galactose (EMBgal agar)
to obtain 103--104 colonies per plate. Mutant colonies ap-
peared white, dark—red, or white with a small, dark—red
center.

Diethyl sulfate (DES)——A non—aerated overnight
nutrient broth culture of 35ara—2 was diluted 1:101 into

fresh broth supplemented with 1% galactose (W/V) prewarmed

to 37 C in a screw cap tube. After 3 hr of incubation

17

without aeration, 2 drops of DES were added, and incubation
was continued. Samples of 0.05 and 0.1 ml were withdrawn
and spread on EMBgal agar plates at 45 and 60 min. The time
of incubation and amount of dilution required to obtain 103--

104 cells per plate varied from experiment to experiment.

N—methyl-N—nitroso-N'-guanidine (NG)-—A non—aerated,
overnight nutrient broth (pH 5.6) culture of 35ara-2 was
diluted 1:101 into fresh prewarmed broth (pH 5.6). Incu-
bation was continued 3 hr, and 0.2 ml of 1 mg NG dissolved
in 1 ml ethanol (pH 5.6) were added. After one more hour of

2 3

incubation, 1:10 and 1:10 dilutions were made, and 0.1 ml

samples were spread on EMBgal agar plates.

Selection of mutants deficient in UDP—galactose—4-

epimerase--Epimeraseless mutants of Escherichia coli, Salmon-

 

ella typhimurium and Salmonella enteriditis exhibit lysis

 

when grown in the presence of as little as 10-4M galactose
(Fukasawa and Nikaido, 1959a; Fukasawa and Nikaido, 1959b;
Yarmolinsky, Wiesmeyer, Kalckar, and Jordan, 1959). It was

assumed that similar mutants of Salmonella pullorum would be-

 

have in the same manner.

Strain 35ara-2 was treated by one of the above in-
duction procedures and diluted to obtain approximately 100
colonies when spread on a nutrient agar plate. After 24 hr
incubation, the nutrient agar plate was replicated to a
plate of EMBgal agar. Both the master and replicated EMBgal

plates were then incubated an additional 24 hr. Colonies

18

which formed on master plates but not on EMBgal plates were
purified further on nutrient agar. After three purifications,
the suspected epimeraseless mutants were again checked for
their inability to grow on EMBgal agar. Those mutants which
remained sensitive were transferred onto agar slants and

were classified as probable epimeraseless mutants. Each
presumptive mutant was assayed for UDP—galactose—4-epimerase

activity before final classification.
Preparation of Phage Lysates

Donor phage—~Mutants (35ara+ga1-) were streaked on
EMBgal agar plates. After 48 hr of incubation, an isolated
colony was transferred to Mr9 broth, and the broth was aerated
overnight at 37 C. The optical density was determined in a
Bausch and Lomb Spectronic 20 Spectrophotometer, and the
culture was diluted to approximately 2 x 108 cells/m1 with
fresh, prewarmed M-9 broth, 10 m1 final volume, or the over-
night culture was diluted 1:20 with fresh prewarmed broth.
The 10 ml diluted cultures were aerated 3 hr, and phage
grown on 35W were added to a final concentration of 1-2 x 106
plaque—forming units per m1 (pfu/ml). Aeration was continued
for 8—10 hr. The presence of free phage in the cultures was
usually indicated by the appearance of white, viscous
Inaterial around the top of the culture tubes.

Lysed cultures were centrifuged at 6000 x G, and the

Supernatant fluid passed through a 0.45’p pore—size millipore

19

filter membrane. Filtrates were stored in screw cap tubes
at 4 C and checked for bacterial contamination by spotting
0.1 ml samples on nutrient agar plates.

Plaque-forming units per ml were determined according
to the soft agar overlay method of Adams (1959) using strain
25W as an indicator organism. Phage lysates were diluted in

T2 buffer and plated in duplicate on nutrient agar plates.

Homologous phage-—Homologous phage were prepared as
above on 35ara-gal-strains to be used as recipients in the

transduction tests.

Transduction Procedure

Preparation of recipients-—Isolated colonies from 48
hr streak cultures on EMBgal agar (epimeraseless mutants
were streaked on nutrient agar) were transferred to 10 ml of
M—9 broth and incubated overnight with aeration. Overnight
cultures were centrifuged, washed twice with equal volumes
of T2 buffer, and resuspended in 1-2 ml of T2 buffer. The
optical density was recorded, and the cultures were diluted

to 1 x 1010 cells/ml with T2 buffer.

Preparation of phage lysates for transduction tests—-
Phage lysates were diluted with T2 buffer to l x 108 pfu/ml.
Ten ml of the diluted lysates were shaken mechanically while
Tbeing irradiated with ultraviolet light for 60 sec at a

distance of 51 cm. Precautions were taken to avoid photo-

reauztivation effects. Under these conditions, 99—99.9% of

20

the plaque forming units were inactivated. Irradiated ly-
sates were placed in screw cap tubes, wrapped with aluminum

foil, and stored at 4 C at least 12 hr before use in trans-

duction tests.

Transduction test——Nine-tenths of a m1 of irradiated
phage lysates were transferred to 13 x 100 mm test tubes and
allowed to equilibrate to 37 C. After 10 min of temperature
equilibration, 0.1 m1 samples of recipient cultures were
mixed with the lysates. The multiplicity of input was ap-
proximately 1 phage per 10 bacteria. A period of 10-15 min
was allowed for phage adsorption (approximately 99% adsorp-
tion) before 0.1 m1 samples were spread on each of 3 TTCgal
and 3 TTCara agar plates. Spread plates were kept at room
temperature for 2 hr before continuing incubation at 37 C.
Carbohydrate—positive dark-red papillae appeared on the
white carbohydrate-negative background growth after 48—72 hr
of incubation.

Each recipient used in a transduction test was
treated as above with homologous phage to serve as a control
for backmutation. One-tenth m1 samples of phage lysates and
tflue T2 buffer used in the test were spotted on nutrient agar
ffilates to serve as controls for bacterial contamination.

Transductants arising on the homologous phage control
Plaites were subtracted from the number of transductants
ariwsing on the test plates, and the ratio of galactose-

traJisductants to the galactose-transductants plus

21

arabinose-transductants (gal/gal+ara) was computed.
Arabinose, not linked to galactose in transduction tests,
served as an internal control. The number of galactose-
transductants arising from a particular test was assumed to
be a function of the distance between the alleles under in-
vestigation. However, the number of arabinose-transductants
was assumed to be independent of galactose and to arise from
transducing particles independent from galactose-transducing
particles. In other words, two classes of the recombination
event took place in different bacteria. Consequently, vari-
ations in the number of arabinose-transductants were expected
to be due to variations in the efficiency of the transduction
process and not to peculiarities of the arabinose allele
(eg.. differences in transducing potency of individual phage
preparations, media, temperature, dilution errors, host
susceptibility to the phage, etc.). Since experimental vari-
ations were presumed to affect the frequency of galactose

and arabinose transductants alike, the ratio gal/ga1+ara
transductants should be essentially constant between several
tests of the same two gal alleles. The ratio of gal/gal+ara
transductants then served as a more accurate representation
of the recombination frequency between two particular gal

alleles.

In. Ir-r1

Ff

Cc

mi

22

Enzymatic Determination of
Galactose Enzymes

 

Chemicals--For enzyme induction, Sigma grade galac-
tose, essentially free of glucose, was used (Sigma Chemical
Company). D—galactose—l-phosphate was purchased from Mann
Research Laboratories or the California Corporation for Bio-
chemical Research. Adenosine-S'-triphosphate (ATP) was ob-
tained from Mann Research Laboratories. D—galactose—l-C-l4,
glucose—6-phosphate dehydrogenase (G—6-PdeH) and phospho—
glucomutase (PGM) were also obtained from the California
Corporation for Biochemical Research. Other chemicals
(purchased from the Sigma Chemical Company) were: nicotina-
mide adenine dinucleotide (NAD), reduced nicotinamide adenine
dinucleotide (NADH2), nicotinamide adenine dinucleotide phos-
phate (NADP), reduced nicotinamide adenine dinucleotide
(NADPHZ), uridine diphosphate glucose (UDPG), uridine-5'-
diphosphoglucose dehydrogenase (Type III) (UDPGdeH) and 6-
phosphogluconic acid (6-P—g1uconate). Uridine-5'-diphospho—
galactose (UDPgal) was either obtained from the California
Corporation for Biochemical Research or prepared according

to the method described by Wiesmeyer and Jordan (1961).

Preparation of cell free extracts—-An isolated colony
from an EMBgal agar plate (nutrient agar was used for
epimeraseless mutants) was transferred to 10 ml of tryptone
broth and aerated 18—20 hr. A 0.1 ml sample was inoculated

into 100 m1 of tryptone broth contained in an 8 oz

23

prescription bottle equipped with a metal cap through which

a Pasteur pipette was passed. After 18 hr of incubation with
forced aeration, the culture was diluted (approximately 2 x
108 cells/ml final concentration) into 700 ml of tryptone
broth contained in a 32 oz prescription bottle equipped with
a gas dispersion tube. For enzyme induction 0.01 M galactose,
essentially glucose free, was present for at least three
generations. Aeration was continued until a concentration

of 1—2 x 109 cells/ml was reached.

The bacteria were collected at 7000 x G in a re—
frigerated centrifuge and washed twice with 150 m1 quantities
of 0.005 M potassium phosphate buffer (pH 7.0). Packed cells
were then resuspended in 10 ml of 0.02 M potassium phosphate
buffer containing 0.01 M mercaptoacetic acid and 0.001 M
EDTA at a final pH of 7.0 (Sherman and Adler, 1963).

A Nossal cell disintegrator equipped with a C02 jet
(McDonald Engineering Company, Bay Village, Ohio) was used
to disrupt bacterial cells. Resuspended cells (10 ml) were
transferred to an 18 ml stainless steel capsule containing
10 g of glass beads, 125—177 p in diameter. The glass beads
and capsule were precooled to -20 C. Shaking in the dis-
integrator was stopped every 30 sec to allow cooling. The
CD2 jet was Operated intermittently to keep frost on the
Capsule. Usually two 30 sec periods were sufficient to
break the cells. The mixture was subsequently spun for 10

min at 30,000 x G in a refrigerated centrifuge, and the

24

supernatant fluid was then spun in a Beckman model L centri-
fuge at 106,000 x G for 1 hr. Each extract was divided into
two portions; one was stored at -20 C and the other placed
on ice for subsequent enzymatic assay.

Protein was determined by the method of Lowery _E._l-
(1951) after precipitation with trichloroacetic acid (final
concentration of 10%). Protein content ranged from 1-5 mg/ml.
Enzyme preparations to be assayed were diluted to 1 mg/ml of
protein with 0.02 M potassium phosphate buffer containing

0.01 M mercaptoacetic acid, 0.001 M EDTA, and 100 pg of

bovine serum albumin per m1, at a final pH of 7.4.

Assay of galactokinase-—Galactokinase was assayed by
the procedure of Sherman (1962) and Sherman and Adler (1963)
with the following modifications: (1) enzyme extract, 20‘p1
(20 pg of protein), was added to the assay mixture, and after
30 min of incubation, 204p1 of the reaction mixture were
spotted on Whatman paper DE—20 (DE-81); (2) after elution,
discarding the bottom 5 cm of the DE-20 paper strip, and
drying, the paper was placed in a scintillation vial so as to
form a circle around the inside of the vial; and (3) the
vials were counted in a Packard Tri-Carb liquid scintillation
spectrometer at tap 4 and window settings of 10-50-50 volts.

Very carefully cleaned glassware was used, and
reagents were prepared with highly deionized water. Horowitz
(1962) recommended rinsing pipettes with 0.0002 M EDTA in

ckaionized water before use since heavy metals are apparently

25

antagonistic to galactokinase. The radioactive homogeneity
of the gal-l-C-l4 was checked by chromatography together with
galactose (essentially glucose—free) and with glucose in
n—butanol—pyridine-water, 6:4:3 by volume (French, Knapp and
Pazur, 1950).

The amount of galactose phosphorylated was directly
proportional to the amount of galactokinase added, at least
up to 20% utilization of the galactose, confirming the obser-
vations of Sherman and Adler (1963). One unit of enzyme was
defined as that amount of enzyme which phosphorulates l‘pmole
of D—galactose per hour under the conditions described
(Sherman and Adler, 1963). Specific activity was expressed
as units per milligram of protein.

Contrary to reports by Kalckar, Kurahashi and Jordan
(1959) and Sherman and Adler (1963), the author was not able

to freeze galactokinase and retain appreciable activity.

Assay of galactose-1-phosphate-uridy1 transferase--
The method of Kalckar, Kurahashi and Jordan (1959) was fol-
lowed with some variance in the amounts of components added
to a 1 m1 cuvette. The reaction mixture contained the
following: cysteine solution (33 mg/ml, pH 8.5), 30’pl;
MgCl2 (0.01 M), 10‘p1; glycine (l M, pH 8.7), 60 p1; PGM
(105 enzyme units/ml in 0.15 M acetate buffer, pH 5.5), 10
,Pl7 NADP (20 mg/ml), lO‘pl; G—6—PdeH (2 enzyme units/m1),
10‘p1;'UDPG (10 pmole/ml), 20_p1; cell free extract (1 mg

protein/ml), 50-100‘p1; and water, 420—370‘p1. Gal—l-P,

26

30 p1 of a 10_umole/m1 solution, was added to start the re—
action after allowing 10 min for equilibtation to 37 C. The
rate of NADP reduction was followed directly in a Beckman
spectrophotometer equipped with a Ledland log converter and
a Sargent recorder.

Endogenous NADP reduction was checked by omitting
gal-l-p from the above reaction mixture. NADPH2 oxidase
activity was negligible. Reducing the amount of enzyme
extract added to the reaction mixture by one half was done
to ensure that the indicator enzymes were in excess.

The specific activity of galactose-l-phosphate—

uridyl transferase was defined as the‘pmoles of gal-l-p con-

verted to UDPgal per hour per mg of protein.

Assay of uridine-diphosphogalactose-4-epimerase——
UDPgal epimerase was determined by the one step procedure of
Kalckar, Kurahashi and Jordan (1959) with the addition of
0.005 M KCN (Jordan, Yarmolinsky and Kalckar, 1962).
Controls similar to those used for the UDP—gal-transferase
assay were done. The specific activity of UDPgal epimerase
was defined as the‘pmoles of UDPG formed per hour per mg of

protein.

RESULTS

Carbohydrate Utilization of
Mutants Isolated

Carbohydrate utilization patterns of some of the
mutants isolated are presented in Table 1. Few mutants give
identical reaction patterns with the three media supplemented
with galactose. The reaction patterns in Bromcresol purple
broth were the most inconsistant. Colonies producing no
sheen or dark center on EMBgal agar were considered gal
mutants and used in this study.

Strains of Galactose—Negative
Mutants Used

 

All the gal- mutants used are isolates from strain
35ara—2 as given in the materials and methods. The number
designating the strain number of ara will be disregarded.
Galactose mutants are designated by a series of numbers
identifying the mutant in the order isolated. The ara+ga1_
mutants used as sources of donor phage for transductions are

given in Table 2.

27

E!

[-1

E-I

28

 

 

 

 

m+ (VI) TU I BmH + + I I + I I mma mm
AW .6 G I + + + I I + I I mmn R
+ + + I Bm+ + + I I + I I man om
m+ .H I I .H I I I I I I I mma mm
@ + + I I + + I I + I I man am
am... I. ..
® H ® I + I + I H + I I mmo mm
@ B 9., I IH. I + I I + I I man an
m@ nu. + I + I + I I + I I mma om
«NV I. aw .I .H. I + I I .+ I I mmn ma
6 ® no I H I + I I + I I mmo ma
0v 9 ® I H I + I I + I I 93 :
m+ + + I Bm+ I + I I + I I man 6H
+ + + I I + + I + + I I maIm ma
+ + + I I + + I + + I I mmIm vH
+ + + I I + + I + + I I meN ma
+ + + I I + + I + + I I m<Im NH
+ + + I I + + I + + I I 97m 3
+ + + I I + + I + + I I mcIm 0H
+ + + I I + + I + + I I mfiIm m
AHV nHV nmv AHV flu .H + + + + + + 3mm
xmp me HuE mum Hmm wa Hue mum Ham HuE mum Hmm commas: cflmuum
mom mm mZm

 

coaumuflaflus mumuphnonumo

 

 

*.mucmuse Imumlamm Mo moHUmHumuomumnu .H wanna

29

 

 

 

 

 

 

 

 

® e G I H H I I + I I mmo mm
+ em+ + I em+ + I I + I I mmo mm
+ em+ + I Bm+ I I I + I I mmo mm
6 O 6 I 4M I I I + I :7 mun 6m
mw em+ G I + I I I + I I mmo mm
GU + + I Bm+ I I + + I SI 93 gm
Bm+ I Bm+ I I h I I + I I mmo mm
+ Bm+ + I Bm+ + I I + I I mmo mm
em+ my O I H H I I + I I man 3
@ Bm+ @ I + + I I + I I mmo om
ow % I III I I I + I I mmo 2.
am? .H I + + I .II + I H mmo ma
@ + + I Bm+ I I + + I H mmn av
.56 em+ Bm+ I Bm+ + + + + I + 93 mm
6 SE 9 I am+ I I + + I IHI mmo \Im
@ 9 + I Bm+ I H. + + I H man om
Bm+ em+ + I Bm+ + + + + I H mmn mm
Bm+ + my I Bm+ + I + + I + mmo mm
9 0 o I I + I I + I I .21 om
xmp wa Hue mum Hmm ahx Hue mum Ham Hue mum Hmm commusz cflmuum
mom mm mZM
COHUMNHHHps mpwHU>SOQHmU
Umscflucoo .H wanna

3O

 

 

 

 

 

 

 

 

H 9 @ I em+ I I H mI mI .7 man m»
GU Bm+ @ I + I I + + I El me VF
MO I Q I @ I I I + I I mmo mm
6 I w I H I I I + I I mmn N»
no aw a I + + + I mI I I man 3
no mm I GU + .II. I + I I mmn on
+ 6 6 I + + + I + I I mmo mo
9 6 9 I III H I I + I I mmo mo
6 6 6 I O H I I I I 2. an S
+ AW @ I em+ + I I + I I 93 mo
GU e + I M H I + I I 93 mo
G n t I em+ H I I + I I mmo am
Bm+ I + H H I + I I 93 mo
9 ® I @ + + I + I I mmo mo
Bm+ Bm+ % I am: + + I + I I mmn Ho
.4 Q I em+ I I I + I I mmo om

xmp aux HuE mum Hmm axx Hue mum Hmm Hue mum Hmm commas: cflmuum

mum mm mzm
COHDMNHHHDD wumupwnonumu
omscflucoo .H mHQmB

.wsan ma pummcfl mUHmCH QDOMQ .BOHHmm ma unmmcfl mo wpflmuoo zvoun M 9m “Couscoum

mmm n o “3oaamx unmfla ou moan pnmfla n H.1mmcmno o: u I “Boaaom u +IImdeu mom
“Boaawm mum
moflcoHoo UmumHOmH non wufln3 Ho pom ma xmoupm n H.“Umu u I “30Hawh u +IIummm mm
Ufloose ".2 um>fluflmcmm u m
“Hmucwo xumo nuHB muHLS mum mmflcoHoo UmDMHOmH “on cwmnm cmwum m>mn no: woe
no mmE xmwuum u H.1Hoaoo xCHm on muflnz n I “coonm comum u +IIHmmm mZM uwHOQEHm

.mmonuxmp U xmp “mmoahx

n max “Houflccme u HOE “omocflnmnm n mum “mmouomamm n Ham “moms“ Emnusp CH nuoun mamusm

l HOmeo EOHQ n mum “Hmmm pow Hocmnm n mmuummm.msafl mamawnumE.:Hmom mZM “moapflcmsmouuflcl_z
.3 IOmouuHCIzIfiwflumEIz H 02 “mcflusm OCHEm Nnm<lm “mumwasm ahnuwflp n man "mCOHuMH>mHQQ<«

 

 

 

 

 

ml ml oz maa
I mI oz boa
I I OZ em
I I mfilm mm
xmp me Hus mum Ham axx HDE mum Hmm Hue mum Hmm cmmMDSS :Hmuum
mom mm mzm

 

coHpmNHHHus wumupmnonhmo

 

 

 

wmscaucoo .a magma

32

Table 2. Ara+gal— strains used as sources of donor phage
for transduction.

 

 

 

Strain Class Strain Class Strain Class

9 K 48 T 106 E
10 K 49 T 113 E
11 K 55 T

12 K 58 T

13 K 60 T

14 K 85 T

15 K

26 K

28 K

53 K

57 K

64 K

78 K

 

33

Classification of Galactose
Mutants According to Enzyme

 

 

Deficiencies

 

Results of the assays for galactokinase, UDPbgalactose
transferase, and UDP-galactose-4—epimerase specific activities
and inducibility ratios (I) are shown in Table 3. Mutant
strains are classified as KI when their kinase activity is
less than 50% of 35W while transferase and epimerase are
present with specific activities equal to or greater than
35W constitutive levels. Strains classified as T- have spe—
cific activities of less than 0.1. Two strains, 106 and 113,
are classified as E_ since the non-induced activity is re-
duced by a factor of 19 and the induced activity by a factor
of 134. A fourth class of mutants, U, is a group of un—
classified strains.

Constitutivity of the Enzymes
in Various Mutants

 

As evident from Table 3, in all the kinaseless
mutants the I ratios of transferase and epimerase are
markedly decreased as compared to 35W. Mutant l9 lacks in—
ducibility of transferase but is constitutive for epimerase.
Varying degrees of constitutivity for transferase and epimer-
ase are apparent among the kinaseless mutants. Thus, the
mutations in the kinase gene exert a pleiotropic effect to
make the synthesis of transferase and epimerase constitutive.

This pleiotropic effect of kinase mutants is also noted in

34

 

 

 

 

 

0.0 0.0 0.0 0.00 0.00 0.0 0.0 0.0 III 01v 0.v 00 x
0.0 0.0 0.0 0.00 0.0.0 0.0 0.00 0.00 III 0v 0.v E V0
0.0 0.0 0.0 0.00. 0.00 0.0 0.0 0.0 III 0.v 0W 0.0 v0
0.0 0.0 0.0 0.00 0.00 0.0 0.0 0.0 III 00v 0. 00 x
0.0 0.0 0.0 0.00 0.00 0.0 0.00 0.0 III 0.v 00v 00 m
0.0 0.0 0.0 0.00 «.00 0.0 0.00 0.0 III 0.v 0.v 00 x
0.0 0.0 0.0 0.00 0.00 0.0 0.00 0.00 0.vn.mv. .0v 00 x
0.0 0.0 0.0 0.00 0.00 0.0 0.00 0.00 III 0.v 0.V 00 x
0.0 0.0 0.0 0.00 0.00 0.0 0.0 0.0 III 0.v. 00v 00 x
0.0 0.0 0.0 0.00 0.0 0.0 0.0 0.0 III 0.v 00v 00 x
0.0 0.0 0.0 0.00 0.00 0.0 0.00 0.0 III 0.v 0.v 00 x
0.0 0.0 0.0 0.00 0.00 0.0 0.00 0.0 0.0 00. 00. 00 x
0.0 0.0 0.0 0.00 0.00 0.0 0.00 0.0 III 0.v. 0.v 00 x
0.0 0.0 0.0 0.00 0.00 0.0 0.0 0.0 0.0 00. 00. «0 m
0.0 0.0 0.0 0.00 0.00 0.0 0.00 0.0 III 0. 0.» 00 x
0.0 0.0 0.0 0.00 0.0 0.0 0.00 0.00 III 0.v 0.v 00 x
0.0 0.0 0.0 0.00 0.0 0.0 0.00 0.0 III 00v 04v 00 x
0.0 0.0 0.0 0.00 0.00 0.0 0.00 0.00 0.00 00. 00v 00 m
0.0 0.0 0.0 «.00 0.0 0.0 0.00 0.00 0.00 00. 00v 0 x
0.0 0.0 0.0 0.00 0.0 0.0 0.0 0.0 0.0 0.0 v0. 300 0003
0mm+ HmmI H Hmm+ HMmI H Hmm+ 0mmI H 0mm+ 0mml cflmuum mmMaO
m\B mmmumEHQm mmmummmcmue mmmcflx
Acawuoum mE\0£\uUspoum mmHOE 1V
mw000>fluom UHMHmem mexucm
.mucmusE
Iammlmum mo mmemucw mmouomamm mo wuHHHQHUSUCH paw coflumflnowwp 00m>uocm£m .m GHQMB

35

 

 

 

 

 

0. 0. 0 0 00. 00. 0.0 0.0 0.0 0.0 00. 00. 000 m
0.0 00. 0.0 00. 00. 0.0 0.0 0.0 0.0 0.0 00. 000 m
.050
0.0 «.0 0.0 III 0.v 0.v 0.0A 00. 0.V 00 a
«“0 0.00 0.00 III 0.v 0.w. 0.00A v.0 0.v 00 a
0 000A 0.00 0.v III 0.v 0. 0.0A 00. 0.v 00 a
0.0 0.00 0.00 III 0.w 0.v. 0.00 00. 00v 00 a
0.0 0.00 0.00 III 00\ 0uv. III 0.v 0.v 00 a
0.0 0.00 0.0 III 00v 0.v III 0.v 0.v 00 a
00. 0. 0.0 0.00 0.00 III 00v 0..V 0.0 00. 00. 00 0
III III 0.0 0.00 0.0 III 0.v. 0.v 0.0A 00. 00v. 00 0
III III 0.0 0.00 0.0 III 0.v 0.v 0.o« 00. 00v. 00 a
m}
0.0 0.0 0.0 0.00 0.00 0.0 0.0 0.0 III 0.v 00v 000 x
0.0 0.0 0.0 0.00 0.00 0.0 0.0 0.0 III 0mv 0.v 00 m
Hmm+ Hmml H Hmm+ Hmml H Hmm+ HmmI H Hmm+ Hmml :Hmuum mmMHO
m\e ommnmEHmm wmmummmcmue mmmsflx
Achuoum mE\0£\u05©oum mmHOE AV
mmHuH>Huom UHHHUmmm oewncm
000000000 .0 00000

36

.0.0 I.mu
mH :oHpMH>mU UHMUCMDm moo 00H mHMmmm mEHNcw mmmcHx 0:» mo 00000 Unmpcmuw one

.mmmfl m mm 3mm
00 Uwumzflpm mmHuH>Huow UHHHUmmm wmmumEHmw ou wwmummwcmuu mo 00000 opp mH m\e

.UmoschIcoc on pwodpcfl .m0000>0000 Ho 00000 may 00 H

.mCOHumumcmm

 

 

000:0 How omouomHmm 2 NIOH mo mmcmmnm Ho wocomwum on» CH £u3onm ou mewn I paw +
0.0 0.00 0.00 0.0 0.0 0.0 0.0 0.0 00. 00 0
0.0 0.00 0.00 0.0 0.0 0.0 0.00v 0.v 00. 00 0
0.0 0.00 0.00 0.0 0.0 00. 0.0K 00. 0.v 00 0
0.0 0.00 0.00 0.0 0.00 0.00 0.0 00. 00. 00 0
0.0 0.00 0.00 0.0 0.0 0.0 0.0 00. 0.0 00 0
0.0 0.0 0.0 0.0 0.0 00. 0.0V 0.v 00. 00 0
0.0 0.0 0.0 0.00 0.0 00. 0.0A 00. 0.v. 00 0

Hmm+ HmmI H Hmm+ HmmI H Hmm+ HmmI H Hmm+ Hmml GHmem mmmHO

M\B wwmumEHmm wmmummmcmue wwwcHM

 

Achuoum mE\0£\uospoum wwHOE 1v
mmHuH>Huom UHHHummm mEHNcm

 

 

 

000000000 .0 00000

In th
kinas
Trans
induc
m‘ItaI

gene

‘fz?
58 PI
)

.« /..

/”E‘/

.0}
(f)

37

g. 9911 (Kalckar, Kurahashi, and Jordan, 1959; Jordan,
Yarmolinsky, and Kalckar, 1962) and in 9. typhimurium
(Fukasawa and Nikaido, 1961b).

With the exception of mutants 50 and 55, a mutation
in the transferase locus does not affect the inducibility of
kinase; however, the specific activity levels may be reduced.
Transferase mutations can have no effect or can cause non-
inducibility of epimerase. Moreover, some transferase
mutants are constitutive. Thus, mutations in the transferase
gene may also exert pleiotropic effects.

Although only two epimeraseless mutants are repre—
sented, they exhibit different characteristics. Strain 106
is partially constitutive for transferase, but kinase is in-
ducible. The mutation in the epimerase locus of strain 113
does not seem to affect either kinase or transferase. Two
mutants are not enough to support definite conclusions, but
there is an indication that a mutation in the epimerase gene
can also exert a pleiotropic effect on another galactose
sequence gene.

Transduction Frequencies of
P38/35WrDonor Phage with Selected
Ara Gal Recipients

Three kinase and 3 transferase mutants were selected
as recipients for transduction by P38/35W. A small survey
was performed to determine possible effects of various re—

cipients upon the frequency of transduction. Results of the

Ylhu "Id

38

survey are shown in Table 4. Galactose frequencies range
from 0.9—1.5 x 10_5, and arabinose frequencies range from
1.3-1.7 x 10’5 indicating little effect of the selected

hosts upon the frequency of transduction.

Nomenclature (Hartman, LOper and Serman, 1960)

 

The following terms will be used in this thesis:
(1) gene, a discrete section of the chromosome governing the
presence of a specific protein molecu1e(s); (2) locus, or
gene locus, the place on the chromosome occupied by a gene
in any of its forms, wild—type or mutant; (3) allele, an en—
tire gene, regardless of its exact configuration (wild-type
or any one of a number of mutant forms); (4) site, a region
of a gene, altered in the mutant, which can recombine with
adjacent sites.
Use of the Comparative Recombi-
nation Frequencies of the Kinase,
Transferase, and Epimerase Loci

According to Hartman, LOper and Serman (1960), the
nature of transduced material and the phenomenon of inter—
ference preclude the assignment of classical recombination
values for the histidine region of Salmonella typhimurium.
Moreover, the interference effect renders the results of a
two-point test ambiguous. The authors express comparative
recombination frequencies in the histidine region by the use

of average values for a single mutant in tests with several

39

.soH0USUwc000 £000 000 00000050
Imc000 00 00085: 030 8000 0000000050 0003 0mm:& msomoHOEo: £0H3 00000 H000soo

.0m0nm 0smcH mOH x H\m0s00ospmc000 mo .0: u >000sw00m .000CSOU 0003
0000Hm m 00 H0000 0 0cm 000Hm 00m ©00nmm 003 0050xHE coH0ospmc000 0:0 00 HE H.o*

 

 

 

 

m.H m.H ¢.H m.o 0.H ¢.H m.H H.H m.H H.H m.H N.H IQH x H
MUC0SU0HH
mow ovm mmm 00m 000 00m mmm mmm mmm mom mmm mmm mm3\mmm Hocoa
000 H00 000 H00 000 H00 000 H00 000 H00 000 H00 00M002
mm m0 om mm 0H m 0:0HmHo0m
a M

 

 

*.mm3\mmm >0 00C00se IE 0:0 LM mo >0c050000 coH0oswmc0HB .0 0H008

40

other mutants closely linked to one another as a means of
circumventing the above mentioned difficulties.

Several mutants of a locus (A) are used as donors to
one mutant in another linked locus (B), and the average
value (p) is the mean probability of independent integration.
The procedure is repeated using the same donors in tests
with several other mutants of locus B and averaging the p
values obtained. Mean p values represent the distance be-
tween an average point in locus A to an average point in
locus B. Transducing from locus A to a third linked locus
(C) and from B to C should represent the average distance
from A to B and B to C respectively. Hence, it is possible
to order the loci with respect to each other. This procedure
tends to randomize the deviations due to individual recombi-
national behavior.

Results from such an experiment are presented in
Table 5. An order K—T—E is assumed to serve as a model.

The mean p value from K to T (mean pKT) is 0.254 and from E
to T (mean pET) is 0.241. Since the mean average probabili—
ties of independent integration are approximately equal, K
and E are equal distances from T. Therefore K and E must be
on Opposite sides of T. The mean p value from T to K (mean
pTK) is 0.078 and from E to K (mean pEK) is 0.364 indicating
T is closer to K than B. Consequently, the order K—T-E is
probably correct. However, the data does not differentiate

between the order KrT—E or E-T-K.

41

.00000 CH U0NHHH05 msooH 0:0 00 00H0HH0 0C00000H© mo 00QECC 0C0 00000H0CH .oz

 

 

00m. who. HvN. va. . C002
How. N OOH. N HmIM mom. N omo. 0H mmI
mmm. N omo. N mHIM mMN. N NvN. 0H mol
Hmm. N mmo. N NHIM mom. N vON. 0 mmI
0Nm. N 000. N mHIM NHm. N @HH. 0 mmI
ONm. N omo. N vHIM woo. N 000. o omI
000. N HHH. N NHIM mmN. N mmH. 0H mvl
mwm. N 000. N OHIM mHH. N mum. mH ONIB
m. .02 m. .02 0C0HmH00m m. .02 m. .02 0C0HmHo0m
m E m M

HooH HOCOQ

 

 

.muoCop 00 HooH m 0C0 B .M 00 00C00CE mCH>Ho>CH 00C0HQH000 Lx C0>0m
0C0 IE C0>00 How C3000 0H Amy COH000®00CH 0C0©C0m0oCH mo >0HHHQ0QOHQ C002 .0 0H008

42

If comparative recombination frequencies eliminated
effects on recombination due to individual alleles, mean
pKT would be expected to equal mean pTK. Observation of a
difference of 0.176 indicates not all the effects are elimi-
nated. This result can be partially explained since only
two T_ donors were available and are not considered to be a
good average representation for comparative mean recombination
frequencies.
Use of the Ratio Test to Order
Individual Mutant Sites within
the UDP—qal Transferase Locus

A mean probability of independent integration (mean
pABl) between locus A and a specific mutant (B1) in the
linked locus B represents the distance between an average
point in locus A to the specific mutant in locus B. A simi-
lar distance (mean EcBl) from a third locus C linked to the
B locus, where B lies between A and C, can be obtained. The

ratio of mean pABl/mean pC indicates the relationship of

B1
B1 to A and C. Repetition of the procedure with other mutants
of locus B and comparison of ratios is presumed to give an
indication of the relative order of mutant sites within the
B locus.

All recombination values can be adjusted to common
terms by the ratio (mean pA/mean pC)/(l + mean pA/mean pC)

which equalizes the relative map distance from the average A

mutant.

43

The data of Table 6 are from individual tests of 7 T
mutant recipients with 14 K and 2 E donors. The data for K
donors 9—15)< T recipients 20, 48, 69, and 85 represent the
average of 3 experiments, and data for K donors 26, 28, 53,
54, 57, 64, and 78 x T recipients 50, 55, and 58 represent
the average of 2 experiments. Data for K donors 9—15 X T
recipients 50, 55, and 58 and E donors 106 and 113 x T re—
cipients 20, 48, 50, 55, 58, 69, and 85 are from single
experiments. Variation between experiments generally did
not exceed 35% of the value shown.

Table 7 presents a summary of the data of Table 6.
The mean probabilities for recombination separating T mutants
from the average K and E mutants are presented in the columns
headed mean pK and mean pE respectively. The ratio of these
two values is presumed to give an indication of the order of
mutant sites (column headed Ratio: pK/pE). Column five is
the equalized relative map distance from the average K mutant.
The map obtained is depicted in Fig. 1.
Use of the Ratio Test to Order
Individual Mutant Sites within
the Galactokinase Locus

An important feature of the ratio test, as applied
to the UDP-gal transferase locus, is that a specific mutant
site is mapped from opposite directions in relation to aver—
age points of two adjacent loci. Individual recombinational

behavior is assumed to be significantly reduced by this

44

 

 

 

 

000.0 000.0 000. 000. 000.0 000.0 000.0 0.00.
Hmm.o omN.o Hmm. Nom. mOH.o HON.O 00H.o mHHI
NMN.o 0NN.o 00m. moN. 000.0 vhN.o 000.0 QOHIm
000.0 NON.0 HON.o oHH.o 000.0 mmH.o 00m.o mm.
OmH.o NoH.o 00H.o mnl
00H.o Nmm.o mNm.o 000.0 wol
000.0 mmN.o omo.o NNm.o hmI
000.0 vom.o NHH.o NHv.o wmI
Nmo.o HmN.o mmo.o mmN.o mmI
hmo.o mmH.o OHo.o mHN.o mNI
hNo.o NNN.o mmo.o mm¢.o mNI
mmH.o mmH.o oHN.o mvH.o mwv.o mmN.o mam.o mHI
000.0 mmH.o mmH.o mno.o Hmm.o HNN.o oom.o HHI
000.0 00H.o mmH.o 00H.o MNm.o HmN.o o0m.0 MHI
©ON.o Nom.o noN.o NNN.o o¢m.o oom.o mmm.0 NHI
HoH.o moN.o HmH.o 000.0 000.0 mmH.o 00m.o HHI
HNo.o mmH.o HOH.0 mwo.o mnm.o HmN.o 0mm.o OHI
000.0 th.o Nom.o mNH.o III HON.o mmv.o mIM
mmle moIB mmIB mmIB omIB meB ONIB 000COQ
00C0HmHo0m
.000C00 m N
0C0 Lx 0H mCH>Ho>CH 00000 CH 00C0HQH00H B C0>00 How C3050 000 Amv C0H0000M0CH
0C0UC0Q0GCH Ho >0HHH00000m 0:0 0C0 m0N>H0C0 0C0H00300C0H0 mo 00QECC H0009 .0 0H008

45

 

 

000.0 00.0 000.0 000.0 00:
000.0 00.0 000.0 000.0 00:
000.0 00.0 000.0 000.0 00:
000.0 00.0 000.0 000.0 00:
000.0 00.0 000.0 000.0 00I
000.0 00.0 000.0 000.0 00:
000.0 00.0 000.0 000.0 00:0
00\xm+0 mn\xn. "00000 0000002 0000052 000002
mm.c002 xn.c002

 

mm\xm. "000mm

 

 

 

.E5000050 0000008000 00 00000 B 000
000003 00000 000056 00000000000 :0 005 00000 000 0 00009 00 0000 00 0008850 .0 00008

46

 

.ESHOHHDQXMHHwCOEHmm
mo msooH Ammmuwwmcmuu mmouomammlmnbv B may CHSHHB mucmuse
b m0 mcofiumooa m>HumHmH wnu mCHBOSm cam mamom op c3muw mm:

 

 

.H

005

 

N X VA VAIIINIIIXIIVA
om ON . mo mfl‘ mm mm mm

 

m E

 

47

approach. Ordering of mutants within the kinase locus is not
possible by this method because the adjacent locus on one
side of the kinase locus is not known. However, an analysis
to minimize individual recombinational effects of mutant
sites can be done by approaching kinase sites from 2 loci on
the same side.

Mutant site Al lies within locus A which is linked to
loci B and C in the order A-B-C. The distance from an aver—
age point in.B to Al (mean pBAl) and the distance from an
average point in C to Al (mean EcAl) is determined by trans-
ductional analysis. The ratio of the distances mean pBAl/mean
pCAl, is presumed to give an indication of the relative order
of Al within the A locus when compared by the same method to
other A mutant sites.

Negative interference effects are assumed to be less
distance since

for the mean pC distance than the mean pB

Al Al

the mean 5c distance is greater. Where negative inter-

Al
ference has been investigated, the effect seems to be re-

duced the further apart two mutant sites are separated.

Therefore, individual recombinational effects present in the

mean pB should be moderated in the ratio mean pBAl/mean

Al

EcAl by the dampening effect of the mean pCAl distance.
Negative interference phenomena are not assumed to

be reduced to the same degree as when a mutant site is ap-

proached from opposite directions. Therefore, the order of

sites determined for the galactokinase locus is not as

48

reliable as those determined for the UDP-gal transferase
locus.

Table 8 represents data for individual tests of 19 K
mutant recipients with 2 T and 2 E donors. The data for T
donors x K recipients is the average from two experiments,
and the data for E donors X K recipients is from one
experiment.

Table 9 is a summary of the data from Table 8. The
mean probabilities of independent integration separating K
mutants from the average T and E mutants are presented in
the column headed pT and pE respectively. The ratio of
these two values (column headed Ratio: pT/pE) is used to
give an indication of the order of mutant sites within the K
locus.

A composite map of the most probable order of mutant
sites within the kinase and transferase loci is given in
Fig. 2. Although no attempt is made to order the epimerase—
less mutants 106 and 113, inspection of tables 6 and 9 indi-
cates that 106 is probably closer to the transferase locus
than is 113.

Theoretical Aspects of the
Linearity of the Galactose Loci

 

 

A diagram of the distances under investigation is
depictedwon p.52.fmu3ratio pTKl/pEKl is the ratio used to

order mutant sites within the K locus as described above.

49

 

 

oom.o Hmm.o 0mm.o Ho¢.o ham.o mom.o 5mm.o mvm.o oa¢.o mm.
oom.o nmm.o mao.o mov.o mmm.o oHv.o mom.o 0mm.o mm¢.o mHHI
wmm.o mwm.o mwm.o Nmm.o Hom.o nmm.o ham.o oam.o omm.o ooalm
mno.o nmo.o Hom.o ooa.o mmo.o oo.H moo.o mmo.o ooa.o em.
mmo.o «mo.o mmm.o mvo.o mao.o vmo.o wmo.o who.o mmo.o mml
mwo.o ooa.o mom.o omH.o nmo.o moa.o hwo.o mmo.o NHH.0 mole
omIM whim fioIM nmLM mmim Nmix mNLM omix mmix
mmm.o Hmm.o mum.o 0mm.o omm.o mmm.o o¢¢.o mom.o mom.o mmm.o mm.
mmm.o mmm.o mmm.o oom.o vcm.o wwm.o va.o mom.o mmm.o 5mm.o maal
mam.o mom.o mmm.o 0mm.o mom.o Nmm.o mm¢.o mmm.o mmm.o mmm.o moaum
omo.o mmo.o omo.o ovo.o omo.o Hwo.o HBH.O moo.o vwo.o owo.o Em.
Hoo.o «mo.o ovo.o mmo.o Hmo.o omo.o oom.o mmo.o mmo.o mmo.o mm:
wmo.o mma.o Hma.o ovo.o Hmo.o omo.o mma.o moo.o wmo.o mmo.o mvle
mHIM balm calm malm calm mHIM NHIM HHIM OHIM mum muocon
mucmflmflomm

 

 

.mHOCOp um N
paw IE m GCH>HO>CH mummy CH mucwflmflowu LM ma Mom c3onm mum Amy coflumumwucfl

pampcmmmUCH mo wufiaflnmnoum on“ com Umuhamcm mcofluospmcmuu mo nomad: Hmuoa

.m mHQMB

50

 

 

 

 

Table 9. Summary of data in Table 8 and its use in ordering
mutant sites within the K region of Salmonella
pullorum.

Mean ET Mean EB ._

Mutant Mutants Mutants Ratio: pT/pE

Ké53 0.023 0.317 0.073

-14 0.036 0.320 0.113
-9 0.040 0.335 0.119
—13 0.041 0.335 0.122
-15 0.040 0.320 0.125
—19 0.050 0.339 0.147
-26 0.056 0.345 0.162
-11 0.063 0.349 0.181
-10 0.064 0.349 0.183
-28 0.068 0.357 0.190
-86 0.072 0.360 0.200
—17 0.083 0.381 0.218
—16 0.086 0.379 0.227
-77 0.097 0.391 0.248
-57 0.100 0.401 0.249
-52 0.100 0.399 0.251
-23 0.106 0.410 0.259
—12 0.171 0.440 0.389
-64 0.561 0.580 0.967

 

51

 

.EsuoHasm.maamCOEHmm mo
mucmuze coflmwu mmouomHmm mo Hmpuo o>HumeH wHQonum umoE wag mo mm: .m .mflm

.prHEHmumU uoc mH mumxomun CH mucmuoe mo Hmpno one

 

817 ‘
89
ss
98
£9
SI
61
9a
II
OI
8z

VT
6
ET

09

OZ
69
98

LT

91
LL
LS
ZS
EZ
ZT
t9

 

 

 

'901)

(EU

 

52

 

 

 

K1

L _ 4

'F 7

I _ ..

r—pTKl ——+— PE. —1
I I

1 L‘ J J | . l

T

A point K1 in the K locus is moved away from the T
locus by constant increments, and the ratio pTKl/pEKl is

plotted as a function of pT The resulting plot is non—

Kl'
linear and approaches 1 as a limit. Although the relative
order of mutant sites within the K locus is maintained, the
method introduces non-linearity into this order.

The distances depicted in the diagram can be trans-
formed into an equation for a straight line. Let X = pTKl,

A = pET which is assumed to be constant when ordering mutant

sites within the K locus, and C = pEKl. Then C = pET + pT

 

 

Kl'
Let the function (f) =-z. It follows that f = X and
C X+A
X = f(X+A). By transposing, X = f(X+A) into-% =-§§A, the
equation of a straight line-% =-% A + l is obtained with an
ordinate intercept of l and a slope of A.
l . l .
A plot of :——-:—— as a function of _ is repre-
pTK/pTE pTK

sented in Fig. 3. It can be seen that a straight line with

53

msu mcHCHEkump aw mucmuse mmmcflx

MBm
H

0m

«1—

(#0

.m ou H Eoum mocmumap
mafiuwono Eouw mump mo mm:

om

ON

.m

.mam

 

54

an ordinate intercept of l and a slope A of 0.292 is obtained.
The predicted value of 0.292 for A, which is the distance

pET, is in close agreement with 0.241 obtained by trans-
ductional analysis (see Table 5). Agreement of the theoreti—
cal prediction with experimental results indicates that the
assumptions used in mapping are probably correct.

Attempts to Transduce Gal-
Epimeraseless Mutants to Gal

 

All attempts to transduce galactose utilizing ability
from 35W, 35ara+ga1-9, 35ara+gal-15, 35ara+ga1-48, and 35ara+
gal-85 to 35ara-ga1—106 and 35ara-ga1—1l3 failed. Epimerase-
less mutants of g. typhimurium and §°.22ll: grown in the
absence of galactose, contain glucose in their cell walls
but no other detectable carbohydrates (Nikaido, 1961;
Fukasawa and Nikaido, 1961a). Addition of galactose to a
culture during the exponential phase of growth initiates
severe bacteriolysis after 30-60 min. However, addition of
glucose can completely repress the action of galactose, and
galactose will be present in the cell wall.

Epimeraseless mutants of §. typhimurium, grown in
the absence of galactose, are resistant to phage P22. Epimer-
aseless mutants regain their sensitivity to P22 and support
phage multiplication when grown in the presence of 1% glu-
cose and 0.1% galactose (Fukasawa and Nikaido, 1961b).
Mutants 106 and 113 will support multiplication of phage P38

with or without galactose in the media.

55

When mutants 106 and 113 were used as recipients
they were grown in M—9 broth, with or without galactose, and
nutrient broth, with or without galactose. Plaques were
visible on the transduction plates regardless of the growth
medium used. Consequently, 106 and 113 must have adsorbed
P38, either during the incubation period allowed for phage
adsorption or after the mixture was plated. Resistance to
P38 can be eliminated as a reason for lack of transduction
of galactose utilization. Furthermore 106 and 113, plated
on TTCara agar or on nutrient agar supplemented with 2,3,5—
triphenyl tetrazolium chloride and arabinose, were trans-
duced from ara“ to ara+ at a slightly reduced efficiency
than reported in Table 4. Mutants 106 and 113 are extremely
sensitive to EMBgal agar and revert to gal+ at an extremely
high rate (l—2 x 103 revertants per 1 x 108 cells plated).
Observation of transductants using EMBgal agar would be very

difficult due to the high reversion rate.

DISCUSSION

The pleiotropic effect of a mutation in the kinase
gene upon the transferase and epimerase genes is apparently
a complex phenomenon. All of the 21 kinaseless mutants
studied exhibited some effect upon transferase and epimerase
in one form or another. It has been proposed that kinase
acts in some manner by which it removes an endogenous in-
ducer° Therefore, cells with functional kinase would re-
move the endogenous inducer and subsequently be inducible in
the presence of an exogenous inducer (galactose or fucose).
A defect in kinase would permit accumulation of an endogenous
inducer allowing apparent constitutive synthesis of trans-
ferase and epimerase (Kalckar and Sundararajan, 1961; Jordan,
Yarmolinsky and Kalckar, 1962; Jordan and Yarmolinsky, 1963).

Buttin (1961) reported that methyl-JQ-D-thiogalacto-
side (MTG), a gratuitous inducer for.£?-galactosidase, re-
presses induction of the galactose enzymes. Constitutive
synthesis of transferase and epimerase associated with a de-
fective kinase gene is also inhibited by MTG, but it can be
reversed by high concentrations of galactose (Jordan,
Yarmolinsky and Kalckar, 1962).

The observed constitutive synthesis of transferase

and epimerase would be the result of an 0C type mutation.

56

57

An 0C mutation is presumed to prevent a repressor molecule
from inhibiting transcription of DNA. Therefore, MTG should
have no effect on the constitutive synthesis of kinase and
epimerase. Since MTG is observed to repress the constitutive
synthesis of transferase and epimerase, a mutation in the
kinase locus can not be an 0C mutation.

Inhibition of constitutive synthesis of transferase
and epimerase by MTG was not attempted with S. pullorum
since the experiment would probably be more complex than
when performed with g. 291;. Salmonella pullorum does not
utilize lactose and probably has a deletion for at least
part of the lactose operon. No spontaneous backmutants have
been observed in this laboratory. In order for MTG to be
repressive it must enter the cell and would require lactose
permease. At present, it is not known if g. pullorum con-
tains lactose permease in its genome.

Six kinase mutants (9, 10, l4, l6 and 28) are ap-
parently partial mutants with reduced kinase activity. If
the endogenous induction hypothesis is correct, a 50% re-
duction in kinase activity is sufficient to impair removal
of the endogenous inducer and to produce at least partial
constitutive synthesis of transferase and epimerase.

If the gal sequence enzymes are under the control of
one operator, all three enzymes should be coordinately in-
duced. Fucose, a gratuitous inducer for the gal sequence,

induces kinase, transferase and epimerase coordinately in

58

g. 221; (Buttin, 1961). Coordinate induction should be
reflected in the ratios of the non—induced and induced spe-
cific activities of the three enzymes when adjusted to common
terms. Induction of the gal sequence should increase the
specific activities in the same proportion as they were
present when in the non—induced state. Induction should in-
crease the amount of DNA transcription of the galactose
operon. If there is maximum fidelity and efficiency in
translation, all three gal enzymes should be increased to

the same degree. The columns headed T/E in Table 3 represent
the proportional content of transferase and epimerase in
kinaseless mutants when in the non-induced and induced state.
The ratio T/E has been adjusted so that 35W values are equal
to one. Comparison of the T/E ratios of induced and non—
induced kinase mutants, with the exception of mutants 10, 12,
15, 26 and possibly 13, 28 and 57, indicates the gal sequence
in S. pullorum is under the control of one operator.

The ratio K/E was determined only for transferaseless
mutant 49 since the kinase activities indicated as less than
0.1 would make the ratio meaningless. Mutant 49 appears to
be under the control of one operator. The K/T ratios calcu-
lated for the epimeraseless mutants indicate one operator for
mutant 113, but 106 may possibly have more than one operator.

Mutations in the transferase locus of 5. 291i and S.

typhimurium do not affect the inducibility of kinase and

 

epimerase (Kalckar, Kurahashi and Jordan, 1959; Jordan,

59

Yarmolinsky and Kalckar, 1962; Fukasawa and Nikaido, 1961b).
Mutant 48 of S. pullorum is similar to those found in E. coli

and S. typhimurium.

 

The apparent pleiotropy observed by a mutation in
the transferase locus may have several explanations. Mutant

69 is also similar to E. coli and S. typhimurium transferase-

 

less mutants with respect to kinase, but epimerase may be
constitutive. The epimerase I value of 2.4 is very close to
the value 2.5 used by Jordan and Yarmolinsky (1963) as the
dividing point between inducibility and constitutivity.
Mutant 69 is probably constitutive since the non-induced epi—
merase level is 6 units above the non-induced level of 35W.

If an Operator governing the gal locus is located at
the extremity of the epimerase locus as Buttin (1963b) sug-
gests, then a mutation in the T locus could interrupt the
polorized transcription of DNA. Jacob and Monod (1961) sug-
gested that interrupted polorized transcription occurred in
the lactose system. Transferaseless mutants 20, 49, 50 and
60 may fit into this category. The nine mutants classified
as transferaseless were isolated after induction with DES.
Diethyl sulfate is a depurinating agent (Freese, 1963) which
deletes purines from the DNA molecule and could produce
mutants such as those described above.

Mutants 55 and 58, which lack transferase and kinase
but are constitutive for epimerase, may be double mutations.

Galactose initiates bacteriostasis in E. coli transferaseless

6O

mutants when grown in a synthetic medium. The bacteriostasis
can be almost completely overcome by the addition of yeast
extract or Casein hydrolysate to the medium (Kurahashi and
Wahba, 1958). The bacteriostatic effect of galactose upon
transferaseless mutants is connected in some manner with the
gal-l—p which they accumulate (Kalackar, 1958; Sundararajan,
1963). A second mutation occurring in the kinase gene would
prevent accumulation of gal—l-p and relieve the bacterio-
stasis. These conditions provide a very efficient method
for selection of strains mutant in both the kinase and trans-
ferase genes (Kalckar, Kurahashi and Jordan, 1959; Soffer,
1961). Since all transferaseless mutants were initially iso-
lated and subcultured on EMBgal agar, mutants 55 and 58

could readily be double mutants.

Confirmation that 55 and 58 are double mutants could
be confirmed by complementation tests using an Fgal particle.
If 55 and 58 are single mutations in the transferase locus,
then an Fgal K-TOE particle should complement the bacterium
(KT—E0) when in a partial diploid state (FgaleTEO/KT‘EO).

A positive complementation test would confirm that the muta-
tions of 55 and 58 are confined to the transferase locus.

Mutant 85 could conceivably be an 00 type Operator
mutation. The low levels of kinase and transferase are
analogous to the levels found in g. ggli (Kalckar, Kurahashi
and Jordan, 1959; Jordan, Yarmolinsky and Kalckar, 1962) and

g, typhimurium (Nikaido and Fukasawa, 1961), but the level

61

of epimerase is high (2.4 units for S. pullorum, 0.9 units or

less for E. coli and 0.05 units for S. typhimurium). The

 

higher epimerase content in strain 85 may be due to the
method of protein determination. The cell free extracts pre—
pared for enzyme analysis are probably of lower protein con-
tent for two reasons: (1) the extracts were centrifuged at
106,000 x G for 1 hr, and (2) the protein content was de-
termined from the precipitate formed after addition of TCA
(10% final concentration).

Mutant 85 maps in the extremity of the transferase
locus closest to the kinase locus. Morse (1962) found the
00 mutants to map between the K and T loci, and Fukasawa and
Nikaido (1961b) determined 00 to be closer to the K locus
than to the E locus.

The 00 mutants described in g. £21; and S. typhi-
murium are single site mutants as indicated by their back-
mutation rate. Strain 85 of §. pullorum also reverts to
gal+ indicating a single site mutation.

Complementation tests could distinguish whether 85
is 00 or a true transferaseless mutant. The 00 mutants so
far described complement with none of the three gal sequence
loci.

The 00 mutants mapped by Morse (1962) are the same
mutants Adler and Kaiser (1963) found to locate in the E
locus or between the T and E loci. However, the majority of

the other mutants used in their experiments were not identical.

62

The proposal of Buttin (1963b) for 00 mutants found in g.
3913 K12 will not explain an 00 mutant mapping in any location
other than in the E locus. The possibility of two operators
governing the galactose sequence must still be considered.

The T/E ratios presented in Table 3 indicate that
kinaseless strains 10, 12, 15, 26 and possibly 13, 28 and 57
may not be controlled by a single operator. The map con-
structed in Fig. 2 shows a tendency of kinaseless strains 10,
13, 15, 26 and 28 to cluster. Kinaseless mutant 19 is con-
stitutive for epimerase but is non-inducible for transferase.
The above factors may support the presence of more than one
operator for the galactose sequence in g. pullorum. Further
experiments are required before any definite conclusions can
be made.

Assignment of order is always attended by a measure
of uncertainty, and this is necessarily so with the map con—
structed for the gal region of S. pullorum. The positioning
of loci (T between K and E) is in agreement with the order
established for g. 221; (Buttin, 1963b).

If comparative recombination frequencies eliminated
effects on recombination due to individual alleles, the pKT
(0.254) should equal the mean pTK (0.078). Theoretically,
the mean probabilities of independent integration are from
an average mutant in the donor locus. The two transferase—
less mutants (48 and 85) that were available as donors are

not a good average representation of the transferase locus

63

since they probably represent only the one-third of the
transferase locus closest to the kinase locus. Furthermore,
mutant 85 may actually be an 00 mutation. These factors can
affect the order of mutant sites within the K locus.

Epimeraseless mutants 106 and 113 are not amenable
to transduction though they will serve as sources of donor
phage. Since epimeraseless mutants of g, £21; and S.

typhimurium can be transduced, 106 and 113 may be of a

 

peculiar type and not representative epimeraseless mutants.
Transduction tests with newly isolated epimeraseless strains
may affect the order of mutant sites within the K and T loci
or provide more validity to the interpretation of the data
so far obtained.

The theoretical distances employed to map the K
locus can be transformed into the formula for a straight
line and can be used to predict the distance from the aver-
age T mutant to the average E mutant. The data obtained from
mapping the kinase mutants were used in the formula to de—
termine if a straight line with an ordinate intercept of 1
followed. The slope of the straight line obtained (0.292),
representing the theoretical distance pET, was in reasonable
agreement with the distance determined by the ratio test
(0.241).

Considering the agreement of the calculated and
experimentally determined distance pET, it can be concluded
that, despite the uncertainties mentioned above, the data

presented have some measure of reliability.

SUMMARY

Galactose—negative mutants of Salmonella_pullorum

 

were isolated and their enzyme deficiencies investigated.

The mutants were classified as follows: (1) 21 kinaseless
mutants, (2) 7 transferaseless mutants, (3) 2 epimeraseless
mutants, and (4) 1 possible 00 mutant which could be a trans—
feraseless mutant. The pleiotropic effect of kinase
mutations upon the regulation of transferase and epimerase,

previously observed in E. coli and g. typhimurium, was also

 

observed with_§. pullorum. Apparent pleiotropic effects re—
sulting from transferase mutations are discussed. Evidence
was obtained indicating that the galactose region of s.
pullorum may be under the control of more than one operator.
Transduction, using phage P38, was employed in the
ratio test to map the kinase and transferase loci. An
arabinose marker, unlinked to galactose, was used as an in-
ternal control. The relative order and relative distances
were determined for transferaseless mutants within the trans—
ferase locus. A modification of the ratio test was used to
determine the relative order of mutants within the kinase
locus. The two epimeraseless mutants isolated served as
sources of donor phage but were not susceptible recipients.

No attempt was made to order the epimeraseless mutants.

64

65

The theoretical relative map distances employed in
the modified ratio test were transformed into the formula
for a straight line. Data obtained from the relative order
of mutant sites within the kinase locus were used in the
formula. The slope of the resulting straight line predicted
the relative distance from the average epimerase mutant to
the average transferase mutant. The predicted distance was
in reasonable agreement with the value determined by trans-

ductional analysis.

BIBLIOGRAPHY

Adams, M. H. 1959. Assay of phage by agar layer method,
p. 450-451. In M. H. Adams, Bacteriophages. Inter-
science Publishers, Inc., New York.

Adler, J., and A. D. Kaiser. 1963. Mapping of the galactose
genes of Escherichia coli by transduction with phage-
Pl. Virology 19:117—126.

Arber, W. 1958. Transduction des caracteres gal par le
bacteriophage Lambda. Arch. Sci. (Geneva) 11:259—338.

Buttin, G. 1961. Some aspects of regulation in the synthe-
sis of the enzymes governing galactose metabolism.
Cold Spring Harbor Symp. Quant. Biol. 26:213-216.

Buttin, G. 1963a. Mecanismes regulateur dans la biosynthese
des enzymes du metabolisme du galactose chez
Escherichia coli K12, I. La biosynthese induite de
la galactokinase et-l'induction simultanee de la
sequence enzymatique. J. Mol. Biol. 1:164-182.

Buttin, G. 1963b. Mecanismes regulateurs dans la biosynthese
des enzymes du metabolisme du qualactose chez
Escherichia coli K12, II. Le determinisme genetique
de la regulation. J. Mol. Biol. 1:183-205.

Buttin, G., F. Jacob, and J. Monod. 1960. Synthese con-
stitutive de galactokinase consecutive au developpe—
ment des bacteriophages LAchez Escherichia coli K—12.
Compt. rend. 259:2471-2473.

 

Echols, H., J. Reznichek, and S. Adhya. 1963. Complemen-
tation, recombination, and suppression in galactose
negative mutants of E. coli. Proc. Natl. Acad. Sci.
U.S.I§Q:286-293.

Freese, E. 1963. Molecular mechanism of mutations, p. 207-
269. _ln J. H. Taylor (ed.), Molecular Genetics,
part I. Academic Press, Inc., New York.

French, D., D. W. Knapp, and J. H. Pazur. 1950. Studies on
the Schardinger dextrins. VI. The molecular size
and structure of the ‘ULdextrin. J. Am. Chem. Soc.
12:5150-5152.

66

67

Fukasawa, T., K. Jokura, and K. Kurahashi. 1962. A new
enzymatic defect of galactose metabolism in
Escherichia coli K-12 mutants. Biochem. Biophys.
Research Comm. 1:121—125.

 

Fukasawa, T., and H. Nikaido. 1959a. Formation of "proto—
plasts" in mutant strains of Salmonella induced by
galactose. Nature 183:1131—1132.

Fukasawa, T., and H. Nikaido- 1959b. Galactose-sensitive
mutants of Salmonella. Nature 184:1168-1169.

Fukasawa, T., and H. Nikaido. 1961a. Galactose-sensitive
mutants of Salmonella, II. Bacteriolysis induced by
galactose. Biochim. et Biophys. Acta 48:470-483.

Fukasawa, T., and H. Nikaido. 1961b. Galactose mutants of
Salmonella typhimurium. Genetics 46:1295—1303.

 

Hartman, P. E., J. C. Loper, and D. Serman. 1960. Fine
structure mapping by complete transduction between
histidine-requiring salmonella mutants. J. Gen.
Microbiol.‘gg:323-353.

Hartman, Z., 1956. Induced mutibility in Salmonella typhi—
murium. M. S. Thesis, Cornell University, Ithaca,
New York.

 

Horowitz, H. B. 1962. A sensitive assay for galactokinase
in Escherichia coli. Anal. Biochem. 3:498-513.

 

Isselbacher, K. J., E. P. Anderson, K. Kurahashi, and H. M.
Kalckar. 1956. Congenital galactosemia, a single
enzyme block in galactose metabolism. Science 12};
635—636.

Jacob, F., and J. Monod. 1961. On the regulation of gene
activity. Cold Spring Harbor Symp. Quant. Biol. 26:

193-211.

Jacob, F., and J. Monod. 1962. Elements of regulatory
circuits in bacteria. Unesco Symposium on Biological
Organization.

Jacob, F., D. Perrin, C. Sanchez, and J. Monod. 1960.
L'operon: groupe de genes a expression coordonnee
par un operateur. Compt. rend. 250:1727-1729.

Jordan, E., and M. B. Yarmolinsky. 1963. Control of in-
ternal induction of galactose pathway enzymes in an
Escherichia coli mutant. J. Gen. Microbiol. 39:357—
364.

 

68

Jordan, E., M. B. Yarmolinsky, and H. M. Kalckar. 1962.
Control of inducibility of enzymes of the galactose
sequence in Escherichia coli. Proc. Natl. Acad. Sci.
U.S. @332-40.

 

Kalckar, H. M. 1958. Uridine diphosphogalactose: Metabo-
lism, enzymology, and biology. Advances in Enzymol.
20:111—134.

Kalckar, H. M., E. P. Anderson, and K. J. Isselbacher. 1956.
Galactosemia, a congenital defect in a nucleotide
transferase. Biochim. et Biophys. Acta 29:262-268.

Kalckar, H. M., B. Braganca, and A. Munch—Petersen. 1953.
Uridyl transferases and the formation of uridine di—
phosphogalactose. Nature 172:1038.

Kalckar, H. M., K. Kurahashi, and E. Jordan. 1959. Heredi-
tary defects in galactose metabolism in Escherichia
coli mutants, I. Determination of enzyme activities.
Proc. Natl. Acad. Sci. U.S. 45:1776-1786.

 

Kalckar, H. M., and T. A. Sundararajan. 1961. Regulartory
mechanisms in the synthesis of enzymes of galactose
metabolism, II. Genetic defects in galactokinase
activity and their relations to its functions. Cold
Spring Harbor Symp. Quant. Biol. 26:227-231.

Kosterlitz, H. W. 1943. The fermentation of galactose and
galactose-l—PO4. Biochem. J. 31:322—326.

Kurahashi, K. 1957. Enzyme formation in galactose—negative
mutants of Escherichia coli. Science 125:114—116.

 

Kurahashi, K., and A. J. Wahba. 1958. Interference with
growth of certain Escherichia coli mutants by galac—
tose. Biochim. et Biophys. Acta 39:298-302.

 

Lederberg, E. M. 1960. Genetic and functional aspects of
galactose metabolism in Escherichia coli K%12.
Symposium Soc. Gen. Microbiol. 19:115-131.

 

Leloir, L. F. 1951. The enzymatic transformation of uri-
dine diphosphate glucose into a galactose derivative.
Arch. Biochem. 33:186-190.

Leloir, L. F. 1955. The uridine coenzymes. Proc. 3rd
Intern. Congr. Biochem. Brussels, p. 154-162.

Levine, M., and M. Borthwick. 1963. The action of strepto-
nigrin on bacterial DNA metabolism and on induction
of phage production in lysogenic bacteria. Virology
21:568—574.

69

Lowry, O. H., N. J. Rosebrough, A. L. Farr, and F. J.
Randall. 1951. Protein measurement with the folin
phenol reagent. J. Biol. Chem. 193:265-275.

Morgan, T. H. 1910. Sex limited inheritance in Drosophila.
Science 32:120-122.

Morse, M. L. 1962. Preliminary genetic map of seventeen
galactose mutations in E. coli K—12. Proc. Natl.
Acad. Sci. U.S. 48:1314-1318.

Morse, M. L., E. M. Lederberg, and J. Lederberg. 1956a.
Transduction in Escherichia coli K-12. Genetics 41:
142—156.

 

Morse, M. L., E. M. Lederberg, and J. Lederberg. 1956b.
Transductional heterogenotes in Escherichia coli.
Genetics 41:758-779.

 

Munch-Petersen, A.. H. M. Kalckar, E. Cutolo, and E. E. B.
Smith. 1953. Enzymic production of uridine tri-
phosphate: uridine diphosphoglucose perphosphoro-
lysis. Nature_ll§:1036-1037.

Nikaido, H. 1961. Galactose-sensitive mutants of Salmonella,
I. Metabolism of galactose. Biochim. et Biophys.
Acta 48:460-469.

Nikaido, H., and T. Fukasawa. 1961. The effect of mutation
in a structural gene on the inducibility of the en-
zymes controlled by other genes of the same operon.
Biochem. Biophys. Research Comm. 4:338-342.

Robinson, M. J. 1964. A conjugation system in Salmonella
pullorum. Ph.D. Thesis, Michigan State University,
East Lansing, Michigan.

 

Sherman, J. R. 1962. Rapid enzyme assay technique utilizing
radioactive substrate, ion-exchange paper, and liquid
scintillation counting. Anal. Biochem. 5:548-554.

Sherman, J. R., and J. Adler. 1963. Galactokinase from
Escherichia coli. J. Biol. Chem. 238:873—878.

 

Soffer, R. L. 1961. Enzymatic expression of genetic units
of function concerned with galactose metabolism in
Escherichia coli. J. Bacteriol. 82:471-478.

 

Sturtevant, A. H. 1913. The linear arrangement of six sex-
linked factors in Drosophila, as shown by their mode
of association. J. Exper. 2001. 14:43-59.

7O

Sundararajan. T. A. 1963. Interference with glycerokinase
induction in mutants of E. ggli accumulating gal-l—p.
Proc. Natl. Acad. Sci. U.S._§Q:463—469.

Sundararajan, T. A., A. M. C. Rapin, and H. M. Kalckar. 1962.
Biochemical observations on_§. coli mutants defective

in uridine diphosphoglucose. Proc. Natl. Acad. Sci.
U.S. 48:2187-2193.

Wiesmeyer, H., and E. Jordan. 1961. A simple procedure for

the preparation of uridine diphosphogalactose. Anal.
Biochem. 2:281-284.

Yarmolinsky, M. B., H. Wiesmeyer, H. M. Kalckar, and E.
Jordan. 1959. Hereditary defects in galactose
metabolism in Escherichia coli mutants, II.
Galactose-induced sensitivity. Proc. Natl. Acad.
Sci. U.S. 45:1786-1791.

Zinder, N. D., and J. Lederberg. 1952. Genetic exchange in
Salmonella. J. Bacteriol. 64:679-699.

11111111111111“fillimxmnwmmufit