PATHWAY 0F GALACTETOL CATABOLISM
3N KLEBSIELLA PNEUMOMAE

Dissertation for the Degree of Ph. D.
MECHEGAN ST .‘EE UNE‘JERSITY
fiOHi‘é PAUL MARKWELL
1976

 

 

This is to certify that the
thesis entitled
Pathway of Galactito Z Catabolism in

KZebsieZZa mewmmlae

presented by

John Paul Markwel Z

has been accepted towards fulfillment
of the requirements for

Ph . D. degree in Biochemistry

 

[xvi C1210; 2f L 1 (thtw

V

Major professor

mfi _—__“#b—

Date April 21, 1976

0-7639

   
 

7:" Bantams av“ ‘5
"DAB & SHNS'
”UK BINIJERY INC;

ileum amotns
gunman. menu;

7<..» .a-
w

   
  

 

ABSTRACT
PATHWAY OF GALACTITOL CATABOLISM
IN KLEBSIELLA PNEUMONIAE
BY

John Paul Markwell

Previous investigations in this laboratory established that in
Klebsiella pneumoniae galactitol is modified by phosphorylation with
phosphoenolpyruvate to form L-galactitol l-phosphate. The data in
this dissertation establish that L-galactitol l-phosphate is further
catabolized by a previously unreported pathway, part of which is
identical to the D-tagatose 6-phosphate pathway of lactose and
D-galactose catabolism first established by this laboratory for
Staphylococcus aureus. The pathway reactions for L-galactitol

l-phosphate catabolism in K. pneumoniae are as follows:

(1) L-Galactitol l-phosphate dehydrogenase

L-galactitol l-phosphate D-tagatose 6-phosphate
+ -—-————* +
nicotinamide adenine nicotinamide adenine
dinucleotide dinucleotide (reduced)
+
+
H

(2) D-Tagatose 65phosphate kinase

D-tagatose 6-phosphate D-tagatose 1,6-diphosphate

+ -————-+- +
adenosine 5'-triphosphate adenosine S'-diphosphate

 

Ce

CO

St.

SP5

John Paul Markwell

(3) D-Tagatose 1,6—diphosphate aldolase

 

dihydroxyacetone phosphate
D—tagatose 1,6-diphosphate -—-—-+- +
D-glyceraldehyde 3-phosphate

IThe enzymes catalyzing these reactions were partially purified and
characterized. The products of the individual reactions were identi-
fied by enzymatic, chromatographic, and chemical analysis.

L-Galactitol l—phosphate dehydrogenase was shown to be induced
only by growth on galactitol, and was highly specific for L-galactitol
l-phosphate, D-tagatose 6-phosphate, and nicotinamide adenine
dinucleotide (oxidized and reduced forms). The K.m values for the
substrates were as follows: L—galactitol l-phosphate, 0.42 mM;
oxidized nicotinamide adenine dinucleotide, 0.22 mM; D-tagatose
6-phosphate, 0.10 mM; and reduced nicotinamide adenine dinuCleotide,
0.025 mM. The activity was unaffected by monovalent cations, whereas
activity and stability showed a strong requirement for divalent
cations. The molecular weight of the enzyme was estimated by gel
filtration to be 86,000. The equilibrium constant of the reaction,
as written above, was determined to be 8.0 x 10'.11 moles/liter.

D-Tagatose 6-phosphate kinase was shown to be the same enzyme
as the constitutive D-fructose 6-phosphate kinase. This dual speci-
ficity was established by coincident chromatography on substituted
cellulose, a constant ratio of activity throughout purification,
coincident thermal inactivation, competition between the two sub-
strates, and loss of tagatose phosphate activity in a mutant missing

D-fructose 6-phosphate kinase. The phosphoryl donor and acceptor

specificities were determined, and the apparent Km values for the

John Paul Markwell
phosphoryl acceptors were shown to be lowered by the addition of
adenosine 5'-diphosphate to the assays. The molecular weight of the
kinase was estimated by gel filtration to be 92,000.

D-Tagatose l,6—diphosphate aldolase was induced only by growth
on galactitol. The activity was specific for D-tagatose 1,6-
diphosphate (Km = 0.38 mM, relative Vmax = 1.00) and, to a much
lesser extent, D-fructose 1,6-diphosphate (K.m = 0.86 mM, relative
Vmax = 0.04). The enzyme was activated by both monovalent and
divalent cations. The molecular weight was estimated by gel filtra—
tion to be 157,000.

Three mutants, each missing one of the above enzymes, were
isolated. These mutants were unable to grow on galactitol, estab-
lishing that this is the sole pathway of physiological significance
in this organism. Revertant strains, having regained the ability
to grow on galactitol, were shown to contain the formerly missing
enzyme activities. The enzymes of this pathway, therefore, con-
stitute the first report of a catabolic route of galactitol utiliza-

tion in this or any other organism.

PATHWAY OF GALACTITOL CATABOLISM

IN KLEBSIELLA PNEUMONIAE

BY

John Paul Markwell

A DISSERTATION

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

DOCTOR OF PHILOSOPHY

Department of Biochemistry

1976

ACKNOWLEDGEMENTS

I am pleased to acknowledge the assistance of the following
persons in the completion of this work.

Dr. Richard L. Anderson -~ major professor, mentor, and friend

Drs. Loran Bieber, John Boezi, James Fairley, Arnold Revzin,

Robert Ronzio, and Harold Sadoff -- members of my guidance

committee

Donald Bissett -- for much advice on experimental techniques

Grant Shimamoto -- for work on the first two reactions of
the pathway

James VanPopering -- isolation of mutants VP-8 and VP-l4
Words alone are not sufficient to acknowledge the contribution
of others.

Norman and Eileen Markwell -- parents; contributed love,
encouragement, and wealth, asking nothing in return

Mary Ann -- wife; carried my dream when I grew weary

ii

TABLE OF CONTENTS

LIST OF TABLES O O C O O O I O O O O O O O O O O O O O O O 0
LIST OF FIGURES O O O O O I O O O O I O O O O O O O O O O 0
LIST OF ABBREVIATIONS . . . . . . . . . . . . . . . . . . .
GENERAL INTRODUCTION. . . . . . . . . . . . . . . . . . . .

LITEMTURE REVIEW I O O O O O O C C O O O O O O O O O O O 0

SECTION 1 - ELUCIDATION OF THE METABOLIC FATE OF GALACTITOL .

INTRODUCTION. . . . . . . . . . . . . . . . . . . .
MATERIALS AND METHODS . . . . . . . . . . . . . . .
Bacterial Strain. . . . . . . . . . . . . .
Media . . . . . . . . . . . . . . . . . . .
Sterilization Technique . . . . . . . . . .
Growth of Cells . . . . . . . . . . . . . .
Harvesting of Cells . . . . . . . . . . . .
Centrifugation. . . . . . . . . . . . . . .
Preparation of Cell Extracts. . . . . . . .
Protein Determination . . . . . . . . . . .
Removal of Barium from Sugar Phosphates . .
Preparation of L-Galactitol l-Phosphate . .
Preparation of D-Tagatose 6-Phosphate . . .
Gas-Liquid Chromatography . . . . . . . . .
Preparation of Ion Exchange Resins. . . . .
DEAE-Cellulose Chromatography . . . . . . .
Enzymatic Assays. . . . . . . . . . . . . .
Galactitol Kinase . . . . . . . . . . . . .
Galactitol Dehydrogenase. . . . . . . . . .
Galactitol Epimerase. . . . . . . . . . . .
L-Galactitol l-Phosphate Dehydrogenase. . .
D-Tagatose 6-Phosphate Kinase . . . . . . .
D-Tagatose 1,6-Diphosphate Aldolase . . .
Source of Materials . . . . . . . . . . . .
RESULTS . . . . . . . . . . . . . . . . . . . . . .
Absence of Galactitol Kinase, Galactitol
Dehydrogenase, and Galactitol Epimerase
Activities. . . . . . . . . . . . . . . .
Prediction of the Galactitol Pathway. . . .
Detection of the Pathway Enzymes. . . . . .
DISCUSSION. . . . . . . . . . . . . . . . . . . . .

iii

Page

vii
viii

10

ll
12
12
12
12
12
12
13
13
13
13
13
14
14
14
14
15
15
15
15
16
16
17
17
19

19
20
20
24

SECTION 2 - PURIFICATION AND CHARACTERIZATION OF THE
ENZYMES INVOLVED IN THE CATABOLISM OF
L-GALACTITOL l-PHOSPHATE. . . . . . . . . . . .

INTRODUCTION. . . . . . . . . . . . . . . . . . . .
MATERIALS AND METHODS . . . . . . . . . . . . . . .

RESULTS
A.

Growth Conditions . . . . . . . . . . . . .
Determination of Substrate Concentration. .

L-Galactitol l—Phosphate Dehydrogenase Assays

D-Fructose 6-Phosphate Kinase Assays. . . .
D-Tagatose 1,6-Diphosphate Aldolase Assays.
D-Tagatose 6-Phosphate Kinase Assays. . . .
Alkaline Phosphatase Assays . . . . . . . .
Glucose 6-Phosphate Dehydrogenase Assays. .
Lactate Dehydrogenase Assays. . . . . . . .
Pyruvate Kinase Assays. . . . . . . . . . .
D-Fructose 1-Phosphate Kinase Assays. . . .
D-Fructose l,6-Diphosphate Aldolase Assays.
Ketohexose Determination. . . . . . . . . .
Phosphate Determination . . . . . . . . . .
Bentonite Treatment . . . . . . . . . . . .
Ammonium Sulfate Precipitation. . . . . . .
Sephadex G-25 Chromatography. . . . . . . .
Sephadex G-150 Chromatography . . . . . . .
Hydroxyapatite Adsorption . . . . . . . . .
Alumina Cy Adsorption . . . . . . . . . . .
Hexokinase Assays . . . . . . . . . . . . .
Protein Assays. . . . . . . . . . . . . . .
Preparation of NAD+- or AMP-Sepharose . . .
Dialysis. . . . . . . . . . . . . . . . . .
Polyacrylamide Gel Electrophoresis. . . . .
Paper Chromatography. . . . . . . . . . . .
Ketohexose Differentiation. . . . . . . . .
Effect of NaBH4 . . . . . . . . . . . . . .
Effect of NEM . . . . . . . . . . . . . . .
Dephosphorylation of Sugar Phosphates . . .
Monovalent Cation Activation. . . . . . . .
Divalent Cation Activation. . . . . . . . .
Determination of Dehydrogenase Equilibrium
Constant. . . . . . . . . . . . . . . . .
Source of Materials . . . . . . . . . . . .
L-Galactitol l-Phosphate Dehydrogenase
(L-Galactitol l-PhosphatezNAD+
Oxidoreductase) . . . . . . . . . . . . . .
Purification. . . . . . . . . . . . . . . .
Cell extract. . . . . . . . . . . .
DEAE-cellulose chromatography . . .
Ammonium sulfate precipitation. . .
Sephadex G-150 chromatography . . .
Properties. . . . . . . . . . . . . . . . .
Stability . . . . . . . . . . . . .
pH optima . . . . . . . . . . . . .

iv

Page

27

28
29
29
29
29
3O
3O
31
31
31
31
31
31
32
32
32
32
32
33
33
33
33
34
34
34
34
34
35
35
35
35
36
36
36

36
37
39

39
39
39
41
41
41
44
44
44

SECTION 3 - GENETIC EVIDENCE FOR THE PHYSIOLOGICAL SIGNIFI

Substrate specificity . . . . . . . .
Cation activation . . . . . . . . . .
Kinetic constants . . . . . . . . . .
Molecular weight determination. . . .
Reaction Characteristics. . . . . . . . . . .
Product identification. . . . . . . .
Stoichiometry . . . . . . . . . . .
Determination of equilibrium constant
Discussion. . . . . . . . . . . . . . . . . .
D-Tagatose 6-Phosphate Kinase [ATP:D-Fructose
6-Phosphate (D-Tagatose 6-Phosphate) 1-
Phosphotransferase] (EC 2.7.1.11) . . . . . .
Purification. . . . . . . . . . . . . . . . .
Cell extract. . . . . . . . . . . . .
Bentonite treatment . . . . . . . . .
DEAR-cellulose chromatography . . . .
Properties. . . . . . . . . . . . . . . . . .
Stability . . . . . . . . . . . . . .
Thermal inactivation. . . . . . . . .
pH optimum. . . . . . . . . . . . .
Phosphoryl acceptor specificity and
kinetic constants . . . . . . . .
Phosphoryl donor specificity. . . . .
Sulfhydryl requirement. . . . . . . .
Molecular weight. . . . . . . . . . .
Product Identification. . . . . . . . . . . .
Production of product . . . . . . . .
Isolation of the product. . . . . . .
Identification of the product . . . .
Reaction stoichiometry. . . . . . . .
Discussion. . . . . . . . . . . . . . . . . .
D-Tagatose 1,6-Diphosphate Aldolase
(D-Tagatose 1,6-Bisphosphate D-Glyceraldehyde
3-Phosphate-Lyase). . . . . . . . . . . . . .
Purification. . . . . . . . . . . . . . . . .
Preparation of cell extract . . . . .
DEAR-cellulose chromatography I . . .
Sephadex G-150 chromatography . . . .
DEAR-cellulose chromatography II. . .
Properties. . . . . . . . . . . . . . . . . .
Stability . . . . . . . . . . . . . .
pH optimum. . . . . . . . . . . . . .
Substrate specificity and kinetic
constants . . . . . . . . . . . . .
Effect of cations . . . . . . . . . .
Effect of NaBH4 . . . . . . . . . . .
Molecular weight. . . . . . . . .
Discussion. . . . . . . . . . . . . . . . . .

CANCE OF THE PROPOSED PATHWAY FOR GALACTITOL
CATABOLI SM 0 O O O O O O O O O O O O O O O O O O O

Page

44
46
46
50
50
50
50
52
52

55
55
56
56
56
59
59
59
59

62
65
66
66
66
66
68
68
69
70

73
73
73
73
75
75
79
79
79

79
85
87
89
89

93

INTRODUCTION. . . . . . . . . . . . . . . . . . . .
MATERIALS AND METHODS . . . . . . . . . . . . . . .
Organisms . . . . . . . . . . . . . . . . .
Media . . . . . . . . . . . . . . . . . . .
Enzyme Induction. . . . . . . . . . . . . .
Monitoring of Cell Growth . . . . . . . . .
Enzymatic Assays. . . . . . . . . . . . . .
Mutant A9-l (D-Fructose 6-Phosphate Kinase
Negative) . . . . . . . . . . . . . . . .
Isolation of Mutants Missing L-Galactitol
l-Phosphate Dehydrogenase (VP-14) or
D-Tagatose 1,6-Diphosphate Aldolase (VP-8)
Isolation of Spontaneous Revertants (VP-8R,
VP-14R, and A9-1R). . . . . . . . . . . .
Source of Materials . . . . . . . . . . . .
RESULTS . . . . . . . . . . . . . . . . . . . . . .
Analysis of Enzyme Defects in the Mutants .
Growth of Mutants . . . . . . . . . . . . .
Specificity of Induction. . . . . . . . . .
DISCUSSION. . . . . . . . . . . . . . . . . . . . .

SUMMARY . . . . . . . . . . . . . . . . . . . . . . . . . .

LIST OF REFERENCES. . . . . . . . . . . . . . . . . . . . .

vi

Page

94
95
95
95
95
96
96

96

96

97
98
99
99
99
104
106

108

110

Table

10

11

12

LIST OF TABLES

Oxidation of L-galactitol l-phosphate to
D-tagatose 6-phosphate. . . . . . . . . . . . . .

Cleavage of D—tagatose 1,6—diphosphate to triose
phosphates. . . . . . . . . . . . . . . . . . . .

Purification of L-galactitol l-phosphate
dehydrogenase . . . . . . . . . . . . . . . . . . .

Divalent cation activation of EDTA-treated
L-galactitol 1-phosphate dehydrogenase. . . . . . .

Determination of the equilibrium constant for the
L-galactitol 1-phosphate dehydrogenase reaction . .

Purification of D-tagatose 6-phosphate kinase . . .

Determination of the product stoichiometry of
D-tagatose 6-phosphate kinase . . . . . . . . . .

Purification of D-tagatose 1,6-diphosphate aldolase

Effect of divalent cations on EDTA-treated
D-tagatose 1,6-diphosphate aldolase . . . . . . .

Stimulation of D-tagatose 1,6-diphosphate aldolase
by monovalent cations . . . . . . . . . . . . . . .

Induction of enzymes for the catabolism of
L-galactitol l-phosphate in parental and mutant

strains 0 O I O O O O O O O O C I I O O O O O O 0

Induction of the galactitol pathway enzymes in
K. pneumoniae grown on a variety of carbohydrates .

vii

Page

21

22

40

47

53

57

71

74

86

87

100

105

Figure

10

11

12

13

14

LIST OF FIGURES

Catabolic pathways for D-mannitol and D—glucitol. .

Proposed pathway for galactitol catabolism in
Klebsiella pneumoniae . . . . . . . . . . . . . . . .

Chromatography of L-galactitol 1-phosphate
dehydrogenase on DEAE-Cellulose . . . . . . . . . . .

Chromatography of L-galactitol 1-phosphate
dehydrogenase on Sephadex G-150 . . . . . . . . . . .

pH optimum of L-galactitol l-phosphate dehydrogenase.
Lineweaver-Burk plots of L-galactitol 1-phosphate
dehydrogenase with L-galactitol l-phosphate and

NAD+ as substrates. . . . . . . . . . . . . . . . . .
Lineweaver-Burk plots of L-galactitol l-phosphate
dehydrogenase with D-tagatose 6-phosphate and NADH

as substrates 0 O O O O O O O O O O O O O O O O O O 0

Molecular weight determination of L-galactitol
1-phosphate dehydrogenase . . . . . . . . . . . . . .

Chromatography of D-fructose 6-phosphate kinase and
D-tagatose 6-phosphate kinase on DEAR-cellulose . . .

Thermal inactivation of D-tagatose 6-phosphate
kinase and D-fructose 6-phosphate kinase. . . . . . .

pH optimum of D-fructose 6-phosphate kinase and
D-tagatose 6-phosphate kinase . . . . . . . . . . . .

The effect of D-fructose 6-phosphate concentration
on the kinase activity. . . . . . . . . . . . . . . .

The effect of D-tagatose 6-phosphate concentration
on the kinase activity. . . . . . . . . . . . . . . .

Molecular weight determination of D-tagatose
6-phosphate kinase. . . . . . . . . . . . . . . . .

viii

Page

25

42

43

45

48

49

51

58

60

61

63

64

67

Figure Page

15 DEAR-Cellulose chromatography I of D-tagatose
1,6-diphosphate aldolase. . . . . . . . . . . . . . . . 76

16 Chromatography of D-tagatose 1,6-diphosphate
aldolase on Sephadex G-150. . . . . . . . . . . . . . . 77

17 DEAE-Cellulose chromatography II of D-tagatose
1,6-diphosphate aldolase. . . . . . . . . . . . . . . . 78

18 pH optimum of D-tagatose 1,6-diphosphate aldolase . . . 80

19 Lineweaver-Burk plots for D-tagatose 1,6-diphosphate
aldolase. . . . . . . . . . . . . . . . . . . . . . . . 81

20 Concomitant thermal inactivation of D-tagatose
1,6-diphosphate aldolase and D-fructose
1,6-diphosphate aldolase activities . . . . . . . . . . 83

21 Effect of potassium ion on D—tagatose 1,6-diphosphate
aldolase. . . . . . . . . . . . . . . . . . . . . . . . 88

22 Molecular weight determination of D—tagatose
1,6-diphosphate aldolase. . . . . . . . . . . . . . . . 90

23 Growth characteristics of K. pneumoniae PRL-R3 (0-) . . 101
24 Growth characteristics of strain A9-l . . . . . . . . . 102

25 Growth of strains VP-8, VP-14, A9-1R, VP-8R, and
VP-14R on galactitol. . . . . . . . . . . . . . . . . . 103

ix

If

in

[W
.
\r .

N ”

. .1. 3148‘

NE

DTE

DTT

35$

XXX

PEP

ATP

ITP
GTP
UTP
CTP
TTP
NAD
NADH
NADP

NADPH

a-methylglucoside
DEAE

EDTA

NEM

DTE

BME

XXX

LIST OF ABBREVIATIONS

phosphoenol pyruvate

adenosine 5'-triphosphate

adenosine 5'-diphosphate

adenosine 5'-monophosphate

inosine 5'-triphosphate

guanosine 5'-triphosphate

uridine S'-triphosphate

cytidine 5'-triphosphate

thymidine S'-triphosphate

nicotinamide adenine dinucleotide
nicotinamide adenine dinucleotide (reduced)
nicotinamide adenine dinucleotide phosphate

nicotinamide adenine dinucleotide phosphate
(reduced)

methyl-a-D-glucopyranoside
diethylaminoethyl
ethylenediaminetetraacetic acid
N-ethylmaleimide
dithioerythritol

dithiothreitol
B—mercaptoethanol; 2-thioethanol

absorbance at xxx nanometers

mg

Hz

min

E

Q

Tris

HEPES

Pyr

acceleration of gravity
gram

milligram

Hertz

minutes

milliliter

molar

millimolar

centimeter

millimeter

nanometer

tris (hydroxymethyl) aminomethane

N-2-hydroxyethylpiperazine-N'-2-ethanesulfonic
acid

pyruvate

phosphate

xi

GENERAL INTRODUCTION

The purpose of this work was to elucidate the pathway for the
catabolism of galactitol in Klebsiella pneumoniae. As will be
documented in the Literature Review, galactitol is modified in
extracts of K. pneumoniae by phosphorylation, to form L-galactitol
1-phosphate. No route for the further catabolism of this compound
has yet been reported. It was with this knowledge that I began my
studies with this system.

The whole of this research is divided into three parts. The
experiments of the first section show that the free hexitol is not
modified except by the above-mentioned phosphorylation. A pathway
for the catabolism of L-galactitol 1-phosphate is also proposed,
along with the reasoning behind it. The activities of the required
enzymes (L—galactitol l-phosphate dehydrogenase, D-tagatose
6-phosphate kinase, and D-tagatose 1,6-diphosphate aldolase) are
shown to be present in the extracts of galactitolegrown cells.

The second part of this dissertation consists of the partial puri-
fication and characterization of these enzymes and their reaction
products. These data confirm that the reactions of this pathway
are catalyzed in vitro by the purified enzymes. The results pre-
sented in the third section of this dissertation demonstrate that

mutants missing any one of the pathway enzymes cannot grow on

2
galactitol. This finding confirms the functioning of these enzymes
in galactitol catabolism in this organism.
The pathway is as follows: L—galactitol 1-phosphate-—+
D—tagatose 6-phosphate -—+-D-tagatose 1,6-diphosphate ——+-dihydroxy-

acetone phosphate + D-glyceraldehyde 3-phosphate.

 

 

()

Lt.

{1e

LITERATURE REVIEW

While the literature contains many reviews of carbohydrate
metabolism in general, only a few devote more than a paragraph to
the catabolism of the hexitols. The two extensive reviews of the
hexitols (1,2) are now almost 30 years old, and the biochemical
data very out-dated. It shall then be the primary purpose of this
section to review what is currently known about the bacterial
utilization of the hexitols, especially galactitol.

Four of the hexitols have been found in nature (2,3). These
naturally occurring hexitols are D-glucitol, D-mannitol, galactitol,
and L-iditol. The greatest natural abundance of free galactitol
occurs in plants, especially Madagascar manna (2) and the manna of
Gymnosporia diflexa (4). Large amounts of galactitol have also
been found in Euonymus japonica (5,6) and in some yeast (7).

The presence of free galactitol is not unknown in mammalian
systems. The polyol has been identified in tissues of rats and
chicks given large amounts of D-galactose in the diet (8,9). Humans
deficient in the galactose kinase or uridyl transferase of the
Leloir pathway have also been reported to accumulate galactitol
(10,11). Despite these occurrences, there are numerous reports
Ithat galactitol is not metabolized in mammals (12,13,14,15). Since
galactitol is not a substrate for mammalian polyol dehydrogenases
(16,17,18), it seems likely that it is formed when D-galactose

3

4
concentrations are high enough to act as a substrate for aldose
reductase (19,20) or L-hexonate dehydrogenase (21,22), both of which
are known to oxidize galactitol and to have a wide occurrence in
mammalian tissues (23,24).

D-Mannitol, D—glucitol and galactitol have long been known to
be metabolized by a variety of bacterial species. Growth and fer-
mentation of these hexitols have been used as a basis for the taxo-
nomic classification of the Gram—negative organisms of the colon—
typhoid (25) and the colon-aerogenes groups (26,27). Other micrdbes
that have been reported to utilize galactitol include Salmonella
typhimurium (28), Streptococcus lactis (29), Bacillus subtilis (30),
five species of the yeast genera Torulopsis and Debaryomyces (31),
and four species of the tomato, cabbage, and melon wilt-causing
Fusarium (32).

Enzyme activities that can modify the hexitols by oxidation
have been reported for a number of microbial species. Pseudomonas
fluorescens cell extract is reported to oxidize D-mannitol to
D-fructose, and D-glucitol to a mixture of D—fructose and L-sorbose
(33). Two polyol dehydrogenases were reported from Acetobacter
suboxgdans (34). One of these activities, with an acid pH optimum,
oxidizes D—mannitol and D-glucitol, but not galactitol, and was
cytochrome-linked. The other activity, having an alkaline pH optimum,
is NAD-linked, and again is specific for D-mannitol and D-glucitol,
but not galactitol. Another particulate dehydrogenase, specific
for D-mannitol and D-glucitol, was found in Gluconobacter oxydans:
this is in addition to soluble NAD- and NADP-linked dehydrogenases

catalyzing these same oxidations (35,36). Lin (37) has reported

S

that D-arabitol dehydrogenase from Aerobacter aerogenes, an NAD-
specific enzyme, oxidizes D-mannitol to D-fructose. However, since
this activity is induced by growth on D-arabitol, and not D-mannitol,
it seems unlikely that this enzyme plays a part in mannitol utiliza-
tion for this organism. An NAD-specific polyol dehydrogenase that
oxidizes mannitol and glucitol, but not galactitol, was found in the
yeast Candida utilis (34,38,39). Polyol dehydrogenases of mammalian
tissues have also been studied. These catalyze the oxidation of
D-gluticol and L-iditol, but not D—mannitol or galactitol (16,17,18).

None of the microbial enzymes thus far mentioned has been shown
to oxidize galactitol. Shaw (40), after futile attempts to detect
this activity in Escherichia coli, was able to isolate an airborne
Pseudomonas species that exhibited a galactitol dehydrogenase. The
enzyme is NAD specific, and oxidizes galactitol, D-glucitol, and
L-iditol, while failing to utilize D-mannitol. Glactitol dehydrogenase
activity has also been found in Staphylococcus species (41) as well
as several species of yeast (31,42).

It was only after the discovery of a D-mannitol l-phosphate
dehydrogenase (43) that researchers considered the initial step for
hexitol catabolism to possibly be a phosphorylation. D-Mannitol
1-phosphate dehydrogenase activity has been found in extracts of
Diplococcus pneumoniae (43), Lactobacillus arabinosus (44), Lacto-
bacillus casei (45), Lactobacillus palntarum (4S), Leuconostoc
mesenteroides (45), Aerobacter aerogenes (45,46,47), Bacillus
subtilis (47), and Escherichia coli (45-48). D-Glucitol 6-phosphate
dehydrogenase activity has been detected in Lactobacillus casei

(49), Aerobacter aerogenes (46,47), Klebsiella pneumoniae (50,51),

6
and Escherichia coli (46,48). L—Galactitol 1-phosphate dehydrogenase
has been reported for Aerobacter aerogenes (46) and for Escherichia
coli (46,48,52).

D-Mannitol l-phosphate dehydrogenase from A. aerogenes (47) has
been purified and shown to be a 40,000 molecular-weight protein,
specific for NAD and D-mannitol l-phosphate. This enzyme is present
in D-glucose grown cells, but is found at much higher specific
activities in D—mannitol-grown cells (46,47). D-Glucitol 6-phosphate
dehydrognease from the same organism is specifically induced by
growth on D-glucitol, utilizes NAD rather than NADP, and oxidizes
L-galactitol l-phosphate at 3% the rate of D-glucitol 6-phosphate
(47). Prior to this investigation there had been no report on the
characteristics of an L-galactitol l-phosphate dehydrogenase from
any source .

The presence of these hexitol phosphate dehydrogenases led to
the search for hexitol kinases (ATP:hexitol phosphotransferase)
that would be responsible for the phosphorylation, but none was
ever found (47). This lack of a hexitol kinase activity was
explained in 1964 when Roseman's group discovered the PEP:carbohydrate
phosphotransferase system in E. coli (53). The system is composed
of at least two soluble and one membrane-bound proteins. It is the
membrane-bound protein that is specific for the sugar that is
phosphorylated (54). The reader is directed to one of several
extensive reviews on the subject (55,56,57). It has been shown that
this system is responsible for the phosphorylation of D-mannitol

(58) and D-glucitol (50) that initiates the catabolism of these

7
hexitols. The pathway for D-mannitol and D-glucitol is summarized
in Figure 1.

Phosphorylation of galactitol by the PEP-phosphotransferase
system was recently shown by Shimamoto (59) in K. pneumoniae. It
was demonstrated that the formation of the phosphorylated derivative
of galactitol was dependent on PEP, and each of the three protein
components of the PEP-phosphotransferase system. It was additionally
shown that ATP could not substitute for PEP. Furthermore, the product
was conclusively identified as L-galactitol l-phosphate by gas-liquid
chromatography, mass spectrometry, periodate oxidation, and enzymatic
oxidation. The PEP-dependent phosphorylation of galactitol by E.
coli was subsequently suggested by Lengler (52,60), but he did not
show if the conformation of the product was D—galactitol l-phosphate
or L—galactitol l-phosphate.

As is true for D—mannitol and D—glucitol catabolism, it might
be expected that catabolism of L-galactitol 1-phosphate would
involve sequential oxidation to a ketohexose phosphate, phosphoryla-
tion to ketohexose diphosphate, and aldolase cleavage to triose
phosphates. Oxidation of L-galactitol l-phosphate would be to
D-tagatose l-phosphate or D-tagatose 6-phosphate, the evidence
indicating the latter (48,59). D-Tagatose 6-phosphate is phos-
phorylated to D-tagatose 1,6-diphosphate in Staphylococcus aureus
(61) and Group N streptococci (62) in the process of lactose and
D-galactose catabolism. The D-tagatose 1,6-diphosphate is cleaved
to dihydroxyacetone phosphate and D-glyceraldehyde 3-phosphate (61).
which are then assimilated into the glycolytic pathway. Although

this pathway is the most likely prospect for further catabolism

.Houaooamln can Houaccmelo How mstnumm Usaonmumu

mumsmmosmtm
ocmnooamuoomawsa

 

acmmo
mumndmonmacuw.a
omouooumla
memmo
030 Ill
mumnmmocm ll
osouoom
waoucmnfin II
as
. u.
acmmo
o
acmmo

oumnmmosmlm
omouooumlo

acumo

l

 

x 0?

.H whomfim

opmsmmonmla

 

 

douHGGMZIQ HouwcGMZIo
acmmo
II. II
II IL
Wetr use mum .I
_&WW
mumsmmonmlo
Houflosawla Houwosawla
memmfl
I1
I
.v In II
II A m mmm II

 

 

9

of L-galactitol l-phosphate, the enzymes of the D—tagatose-6-phosphate
pathway have been reported absent from extracts of K. pneumoniae
cells grown on D-glutoxe, D—galactose, and lactose (62).

In summary, the pathway of galactitol catabolism has never
been established for any organism. Cells of K. pneumoniae growing
on galactitol have been demonstrated to form L-galactitol l-phosphate
(59). The oxidation of this product to D-tagatose 6-phosphate has
been suggested (48), while further catablism of D—tagatose 6-phosphate
is as yet unknown in this organism. It is the elucidation of this

catabolic pathway that my Ph.D. dissertation will address.

SECTION 1

ELUCIDATION OF THE METABOLIC FATE OF GALACTITOL

10

INTRODUCTION

As described in the Literature Review, the pathway of galactitol
catabolism has yet to be established for any organism. Although
L—galactitol l-phosphate has been clearly shown to be a product of
the phosphorylation of galactitol by PEP in Klebsiella pneumoniae
(59), no evidence was presented to show that this is the major, let
alone the only, initiation step for the pathway of galactitol
catabolism in this organism. The purpose of this section is to
show that the formation of L-galactitol 1-phosphate is indeed the
prime reaction of the pathway for galactitol utilization. Further-
more, the remaining steps in the catabolic pathway--analogous to
the reactions of the D-mannitol and D-glucitol pathways--will be
postulated, based on the presence of enzymes found in the cell
extract. The proposed pathway is as follows: galactitol -—+:
L-galactitol l-phosphate -+ D-tagatose 6-phosphate ——+ D-tagatose
1,6-diphosphate -—+ dihydroxyacetone phosphate + D-glyceraldehyde

3-phosphate.

ll

MATERIALS AND METHODS

Bacterial Strain. The organism used was Klebsiella pneumoniae

PRL-R3 (formerly designated Aerobacter aerogenes PRL-R3).

Media. Cells were grown in a mineral medium containing 0.15%

KH2P04, 0.71% Na HPO

2 4. 0.3% (NH4)2504, 0.01% MgSO

4, and 0.0005%

FeSO4-7H20. Galactitol was used as a carbon source at a concentra-
tion of 0.4% (w/v). Nutrient broth was made by mixing 8 g of the

dehydrated powder with 1 liter of water.

Sterilization Technique. All media were made sterile by heating
for 20 min at 121° in a Wilmot Castle autoclave, model Thermomatic

60. Carbohydrates were sterilized separately from the mineral medium.

Growth of Cells. Cells were grown in 500 ml of mineral medium
plus carbohydrate, contained in a 2800 ml Fernbach flask. The
flasks were incubated in the dark at 30°, and aerated by constant
motion on a New Brunswick Scientific Co. rotary shaker. The

inoculum was 7 ml of a culture grown in nutrient broth.

Harvesting of Cells. Continued incubation of cells was
halted when a culture reached an A600 of approximately 0.6, as

measured in 18 x 150 cm tubes with a Coleman Jr. spectrophotometer.

Cells were harvested by centrifugation at 6,000 x g_for 10 min.

12

 

wi

13
The cells were suspended in 0.85% (w/v) NaCl and centrifuged again

at 12,000 x g_for 10 min. The cells were used within a hour.

Centrifugation. All centrifugations were done in a Sorvall

 

refrigerated centrifuge, model RC-2B, at 0 to 4°. The rotor radius

was either 4.34 or 5.75 inches.

Preparation of Cell Extracts. Cell extracts were prepared by
suspending the harvested, washed cells in the appropriate buffer
and exposing the solution sonic vibration (10,000 Hz) in a
Raytheon sonic oscillator, model DElOl, for 20 min. During soni-
cation, the apparatus was cooled with a circulating ice-water
mixture. Following disruption of the cells, the solution was
centrifuged at 12,000 x g_for 10 min to remove whole cells and
debris. The resulting supernatant was decanted and termed the

cell extract.

Protein Determination. Protein concentration was determined
by the method of Lowry et a1. (63). Bovine serum albumin was used

as a standard.

Removal of Barium from Sugar Phosphates._ Barium salts of

sugar phosphates were treated with washed, dried Dowex 50W-X8

+ . . . .
(H form). The solution was then neutralized with NaOH and diluted

to volume.

Preparation of L-Galactitol l—Phosphate. L-Galactitol
l-phosphate was synthesized by reduction of D—galactose 6-phosphate

with NaBH4 as described by Shimamoto (59).

14

Preparation of D-Tagatose 6-Phosphate. D-Tagatose 6-phosphate

 

was synthesized from D—galacturonic acid as described by Bissett (64).

Gas-Liquid Chromatography. Acetylated derivatives of the hexi-
tols were prepared by incubating 0.1 mg of the polyol with 0.2 m1
of acetic anhydride and 1 mg of anhydrous sodium acetate for 2 hr
at 100°. Gas-liquid chromatography was performed on a Hewlett
Packard 5830A Reporting Gas Chromatograph, employing a 3% XE-60
column, and an oven temperature of 205°. The instrument was pro—

grammed to integrate and report the area of each peak.

Preparation of Ion Exchange Resins. Commercial resins were
stirred overnight in 2 M NaOH on a magnetic stirrer, allowing the
stirring motor to heat the solution. The slurry was washed in
water until the pH was neutral, and then washed with 4 M HCl. The
resin was then washed again with water, and then washed with the
appropriate acid or base to generate the desired ionic form. The

resins were then dried at 100° and stored in the dry state.

DEAE-Cellulose Chromatography. The DEAR-cellulose was washed
with 1 M NaOH, 1 M HCl, and 1 M NaOH as described by Peterson and
sober (65). When not in use, the exchanger was stored in distilled
water at 4°. To equilibrate a poured column, the column was washed
with buffer until the effluent pH matched that of the buffer. The
protein was then passed through the column, and the column was
washed with buffer until protein ceased to elute. The adsorbed
protein was then eluted with a linear or stepwise gradient of salt.

Prior to reuse, the column was washed with 2 M NaCl to remove

15
tightly bound material. After approximately five runs, the column

was unpacked, and the cellulose rewashed in l M HCl and water.

Enzymatic Assays. One unit of enzymatic activity is defined

 

as the amount of enzyme that catalyzes the conversion of one umole
of substrate to product per minute at 30°. Assays involving the
oxidation or reduction of pyridine nucleotides were monitored in
microcuvettes with a 1.0 cm path length at 340 nm, using a thermo-
stated (30°) Gilford spectrophotometer, model 2400. Other assays
that involved incubations were temperature controlled in a Precision
Scientific Co. water bath. The concentrations of substrate added

to an assay were determined on a weight basis of the added substrate.

Galactitol Kinase. Assays (0.15 ml) contained 10 umoles of

 

glycylglycine buffer (pH 8.0), 0.5 umole of ATP, 1.0 umole of

MgCl 0.05 umole of NADH, 0.5 umole of PEP, 7.5 umoles of galactitol,

2!
excess pyruvate kinase and lactate dehydrogenase, and rate-limiting
amounts of cell extract. Controls were run without galactitol or

ATP to correct for ATPase and PEP:ga1actitol phosphotransferase,

respectively.

Galactitol Dehydrogenase. Assays (0.15 ml) contained 10 umoles

 

+
of Tris buffer (pH 9.0), 0.01 umole of MgCl 1.0 umole of NAD or

2’
+
NADP , and 7.5 umoles of galactitol. Controls without galactitol

were used to correct for NADH oxidase.

Galactitol Epimerase The incubation (5 ml) contained the

 

following: 200 umoles of Tris buffer (pH 7.5); 1.5 umoles of

+ +
NADP ; 1.5 umoles of NAD ; 0.3 umole of MgClz; 50 umoles of

16
galactitol; 2514moles of a-methylglucoside; and cell extract. The
tube was incubated at 30°, and 0.75 ml samples were removed at 30
min intervals. The samples were immediately placed in a 100° water
bath for 10 min to stop any enzymatic activity, and the samples
were then centrifuged at 5,300 x g_for 10 min. One-half milliliter
of the supernatant was diluted to 5 ml with water, and treated with
dry Amberlite MB—3 to remove all ions. Of the deionized solution,
0.5 ml was placed inside a 2-dram vial and dried using a Buchi
Rotovapor. When dry, the samples were acetylated and injected into
the gas chromatograph. Alpha-methylglucoside was used as an internal
standard for this system. Although the absolute amount of a sugar
cannot be accurately determined from its peak area due to variations
in sample volume or evaporation during acetylation, its area
relative to the area of the a-methylglucoside peak can be accurately
determined. The retention times were as follows: o-methylglucoside,
5.12 min; D-mannitol, 8.83 min; galactitol, 9.71 min; and D—glucitol,

11.62 min.

L—Galactitol 1-Phosphate Dehydrogenase. Assays (0.15 ml) con-

 

+
tained 10 umoles of Tris buffer (pH 9.0), 0.05 umole of NAD or

+
NADP , 0.2 pmole of L-galactitol l-phosphate, and limiting amounts

of the enzyme solution to be measured.

D-Tagatose 6-Phosphate Kinase. Assays (0.15 ml) contained 10
umoles of glycylglycine buffer (pH 8.0), 0.5 umole of ATP, 1.0 umole
of MgClz, 0.05 umole of NADH, 0.5 umole of PEP, 0.2 pmole of
D-tagatose 6-phosphate; excess of pyruvate kinase and lactate

I

dehydrogenase; and rate-limiting amounts of cell extract. Controls

17
were run without ATP, D-tagatose 6-phosphate, or both, to correct
for Legalactitol l-phosphate dehydrogenase, ATPase, and NADH

oxidase, respectively.

D-Tagatose 1,6-Diphosphate Aldolase. The assay (0.15 ml)

 

contained: 10 umoles of Tris buffer (pH 7.5); 1.0 umole of MgClz;

0.5 “mole of ATP; 0.05 umole of NADH; 0.05 umole of D-tagatose
6-phosphate; excess D—fructose 6-phosphate kinase, o-glycerolphosphate
dehydrogenase and triose phosphate isomerase; and rate-limiting
amounts of cell extract. Controls were run without kinase or
D—tagatose 6-phosphate to correct for L-galactitol l-phosphate
dehydrogenase and NADH oxidase, respectively. The cuvettes were
incubated at 30° for 5 min before addition of cell extract to allow
phosphorylation of the D-tagatose 6-phosphate to D—tagatose

1,6-diphosphate.

Source of Materials. The following were obtained from Sigma

 

Chemical Co.: galactitol; NaBH4; D—galacturonic acid; D-mannitol;
D—glucitol; DEAR-cellulose; glycylglycine; ATP; rabbit muscle
pyruvate kinase and D_fructose 6-phosphate kinase; Tris (Trizma
base); d-methylglucoside; quinacrine; and an a-glycerol phosphate
dehydrogenase-triosephosphate isomerase mixture. NADH, PEP, NADP,
and rabbit muscle lactic dehydrogenase were obtained from Calbiochem.
The Folin and Ciocalteu phenol reagent were from Harleco. -NAD+

was supplied by P-L Biochemicals, Inc. The following reagents were
obtained from Mallinckrodt Chemical Works: sodium tartrate;

anhydrous sodium acetate; acetic anhydride; and Amberlite MB-3.

l8
Difco Laboratories supplied the nutrient broth, and Sephadex G-25

was purchased from Pharmacia.

RESULTS

To show that the phosphorylation of galactitol with PEP is a
metabolically important reaction, it was necessary to preclude other
likely modifications of the galactitol molecule. These likely
modifications of the free hexitol include phosphorylation with ATP,

oxidation, or epimerization.

Absence of Galactitol Kinase, Galactitol Dehydrogenase, and
Galactitol Epimerase Activities. To test for the presence of these
activities, a culture of K. pneumoniae was grown overnight in mineral
medium containing galactitol. The harvested, washed cells were
sonically disrupted in 0.02 M Tris buffer (pH 7.5) or 0.02 M sodium
phosphate buffer (pH 7.5). The cell extract was then passed through
a Sephadex G-25 column, equilibrated with the same buffer, to remove
low molecular-weight metabolites that would interfere with the
assays. No activity could be detected for galactitol kinase (less
than 0.003 unit/mg protein) or galactitol dehydrogenase (less than
0.0002 unit/mg protein). During the assay for a galactitol epimerase,
large amounts of mannitol were formed in the incubations of up to
two hours (87% of the peak area of galactitol at two hours). However,
no glucitol was detected, and the amount of galactitol remained
unchanged throughout the incubation. Thus, epimerase activity was

not detectable (less than 0.0002 unit/mg protein). It seems

19

20
reasonable to conclude, therefore, that the kinase, dehydrogenase,
and epimerase reactions do not play a part in the initiation of
galactitol catabolism, and it seems very likely that PEP-dependent

L-galactitol l-phosphate formation is metabolically important.

Prediction of the Galactitol Pathway. From the presence of

 

an L-galactitol l-phosphate dehydrogenase, and by analogy to the
catabolic pathways for D-mannitol and D-glucitol discussed in the
Literature Review, one would expect that L-galactitol l—phosphate
would be oxidized to D—tagatose 6—phosphate. This molecule would

be phosphorylated to D-tagatose 1,6-diphosphate, which would in

turn be cleaved to yield dihydroxyacetone phosphate and
D-glyceraldehyde 3-phosphate. At this point, the triose phosphates
would enter and continue in the normal route of glycolysis. Thus,
the proposed enzymes would be L—galactitol l-phosphate dehydrogenase,
D-tagatose 6-phosphate kinase, and D-tagatose 1,6—diphosphate

aldolase.

Detection of the Pathway Enzymes. Assay of L-galactitol

 

l-phosphate dehydrogenase in the cell extracts of cells grown on
galactitol proved difficult because of the high levels of NADH
oxidase. The results of assays using NAD+ were inconsistent, while
no reduction of NADP+ was observed. Similarly, attempts to use
D—tagatose 6-phosphate and NADH as substrates were also hampered
by the masking activity of NADH oxidase. The use of the flavine
analog, quinacrine, at 0.6 mM, almost totally (98%) inhibited the
oxidase, but also quantitatively inhibited the dehydrogenase. It

was then found that if the cell extract was passed through a

21
DEAE-cellulose column, equilibrated in 0.02 M potassium phosphate
buffer at pH 7.5, the NADH oxidase would pass through the column,
while the dehydrogenase was adsorbed. Elution of the protein with
0.5 M KCl in the same buffer resulted in a preparation of the
dehydrogenase free of NADH oxidase. The results of assays with
this fraction can be seen in Table 1. It is apparent that the
reaction is dependent on L-galactitol 1-phosphate, and is specific
for NAD+. It also appears that the concentrations of NAD+ used are
less than saturating, as judged by its stimulatory effect at higher

concentrations.

Table l. Oxidation of L-galactitol 1—phosphate to D-tagatose
6-phosphate

 

 

Reaction mixture Units/mg proteinb
a
Complete 0.00037
Minus Legalactitol 1-phosphate 0.00000
, +
Minus NAD 0.00000
, + +
Minus NAD , plus NADP 0.00000
- +
Complete (0.1 umole NAD ) 0.00080
+
Complete (0.15 umole NAD ) 0.00134

 

aThe complete reaction mixture contained 10 umoles of Tris
buffer (pH 9.0), 1.0 umole of MgClz, 0.05 umole of NAD+, and 0.2
umole of L-galactitol l—phosphate in a volume of 0.15 ml.

bl unit = l umole product produced per min at 30°.

22
D-Tagatose 6—phosphate kinase was found to have a specific
activity of 0.013 unit/mg protein. Again, the competing reactions
of NADH oxidase, ATPase, and L-galactitol 1-phosphate dehydrogenase
require the activity to be measured above a background activity,
thereby diminishing the accuracy of the determination.
The specific activity of D-tagatose 1,6-diphosphate aldolase

was calculated to be 0.066 unit/mg protein (Table 2). As for the

Table 2. Cleavage of D-tagatose 1,6-diphosphate to triose phosphates

 

 

Reaction mixture Units/mg proteinb
Complete 0.249C
Minus kinasea 0.222d
Minus kinase,a minus D-tagatose 6-phosphate 0.183e
Minus D-tagatose 6-phosphate 0.185e
Minus cell extract 0.000

 

aMinus D-fructose 6-phosphate kinase.
b . .
1 unit = l umole of substrate cleaved per min at 30°.

CD-Tagatose 1,6-diphosphate aldolase activity (0.066 units/mg)
equals the total activity (0.249 units/mg) minus the activity of
NADH oxidase (0.183 units/mg).

dCombined NADH oxidase and L-galactitol l-phosphate dehy-
drogenase activities. L-Galactitol l-phosphate dehydrogenase
(0.037 units/mg) equals the combined activity (0.222 units/mg)
minus the activity of NADH oxidase (0.185 units/mg).

6Due to NADH oxidase activity.

23
other two enzymes, the competing activity was very high in this case
also. These data also indicate a value of 0.037 unit/mg protein
for the reduction of D-tagatose 6-phosphate with NADH; this value is

much higher than that reported above (Table 1) with L-galactitol

+
1-phosphate and NAD as substrates.

DISCUSSION

The data of this section have suggested that free galactitol
is not modified by a kinase, dehydrogenase, or epimerase, as these
three activities could not be detected in the cell extract. The
threshold for detection of the galactitol kinase is the highest of
the three because of the competing activities. The limits of
detection for the galactitol dehydrogenase and epimerase were ten-
fold more sensitive, making it more certain that these activities are
not present in the cell extract. Alpha-methylglucoside was used as
an internal standard for the epimerase assay since the only modi-
fication that can take place is phosphorylation by PEP-phosphotransferase
(55,66), and a Sephadex G-25—filtered extract would not have had this
capability. Thus, the amount of a-methylglucoside would remain con-
stant during the incubation, allowing comparison of the amount of
galactitol relative to this standard. It is assumed that the forma-
tion of large amounts of mannitol in the Sephadex G-25-filtered cell
extract was due to the breakdown of some polysaccharide.

Thus, it seems probable that the first step in catabolism is
phosphorylation with PEP as reported by Shimamoto (59). The product
has been unequivocally identified as L-galactitol l-phosphate. A
pathway for the further metabolism of this compound is now proposed
(Figure 2). In this proposal, L-galactitol l-phosphate is oxidized
by a dehydrogenase, with NAD+ as cofactor, to produce D-tagatose

24

.mcwsesamtm cNwawomNM ca Emflaonmumo HouwuomHmm How >m3cumm pomomowm .m mwomflm

mumsmmonmlm
opmsocamuoUSHOIQ

 

 

 

 

oumcmmonmfiplm.a mumcmmonmlo mumnmmonmla
mOmmU omoummmeln mmoummmela Houauomamota HOUfluomHmo
acmmo acmmo acmmo
r.)
2 I. I, l
030
I A] .I l
mumcmmonm
occuoom '- . I ll fl
1% on x a
x 6 s.o ao< mac mo<z omz use mom
I T
0' O l
o. acmmo

acmmo

26
6-phosphate. The D-tagatose 6-phosphate is then phosphorylated by
ATP, in a reaction catalyzed by a kinase, to yield D—tagatose
1,6-diphosphate. The diphosphate is then cleaved to dihydroxyacetone
phosphate and D—glyceraldehyde 3-phosphate by an aldolase. This
pathway is analogous to the catabolism of D—mannitol and D—glucitol
discussed in the Literature Review (Figure 1). It should be noted
that the last two steps of this proposed pathway are identical to
part of the D-tagatose 6-phosphate pathway of D-galactose metabolism
found in other organisms (61,62). Although K. pneumoniae does not
use the D-tagatose 6-phosphate pathway for the catabolism of
lactose and D-galactose (62), cells grown on galactitol have not
been previously assayed for the enzymes of this pathway.

The specific activities of the enzymes may be low, especially
for the dehydrogenase as measured with L-galactitol l-phosphate and
NAD+, but it is doubtful that optimal conditions for assaying the
enzymes were met by the arbitrary concentrations used in these
exploratory experiments. Optimal conditions would best be estab-
lished with partially purified enzymes. As will be shown in Sections
2 and 3, use of optimal assay conditions greatly increases the spe-
cific activity of these activities in cell extracts. The absence
of D—tagatose 1,6-diphosphate aldolase in cells of K. pneumoniae
grown on lactose, D-galactose, or D—glucose (62) could be explained
if the enzymes of this pathway are specifically induced by growth

on galactitol.

SECTION 2

PURIFICATION AND CHARACTERIZATION OF THE ENZYMES

INVOLVED IN THE CATABOLISM OF

L-GALACTITOL 1-PHOSPHATE

27

INTRODUCTION

As was pointed out in the Literature Review, it has been
established that in K. pneumoniae galactitol is phosphorylated
by PEP to form Legalactitol l-phosphate. Section 1 of this dis-
sertation has shown that cell extracts of this organism grown on
galactitol lack kinase, dehydrogenase, or epimerase activities for
the free hexitol. These data were interpreted to mean that forma-
tion of L-galactitol l-phosphate was the probable first step in
the catabolism of galactitol. Furthermore, preliminary evidence
was presented for the presence of L-galactitol l-phosphate dehy-
drogenase, D-tagatose 6-phosphate kinase, and D-tagatose 1,6-
diphosphate aldolase activities in these extracts. Based on the
presence of these activities, a pathway for the conversion of
L-galactitol l-phosphate to dihydroxyacetone phosphate and
D—glyceraldehyde 3-phosphate was proposed.

The purpose of this section is to provide further evidence in
support of the proposed pathway. The proteins catalyzing the
individual reactions were purified, and their properties examined.
The products of each reaction were also identified and shown to be
consistent with the proposed pathway. All relevant data are in

agreement with these enzymes carrying out such reactions in vitro.

28

MATERIALS AND METHODS

All materials and methods not described here were presented

in Section 1.

Growth Conditions. Cells were grown as in Section 1, except

 

that the volume was increased to 1 liter.

Determination of Substrate Concentration. For most assays the

 

concentration of substrate was determined on a weight basis. For
assays such as Km determinations, where the substrate concentration
is critical, the concentrations of L—galactitol 1-phosphate,
D-tagatose 6-phosphate, D-tagatose 1,6-diphosphate, D-fructose
6-phosphate, D-fructose 1,6—diphosphate, and NAD+ were determined
using an appropriate NAD+-1inked end-point assay. The concentration

of NADH was determined by its absorbance at 340 nm.

L—Galactitol l-Phosphate Dehydrogenase Assayy. The assays

 

(0.15 ml) contained: 10 umoles of Tris buffer (pH 8.5); 1.0 umole
of NAD+; 0.01 umole of MnClz; 0.45 umole of Legalactitol l-phosphate;
and rate-limiting amounts of L-galactitol 1-phosphate dehydrogenase.
The reverse reaction was measured in a volume of 0.15 ml containing
10 umoles of Tris buffer (pH 7.75), 0.01 umole of MnClz; 0.1 umole
of D-tagatose 6-phosphate, 0.05 umole of NADH, and rate-limiting

amounts of L—galactitol l-phosphate dehydrogenase.

29

30

D-Fructose 6-Phosphate Kinase Assays. The standard assays used

 

were developed by Sapico and Anderson (67). Each assay (0.15 ml)
contained: 10 umoles of glycylglycine buffer (pH 8.0); 0.5 umole of
ATP; 1.0 umole of MgClZ; 0.05 umole NADH; 1.0 umole of D-fructose
6-phosphate; excess D-fructose 1,6-diphosphate aldolase, triose
phosphate isomerase, and.a-g1ycerolphosphate dehydrogenase; and
rate-limiting amounts of D-fructose 6-phosphate kinase. The control
contained no ATP.

To show the competition between D—tagatose 6—phosphate and
D—fructose 6-phosphate, the following assay (0.15 ml) was used:
10 umoles of glycylglycine buffer (pH 8.0), 0.05 umoles of NADH,

1.0 umole of MgCl 0.5 umole of ATP, 0.5 umole of PEP, non-limiting

2]
amounts of coupling enzymes (D-fructose 1,6-diphosphate aldolase,
a-glycerolphosphate dehydrogenase, triose phosphate isomerase, and

pyruvate kinase), 0.8 umole of D-tagatose 6-phosphate, 4.26 umoles

of D-fructose 6-phosphate, and rate-limiting amounts of the kinase.

D—Tagatose 1,6-Diphosphate Aldolase Assays. The assays (0.15
ml) contained: 10 umoles of glycylglycine buffer (pH 8.0); 0.05

umole of NADH; 1.0 umole of MgCl ~ 0.5 umole of ATP; 0.2 umole of

2.
D-tagatose 6-phosphate; non-limiting amounts of D-fructose
6-phosphate kinase from rabbit muscle, triose phosphate isomerase,
and a-glycerolphosphate dehydrogenase; and rate-limiting amounts

of D-tagatose 1,6-diphosphate aldolase. The cuvettes were incubated
at 30° for 5 min before the addition of the aldolase so that the

ATP and rabbit muscle D-fructose 6-phosphate kinase could quanti-

tatively convert the D-tagatose 6-phosphate to D—tagatose

31
1,6-diphosphate (68). When assayed with D—fructose 1,6-diphosphate
as the substrate, 0.3 pmole of that compound was added to the assay,

and the ATP, MgCl , and rabbit muscle aldolase were omitted.

2

D—Tagatose 6-Phosphate Kinase Assays. The standard assays

 

(0.15 ml) contained: 10 umoles of glycylglycine buffer (pH 8.0);
0.05 umoles of NADH; 1.0 umole of MgClz; 0.5 umoles of ATP; 0.2
umole D—tagatose 6-phosphate; 0.075 umole of ADP; 5 umoles of KCl;

0.03 “mole of CoCl excess triose phosphate isomerase,

2;
a-glycerolphosphate dehydrogenase, and D-tagatose 1,6—diphosphate

aldolase; and rate-limiting amounts of D—tagatose 6-phosphate

kinase.

Alkaline Phosphatase Assays. The assays were as previously

 

described (69), except that they were scaled down to a volume of

0.15 ml.

Glucose 6—Phosphate Dehydrogenase Assays. The assays were

 

as previously reported (70), except that the volumes were scaled

down to 0.15 ml.

Lactate Dehydrogenase Assays. The assays were as previously

 

reported (71), except that the volumes were scaled down to 0.15 ml.

Pyruvate Kinase Assays. The assays were as reported (72),

 

except that the volumes were scaled down to 0.15 ml.

D-Fructose l-Phosphate Kinase Assays. The assays were as

 

described previously (67) with the addition of 5.0 umoles of KCl

(73).

32

D—Fructose 1,6-Diphosphate Aldolase Assays. The assays were the

 

same as for D-tagatose l 6-diphosphate aldolase when assayed with

D-fructose 1,6-diphosphate.

Ketohexose Determination. Ketohexoses were determined by the

 

method of Roe (74) as described by Ashwell (75). D-Fructose,
D—fructose 6-phosphate, or D-fructose 1,6-diphosphate was used as
a standard, depending on whether the compound being measured was a

free ketohexose, or its mono- or diphosphate derivative.

Phosphate Determination. Inorganic phosphate was determined by

 

a modification of the Fiske-SubbaRow procedure (76) as described by
Clark (77). For total phosphate analysis, the samples were

hydrolyzed overnight in l M HCl at 100° and assayed as above.

Bentonite Treatment. The enzyme solution was stirred in an

 

ice-water bath as the proper amount of bentonite was added slowly.
After 10 min, the solution was centrifuged at 12.000 x g, the

supernatant decanted and the pelleted material discarded.

Ammonium Sulfate Precipitation. The enzyme solution was

 

stirred at 0° as the desired amount of enzyme-grade ammonium sulfate
was slowly added to bring the concentration to the desired percent
of saturation (78). After the ammonium sulfate had dissolved, the
solution was stirred for an additional 15 min, and then centrifuged
at 12,000 x g_for 15 min. If the supernatant was to be treated
with additional ammonium sulfate, the above procedure was repeated.
Precipitated protein containing the desired enzyme activity was

dissolved in buffer following centrifugation.

33

Sephadex G-25 Chromatography. The gel was swollen by soaking in

 

water for 24 hours. When not in use, the gel was stored in 0.02%
(w/v) sodium azide at 4°. After the gel had been poured, the column
was washed with 4 volumes of the proper buffer. After use, the
column was washed with 0.02% sodium azide for storage stability.

Sample volumes were usually 5-15% of the bed volume.

Sephadex G-150 Chromatography. The gel was swollen in water
for 2 days, and then heated on a steam bath for 12 hours. The gel
was stored, and columns washed, as with Sephadex G-25. Sample
volumes did not exceed 5% of the bed volume for preparative work or

2% of the bed volume for molecular weight determinations.

Hydroxyapatite Adsorption. The powdered form of hydroxyapatite
was suspended in 0.25 M sodium phosphate buffer (pH 7.5) and left
at room temperature overnight. The hydroxyapatite was then washed
exhaustively with water by alternate centrifugation and resuspension.
Finally the pelleted mineral was suspended in the enzyme solution
containing not more than 5 mM phosphate buffer (pH 8.0). After 10
min, the hydroxyapatite was removed by centrifugation. Phosphate
buffer (0.1 M, pH 8.0) was used in an attempt to elute enzyme

activity from the hydroxyapatite.

Alumina §y_Adsorption. Aged alumina gel Cy was washed exten—

 

sively with water by alternate suspension and centrifugation. It
was then used to treat an enzyme solution as described for

hydroxyapatite.

34

Hexokinase Assays. The assays were as described previously

 

(79), except that all amounts were scaled down to a volume of

0.15 ml.

Protein Assays. Protein concentrations of greater than 0.5

 

mg/ml were determined by the method of Lowry et al. (63). Protein
concentrations of less than 0.5 mg/ml were determined by absorbance
at 220 nm (80). Bovine serum albumin was used as a standard with

both methods.

+
Preparation of NAD - or AMP-Sepharose. Sepharose 4-B was

 

activated with cyanogen bromide as described by Cuatrecasas and
Anfinsen (81). The activated Sepharose was then reacted with
adipic dihydrazide (82). The ribosyl moiety of the nucleotide was
oxidatively cleaved, and attached to the adipic dihydrazide-
Sepharose (82). Coupling was confirmed with the 2,4,6-

trinitrobenzene sulfonate test (81).

Dialysis. Dialysis tubing was prepared by boiling in a 0.2%
(w/v) solution of EDTA for 15 min. The tubing was then thoroughly
washed with distilled water. Dialysis was performed overnight at

4°, against at least 100 volumes of the appropriate buffer.

Polyacrylamide Gel Electrophoresis. Electrophoresis on 7.5%

 

polyacrylamide gels at pH 8.9 was performed as reported by Gabriel
(83). Following electrophoresis, the protein bands were stained

with Coomassie blue (84).

35

Paper Chromatography. Descending paper chromatography was done

 

in a water-saturated phenol solvent (85) with Whatman No. 1 paper
(23 x 56 cm). The paper was soaked in 0.1 M citric acid and allowed
to dry before use. Glycerol was used as a marker for this system
because it appears as a white spot, and migrates faster than the
ketohexoses. After the run, the paper was dried at room tempera-
ture, and the chromatograms developed for ketohexoses using an
orcinol spray (86) or for total carbohydrates with a silver nitrate

spray (87)-

Ketohexose Differentiation. The four different 2-ketohexoses

 

were distinguished from one another by the cysteine-H2804 reaction.
Sorbose was identified by its spectrum after reacting for 20 hours

(88), while fructose, tagatose, and psicose were distinguished

by the rate of color development (89).

Effect of NaBH4. Solutions of aldolase, in the presence and

 

absence of the ketohexose diphosphate substrates, were maintained
at pH 6.0 by adding 2 M acetic acid during the addition of NaBH4

at 0°. Samples were periodically removed for assay. The levels

of NaBH4 in the enzyme solution did not affect the assays.

Effect of NEM. Thiol protectants were removed from the solution

 

by chromatography on a Sephadex G-25 column equilibrated with the
proper buffer. The enzyme was treated with NEM for 30 min at 20°.
The small amount of NEM in the enzyme solution did not affect the

assay.

36

Dephosphorylation of Sugar Phosphates. Sugar phosphates were

 

dephosphorylated by the method of Bissett (64). The sugar phosphate
was combined with Sephadex G-25 filtered wheat germ acid phosphatase
in 80 mM sodium acetate buffer (pH 4.6) at 25°. The sample was then
desalted by passage through a mixed-resin column of Dowex 50W—X8

(H+) and Amberlite IR 45 (OH-); the resins were present in a roughly

1:1 proportion of basic to acidic exchange capacities.

Monovalent Cation Activation. Monovalent cations were removed

 

from substrates and cofactors by treatment with Dowex 50W-X8 (H+)
and neutralization with cyclohexylamine (this must be done very
rapidly with NADH because of its acid lability). Enzyme solutions
were chromatographed on Sephadex G-25 columns equilibrated with

Tris or cyclohexylammonium-neutralized phosphate buffers.

Divalent Cation Activation. To remove divalent cations, enzyme

 

solutions were treated with 10 mM EDTA for 15 min at room tempera-
ture, and passed over a column of Sephadex G-25 which had been
equilibrated with the proper divalent cation-free buffer. A similar
amount of untreated enzyme solution was then filtered through

Sephadex G-25 to act as a control for specific activity determination.

Determination of Dehydrogenase Equilibrium Constant. The
dehydrogenase equilibrium constant was determined by placing known
amounts of L-galactitol l-phosphate and NAD+ into cuvettes with the
standard assay components and 0.01 unit of L—galactitol 1-phosphate
dehydrogenase at 30°. The progress of the reaction was continually

monitored at 340 nm. When the reaction had reached completion,

37
the amount of NADH formed was calculated from the increase in
absorbance. As will be seen later in Section 2, the oxidation of
L—galactitol l-phosphate is coupled to the reduction of NAD+ with
a reaction stoichiometry of 1:1. Since the amount of D-tagatose
6—phosphate formed equaled the NADH, these values could be sub-
tracted from the initial concentrations of the substrates, giving
the final concentrations of NAD+ and L-galactitol l-phosphate.
Determinations were made with three different initial amounts of
substrates at each of two pH values. The pH of the assay mixtures
was checked after each of the determinations to make sure that it

had not changed.

Source of Materials. The following were purchased from Sigma

 

Chemical Co.: D-fructose 6-phosphate; D-fructose 1,6-diphosphate;
D—galactose 6-phosphate; E. coli alkaline phosphatase; D—glucose
6-phosphate; D-fructose 1-phosphate; yeast hexokinase; alumina cw
gel; 2,4,6—trinitrobenzene sulfonate; Coomassie brilliant blue;
orcinol; cysteine; DTT; DTE; BME; AMP; HEPES; TTP; GTP; and
diethylacetyl—D—glyceraldehyde 3-phosphate. Resorcinol and sodium
meta periodate were purchased from Mallinckrodt Chemical Works.
Enzyme-grade ammonium sulfate was purchased from Schwarz/Mann,
while NADPH and Sephadex G-150 were purchased from P-L Biochemicals
and Pharmacia, respectively. The acrylamide, ammonium persulfate,
and N,N,N',N'—tetramethylethylenediamine used for electrophoresis
were purchased from Canalco, Inc. Paper used for chromatography
was Whatman grade 1. Adipic dihydrazide and cyclohexylamine were

purchased from Eastman Organic Chemicals. The following were

38
purchased from the indicated sources: cyanogen bromide, Matheson,
Coleman, and Bell; D-mannose 6-phosphate, Nutritional Biochemicals
Corp.; rabbit muscle D-fructose 1,6-diphosphate aldolase, Calbiochem;
wheat germ acid phosphatase, Worthington Biochemical Corp.;
bentonite, Fischer Scientific Co.: hydroxyapatite HTP, Bio Rad
Laboratories. The following compounds were donated by the indi-
cated person: D-tagatose, Dr. H. Lardy; L-sorbose l-phosphate,
Dr. Ronald Simkins; and D—mannitol l-phosphate and D-glucitol
6—phosphate, G. T. Shimamoto. D-Fructose l-phOSphate kinase was

purified as described by Sapico and Anderson (67).

RESULTS

A. L—Galactitol l-Phosphate Dehydrogenase (L-Galactitol
1—PhosphatezNAD+ Oxidoreductase)

 

 

Purification. During the course of purification of L-galactitol

 

l-phosphate dehydrogenase, some techniques, methods, or conditions
were found to be ineffective or counterproductive. Substantial
losses of activity were found when the enzyme was stored overnight
in buffers with a pH less than 8.0. Acid precipitation of the
activity at pH values less than 6.0 resulted in irrecoverable loss
of activity. The activity was not adsorbed to hydroxyapatite and
alumina gel in 5 mM sodium phosphate buffer (pH 8.0) or Tris buffer
(pH 8.5). Addition of bentonite caused loss of activity at high
concentrations, and only slight purification at low concentrations.
The activity failed to be adsorbed to, or be retarded on, columns
of NAD+-Sepharose or AMP-Sepharose in 20 mM Tris buffer (pH 8.0

or 9.0) in the presence or absence of 15 mM L-galactitol l-phosphate.

Cell extract. The washed cells from 1 liter of galactitol-

 

grown K. pneumoniae were suspended in 0.025 M Tris buffer (pH 8.5)
containing 10% (v/v) glycerol and 0.5 mM DTE. The cell extract was
prepared by sonication. All purification steps were carried out at

0-4°. A summary of the purification is found in Table 3.

39

40

.w ca we >wm>oomm©

.oom on use Mom coauom poopowm mo oHOE: H u was: ac

.mE as me camcoumn

.HE as we esoao>m

 

 

 

om sm.s om.m v.m e.s omeno xmcmsomm

mo moo.o mm.m m.oH N.m mummfiSm ESHCOEEm

on mmm.o m.aa mm mm omCHsHHoolm<mo

ooa mmo.o m.mH 0mm. m.Hm uomuuxo Haoo

pwwo>oomm oeououm m5 oxufl>wuom mo ncfiououm moESHo> GOauomwm
woes: woes: Hmuoe Houoa

 

outcomowtmcop muonmmocmua Houauomammia wo coflumowwewsm .m wand?

41

DEAE-cellulose chromatography. The cell extract was

 

loaded onto a l.2 x 6 cm DEAR-cellulose column that had been
equilibrated with 0.025 M Tris buffer (pH 8.5) containing 10% (v/v)
glycerol and 0.5 mM DTE. A 200 ml linear gradient of 0-0.35 M

KCl in the above buffer was used to elute the protein from the
column (Figure 3). Fractions of 2 ml were collected. Fractions
28 to 43, containing the peak of the activity, were pooled for

further purification.

Ammonium sulfate precipitation. To the 28 ml of the

 

pooled activity was slowly added 5.38 g of ammonium sulfate to
bring the concentration to 35% of saturation. The solution was
stirred for 15 min, and then centrifuged at 12,000 x g for 10 min.
The 30 ml of supernatant was decanted, and 6.6 9 more of ammonium
sulfate added to bring the concentration to 70% saturation. After
15 min, the solution was centrifuged at 12,000 x g for 10 min.

The supernatant was discarded, and the pelleted material dissolved
in 2 ml of 0.025 M Tris buffer containing 10% (v/v) glycerol and

0.5 mM DTE.

Spphadex G-150 chromatography. The 35-70% ammonium-

 

sulfate-precipitated protein was layered onto a 1.4 x 82 cm column
of Sephadex G-150 equilibrated with 0.025 M Tris (pH 8.5) containing
l0% (v/v) glycerol and 0.5 mM DTE. The column was eluted with the
same buffer. Fractions of 1 ml were collected. Fractions 55-60
(Figure 4) containing the peak of the activity were pooled. This
solution of L—galactitol l-phosphate dehydrogenase was 25-fold

purified, and contained 20% of the initial activity. D-Tagatose

42

(rm/6m) utaqoxd

.HE m we mEDHo>
cofluomwm .caououm anv .ommcmmouoxcoo mumsmmo Qua Mo auoo cold A 0V "maonewm
.mmoHoHHmOImmmo co ommcomouownoo oumcmmocmla Houauomam IQ we came umeownu .m ouomfim

wonfioz coauomum
om ow on 00 om 0: on om OH

« J 1 d a H u .o o d

 

wor

m.H

 

o.mr.

 

 

 

(rm/sqrun) aseuaboxpfiqaq aqequoqa—I {onrqoetes-q

43

(Tm/6m) utaqoxd

H.o
m.o

n.o

t\
O O

U)

cofluomum

 

.caououm AOV

.HE H ma mEoHo>

.ommcooouoxnoo oumnmmozmla Houeuomammla AOV
.omalo xmpmcmmm so ommcomoupwnoo oumcmmonmta HOufiuomHmmlq mo annoumoumEouno

umnEoz coauomum

om

0:

on

 

umHonE>m

.+.

musmflm

H
O
toqtnoetee-q

n. m.o

I

l

M

O
nequoqa-

l
:2'
O

 

assuaBOJpAqaq a

 

(rm/Saran)

44
6-phosphate kinase or D-tagatose 1,6-diphosphate aldolase could not

be detected (less than 0.005 unit/ml).

Properties

 

Stability. The dehydrogenase was found to retain most of
its activity when stored at 0° or at -20° for at least two weeks.
The presence of glycerol and DTE were found to be essential for
good recovery (greater than 85%) of activity for periods of time as

short as 12 hours.

pH optima. The pH optimum for the dehydrogenase was
measured for both the oxidation and reduction of sugar phosphate
(Figure 5). The optimum for the reaction using L-galactitol
1—phosphate and NAD+ as substrates was pH 8.75, while the optimum
with D-tagatose 6—phosphate and NADH as substrates was pH 7.75.
In both cases, Tris buffer gave higher activity than glycylglycine

buffer at all pH values tested.

Substrate specificipy. L-Galactitol l-phosphate and

 

D-tagatose 6-phosphate were the only compounds found to serve as
substrates for the dehydrogenase. The following compounds were
tested as substrates for the dehydrogenase at concentrations of

10 mM, with NAD+ as the cofactor, with no activity (<0.0006 unit/
mg) detectable: D—glucose 6-phosphate, D-mannitol l-phosphate,
D-glucitol 6-phosphate, D—fructose l-phosphate, and D-galactose
6-phosphate. Similarly, the following compounds were tested, using

NADH as cofactor, without success: D-galactose 6-phosphate,

45

 

100 '

90 '

80‘

% of Maximum Activity

30~ .

20 P

 

 

l J, 1 1 l
7.0 7.5 8.0 8.5 9.0
pH

 

Figure 5. pH optimum of L-galactitol l-phosphate dehydro-
genase. Symbols: (0) Tris and (O) glycylglycine with
L-galactitol l-phosphate and NAD as substrates; and

(A) Tris and (A) glycylglycine with D-tagatose 6-phos-
phate and NADH as substrates. The maximum Specific ac—
tivities obtained were 1.66 units/mg protein with L-galac-
titol l-phOSphate as substrate, and 0.41 units/mg pro—
tein with D-tagatose 6-phosphate as substrate.

46
D-glucose 6—phosphate, D-fructose l-pbosphate, D-fructose
1,6-diphosphate, D—fructose 6-phosphate, and L—sorbose l-phosphate.
All attempts to detect activity using NADP+ (6.7 mM) or NADPH

(0.33 mM) as the pyridine nucleotide cofactor failed.

Cation activation. The dehydrogenase was assayed in the

 

presence of the following monovalent cations at a final concentra-

tion of 33 mM: NaCl, KCl, NH Cl, LiCl, RbCl, and CsCl. None of the

4
cations caused a significant increase or decrease in the enzyme
activity. On the other hand, divalent cations were shown to have a
great effect on the activity of EDTA-treated dehydrogenase (Table
4). Without added divalent cations, the enzyme had no detectable
activity. The cations causing the greatest restoration of activity
were Cd++, Mn++, and Co++. The restoration of 36% of the control
activity by 6.7 x 10—5 M MnCl2 took place 15 min after the Sephadex

G-25 filtration, but if the same concentrations of MnCl2 were added
after 2.5 hours, only 4% of the control activity was restored. The

enzyme is apparently destabilized by the EDTA treatment and subse-

quent storage in divalent cation-free buffer.

Kinetic constants. The concentrations of L-galactitol

 

l—phosphate and NAD+ were varied, each in the presence of saturating
amounts of the other. The Km values obtained were 0.22 mM for NAD+
and 0.42 mM for L-galactitol l-phosphate (Figure 6). When the
experiment was carried out with D—tagatose 6-phosphate and NADH,

the Km values were 0.10 mM and 0.025 mM, respectively (Figure 7).

47

Tablv 4. Divalent cation activation of EDTA-treated L-galactitol
l-phosphate dehydrogenase

 

Specific activity (% of control)

 

 

Divalent cation 6.7 x 10-6 M 6.7 x 10-5 M 6.7 x 10-4 M
caso4 3 42 53
Mnc12 1 36 41
CoCl2 ND 30 41
ch12 ND 15 14
NiCl2 ND ND 2
MgCl2 ND ND 4
BaCl2 ND ND 2
Cac12 ND 2 14
01504 ND ND ND
FeSO4 ND ND ND

 

N.D.: The activity was not detectable (less than 1% of the control).

 

48

 

 

 

 

 

 

 

 

 

3 P
l
t: o 2 '
O E
o s:
H \
a) c
> ml
\ E
H V 1 p
I l 1 I
’4 8 12 16
vpmg (mm‘l)
" 2
a :2 t
«4 0
U E
<3 G
H \
8 .5
\ e 1 -
H V
l j 4 J 1
1 2 3 ll 5
1/ E-Galactitol 1-phosphatg (mm‘l)
Figure 6. Lineweaver-Burk plots of L-galactitol 1—

phOSphate dehydrogenase with L-galactitol l-phosphate
and NAD'+ as substrates.

49

 

4r
1

\N
I

/veloc1ty
(min/nmole)
tu
I

 

 

 

- r A; 5
30 60 90 120
1/ [NADH] (mM’ 1)

 

 

1
U1 ox
l

   

l/velocity
(min/nmole)

  
   

 

 

I J I

I
10 20 30 HO
1/ p-Tagatose 6-Ph05phata (mM-

 

 

1)

Figure'7. Lineweaver-Burk plots of L-galactitol l—phos-
phate dehydrogenase with D-tagatose 6~ph05phate and

NADH as substrates.

50

Molecular weight determination. The molecular weight of

 

the dehydrogenase was determined by chromatography on a 1.45 x 80
cm column of Sephadex G-150, with standard proteins of known
molecular weight. Fractions of 1 ml were collected. The standard
proteins and their molecular weights are: rabbit muscle D—fructose
1,6-diphosphate aldolase, 150,000 (90); yeast D-glucose 6-phosphate
dehydrogenase, 128,000 (90); E. coli alkaline phosphatase, 86,000
(91); and yeast hexokinase, 51,000 (92). When the fraction contain-
ing the peak of activity for each protein was plotted against the
log of the molecular weight, a linear relationship was observed
(Figure 8). The peak of the dehydrogenase eluted concidently with
the peak of alkaline phosphatase, indicating a molecular weight of
86,000. This method of molecular weight estimation assumes a

spherical shape for the protein molecule.

Reaction Characteristics

 

Product identification. The product of the oxidation of

 

L-galactitol 1-phosphate was identified as D-tagatose 6-phosphate
by Shimamoto (59), using L-galactitol l-phosphate dehydrogenase
supplied by this investigator. The product was identified by means
of gas-liquid chromatography of the trimethylsilyl derivative
before and after dephosphorylation, acid lability, ketohexose to

phosphate ratio, periodate oxidation, and enantiomeric configuration.

Stoichiometry. For each mole of L—galactitol l-phosphate,

 

+
as measured by the organic phosphate assay, 0.94 moles of NAD was

reduced to NADH when the reaction was allowed to run to completion

Molecular Weight

51

 

150,000 P‘

  
  
    

GPDH L-Galactitol l-
Phosphate

Dehydrogenase
100,000 '-
-
80, 000 "'
70,000 '
60,000 ~—
50,000 -

40,000 -

 

 

r n 1 1 1 1 l l
48 50 52 54 56 58 60 62
Fraction Number

 

Figure 8. Molecular weight determination of L-galac-
titol l-phosphate dehydrogenase. The molecular weight
markers are: RMA, rabbit muscle D-fructose 1,6-di-
phosphate aldolase; GPDH, D-glucose 6-phosphate de-
hydrogenase; AP,.E. coli alkaline phosphatase; and

HK, hexokinase,

52
n l I U +
at pH 9.0. Thus, the reaction stOichiometry is one mole of NAD

reduced for each mole of L-galactitol l-phosphate oxidized.

Determination of equilibrium constant. The equilibrium

 

constant was determined by placing known amounts of NAD+ and
L—galactitol l-phosphate into cuvettes with 0.01 units of the
dehydrogenase, buffered at one of two different pH values. The
absorbance at 340 nm was monitored and the reaction allowed to go
to completion at 30°. From the increase in absorbance, the amount
of NADH formed could be calculated from its molar absorbtivity.

As shown by the reaction stoichiometry, the amount of D-tagatose
6-phosphate formed was equal to the amount of NADH, and both could
be subtracted from the initial concentrations of NAD+ and
L-galactitol l-phosphate, to give the final concentrations of these
two substrates. Determinations were made at both pH 8.5 and pH 9.0

(Table 5). The equation used for equilibrium constant is

+
[NADH][D-tagatose-6-P][H ]

 

eq = [NAD+][L-ga1actitol-l-P]
The mean value for these determinations was 8.0 x 10.11 M with a
. . -11
standard deViation of 1.1 x 10 M.

Discussion. The data presented for the partially purified

 

L-galactitol l-phosphate dehydrogenase establish that this enzyme
is indeed capable of functioning in the prOposed pathway, oxidizing
L-galactitol 1-phosphate to D-tagatose 6-phosphate. D-Tagatose
6-phosphate has conclusively been shown to be the product of this

oxidation by Shimamoto (59). My data have also shown that the

53

 

 

 

 

HHIOH H.H cofiumfl>op ouoocoum
HHIOH o.m mDHm> cmoz
HHIOH o.m z ouoa x om.mm mica m.mo Z oioa x o.mma 0.0
Hauos m.m z oaoa x om.os macs «.ms 2 ouoa x m.mo o.m
HHIOH b.o z onoH x cv.0H mnoH o.nm z mica x m.mo o.m
HHIOH m.m z mica x oo.vH mica m.mn 2 area x o.mma m.m
HHIOH h.h z mica x mo.m oioa N.mo z mica x m.mo m.m
HHuoH m.o z mica x om.o oioa o.hm z mica x m.mo m.m
om Hmmflzg ”mumsmmoxmla HoufluomHmmtqa H+szH mm
Hoses HmauHcH HofiuwcH
QOwuommu
ommcomowo>soc ouccmmonmua Houauomammlq one now ucoumcoo Eofiunaafioom ecu mo coauMCHEhouoo .m magma

54

dehydrogenase was Specific for NAD+ as its pyridine nucleotide co-
factor, and the reaction had a stoichiometry of one NAD+ reduced
for each molecule of L—galactitol l-phosphate oxidized to D-tagatose
6-phosphate. The Km value for L-galactitol 1-phosphate (0.42 mM)
seems reasonable for a reaction that physiologically follows a
phosphorylative step, with its resultant large decrease in free
energy. The K.m value for NAD+ (0.22 mM) is higher than for most
dehydrogenases, but is not without precedence [e.g., D-mannitol
l-phosphate dehydrogenase (0.38 mM) and D—glucitol 6—phosphate
dehydrogenase (0.42 mM) (47)]. The specificity of the dehydrogenase
appears to be rather high, with only L-galactitol l-phosphate and
D-tagatose 6-phosphate acting as substrates, out of all the com-
pounds tested.

The activity was unaffected by monovalent cations, whereas
divalent cations were necessary for activity and stability.
Although Cd++ gave the highest amount of activity, Mn++ was used
in routine assays of activity because its greater natural abundance
led me to assume it would more likely be the physiologically
important cation.

As discussed in the Literature Review, it was reported that
D—glucitol 6-phosphate dehydrogenase from A. aerogenes is capable
of oxidizing L—galactitol l-phosphate (47), presumably to D—tagatose
6—phosphate. It can be clearly shown that the L-galactitol
l-phosphate dehydrogenase reported here is not in fact the result
of the non-specificity of D—glucitol 6-phosphate dehydrogenase.
First, the L-galactitol 1-phosphate dehydrogenase does not use

D-glucitol 6-phosphate as a substrate. Second, D—fructose

55
6—phosphate does not serve as a substrate, as would be expected for
D-glucitol 6—phosphate dehydrogenase. And third, as will be seen
in Section 3, L—galactitol l-phosphate dehydrogenase is specifically
induced by growth on galactitol, and was not detected in extracts
of cells grown on D-glucitol, which induces the D—glucitol
6-phosphate dehydrogenase (50).
B. D—Tagatose 6-Phosphate Kinase [ATP:D-Fructose

6-Pho§phate (D—Tagatose 6-Phosphate) l-Phospho-
transferase] (EC 2.7.1.11)

 

Purification. During early attempts at purifications of the

 

D-tagatose 6-phosphate kinase activity, some methods were found to
be ineffective. These procedures are as follows. No binding or
retardation was found when the activity was passed through a column
of AMP—Sepharose in either the presence or absence of D-fructose

6-phosphate (l or 10 mM), with and without 1 mM MgCl Treatment

2.
with hydroxyapatite resulted in a loss of most of the activity,

which could not be recovered by elution with 0.5 M phosphate buffer.
The enzyme did not adsorb to alumina gel Cy in 5 mM phosphate

buffer. Adjustment of the pH to 5.0 with 2 M acetic acid resulted

in high recovery only when the protein concentrations were kept

fairly high (about 10 mg/ml or greater), but gave little purifica-
tion beyond that of the bentonite treatment. High protein concen-
tration was also necessary for the high recoveries reported previously
with Sephadex chromatography and ammonium sulfate precipitation (67).
The use of a heat step (5 min at 60°) gave excellent recovery of

activity, but little purification was achieved beyond that of the

bentonite treatment.

56

Cell extract. The washed cells from 1 liter of a

 

nutrient-broth-grown culture were suspended in 20 m1 of 0.02 M
sodium phosphate buffer (pH 7.5) containing 10% (v/v) glycerol.
The cells were sonically disrupted. Whole cells were removed by
centrifugation at 12,000 x g_for 10 min. The extract was then
diluted to 40 ml with the above buffer. All the following steps
took place at 0-4°. A summary of the purification is presented

in Table 6.

Bentonite treatment. To the stirred cell extract was

 

slowly added 1.2 g of bentonite. The mixture was stirred for 10
min, and then centrifuged at 12,000 x g for 10 min. The supernatant

was decanted and reserved.

DEAR-cellulose chromatography. The bentonite supernatant
was passed through a l x 8 cm DEAR-cellulose column that had been
previously equilibrated with 20 mM sodium phosphate buffer (pH 7.5)
containing 10% (v/v) glycerol. The column was washed in the same
buffer. A linear gradient of O to 0.2 M ammonium sulfate in 160 m1
of the above buffer was used to elute the adsorbed protein. Eluted
fractions of approximately 2 ml were collected. The elution profile
can be seen in Figure 9. Fractions 43-62 were pooled for further
studies, and dialyzed against 50 mM sodium phosphate buffer (pH
7.5) containing 10% (v/v) glycerol and 0.1 M ammonium sulfate.

The enzyme was 15.6-fold purified following the DEAR-cellulose
step, with a recovery of 66%. The cells were grown on nutrient
broth, rather than a medium containing galactitol, so there was no

detectable L-galactitol l-phosphate dehydrogenase or D-tagatose

.mmmcax
mumzmmonmlo mmoummmuno cu meCflx mumnmmormno mmOHUSMMIQ mo wufl>fluom Uflwwommm mzu mc ofiumms
a

.w ca ma >um>oomm©

.oom um CHE Mom Uwfihow uosvoum m0 mace: H u ufics Ho

.05 CH ma Camuoumn

.HE ca ma mEDHo>m

 

57

 

 

m.H be mmh.o m.m m.HH ov mmoHsHHmonmmmo

o.H moa 0mm.o v.me ow m.vm wuwcoucmm

m.H OOH mmo.o H.ma mvm ow uomuuxm Hamu

mxmoe\wmwm ©>Mm>oomm cflwuonm ms oupfl>fiuom ncflmuoum mmEsHo> coflucmum
mafia: Hmuoe Hmuoe

 

mmmcflx wumzmmocmuo mmoummmulo mo coHumoflmawsm .m magma

58

.cflmuoum AOV can ammmcflx muwzmmonmlo mmoummmuln A‘V ammmcfix
mumnmwonmlw mmouosuMIo AOV undoneam .mmoasaamUImfimQ co mmmcfix wumnmmonm
Io wmoummmula paw mmmcflx mumnmmonmlo mmouUDHMIQ mo mammumoumeouno .m wusmwm

HmQEDZ COflHUMHm

 

2. oo om 0: on cm 3
q d d u q u q
1 H.o
.0 O O O O o. O o

J
N
O

o
J
N\
C)

l
:1-
C)

o
1*
Ln
C)

 

f
1
\o
o"
(Im/sqtun) aseurx eqequoqd-g esoqefiem-q

(Im/sqrun) aseurx aqequoqa—g esoqonJa—Q

l
5.
C)

 

 

r

(rm/6m) ureqoxa

59
1,6-diphosphate aldolase in the cell extract. Because of the absence
of these two other enzymes from the galactitol catabolic pathway,
as well as the competing NADH oxidase activity, it was not deemed
necessary to further purify the kinase. The constant specific
activity ratios for D-fructose 6-phosphate kinase and D-tagatose
6-phosphate kinase during purification, as well as their coincident
elution from DEAR—cellulose, suggest that these two activities are

catalyzed by the same enzyme.

Properties

 

Stability. The DEAE-cellulose fractions were stable when
kept on ice, having a half-life of approximately two weeks. Samples
stored frozen at -20° for a month showed little loss of activity.
Storage of the enzyme in 20 mM sodium phosphate buffer (pH 7.5) con-
taining 10% (v/v) glycerol for 24 hours at 0° or 4° caused 79% and

100% loss of activity, respectively.

Thermal inactivation. Samples of the kinase treated in a

 

water bath at 64° for various times were assayed for activity with
both D-fructose 6-phosphate and D-tagatose 6-phosphate (Figure 10).
These data show that the loss of both activities is concident, sup-

porting the contention that both activities reside in one protein.

pH optimum. For the D-fructose 6-phosphate kinase, as

 

well as the D-tagatose 6-phosphate kinase activities, the highest
activity occurred in Tris buffer, with the optimum at pH 8.25
(Figure ll). Use of glycylglycine gave an optimum of pH 8.0, while

potassium phosphate buffer shifted the optimum to pH 7.75.

60

 

100

90

80

7O

6O

50

40

% of Activity Remaining

30

2O

 

 

 

l l l l 'L L J
5 10 15 20 25 30 35 40
Minutes at 64°

 

Figure 10. Thermal inactivation of D-tagatose 6-phosphate
kinase and D-fructose 6-phosphate kinase. Symbols: ((5)
D-tagatose 6-phosphate kinase; (0) D-fructose 6-ph03phate
kinase.

61

 

100'-

90"

80-

K]
O
I

 

O\
O
F

Relative Activity
U1
<3
l

.t'
C)
I

\N
C)
l

10*-

 

 

] n l l I 1 l

6.0 6.5 7.0 7.5 8.0 8.5 9.0
pH

 

Figure 11. pH optimum of D-fructose 6-ph03phate kinase and
D-tagatose 6—phosphate kinase. Symbols: D-fructose 6-
phosphate kinase in Tris (O ), glycylglycine (A), and
KP04 (0): and D-tagatose 6-phosphate kinase in Tris (A).

62

Phosphoryl acceptor specificity and kinetic constants. As
has already been shown, both D-fructose 6—phosphate and D-tagatose
6-phosphate act as substrates for the kinase. Plots of substrate
concentration versus activity (Figures 12 and 13) show a sigmoidicity
which indicates allosteric behavior toward these substrates (93).
Substrate concentrations that gave half-maximal velocity were 0.25
mM and 1.1 mM for D-fructose 6-phosphate and D-tagatose 6-phosphate,
respectively, in the absence of ADP, and 0.4 mM for D—tagatose
6-phosphate in the presence of ADP. The value for D-fructose
6—phosphate is in good agreement with the published value of 0.3 mM
(67). It can also be seen that the Vmax for D-tagatose 6-phosphate
is 62% of that for D—fructose 6-phosphate. This is in good agree-
ment with the data for pH optimum where in Tris buffer the D-tagatose
6-phosphate kinase activity was 58% of the D-fructose 6-phosphate
kinase activity at the optimal pH (Figure 11).

To establish that these two substrates are competing for the
same activity, the following experiment was devised. The standard
D-fructose 6-phosphate kinase assay using non-limiting amounts of
rabbit muscle aldolase was employed, except that the D—fructose
6-phosphate concentration was made 1.04 mM (four times its Km
value), and pyruvate kinase and 3.3 mM PEP were added to remove all
ADP. When D—tagatose 6-phosphate was added to a final concentration
of 5.33 mM (approximately five times its Km value), the rate of
D-fructose 1,6-diphosphate formation was inhibited 46%. It was
possible to measure formation of D—fructose 1,6-diphosphate inde-
pendently from formation of D-tagatose 1,6-diphosphate because of

the rabbit muscle D-fructcse l,6-diphosphate aldolase specificity

(umoles/min)

D-Fructose 1,6-Diphosphate Formed

63

 

0.7

 

 

l

C
d»
l

I L

 

1.0

2.0 3.0

D-Fructose 6-Phosphate (mM)

Figure 12. The effect of D-fructose 6-phosphate concen-

tration on the kinase activity.
57.5 pg of the purified kinase.

The assay contained

 

()4

 

 

 

D-Tagatose 1,6-Diphosphate Formed (nmoles/min)

 

 

 

 

l L i l
1 2 3 4 5
D-Tagatose 6-Phosphate (mM)

Figure 15. The effect of D-tagatose 6-phosphate concen-
tration on the kinase activity. Symbols: (0) minus
ADP, (o ) plus 1 mM ADP. The assay contained 67.6 ug

of the purified kinase.

65
for the former (61,94). Because the presence of both substrates in
the cuvette would change the kinetic constants from those obtained
with either substrate present alone, there are not sufficient data
with which to calculate the expected percent inhibition. However,
the fact that a considerable inhibition does occur suggests that
the two sugar phosphates are substrates for the same enzyme.

Other substrates of the kinase would also be expected to cause
an inhibition of the phosphorylation of D-fructose 6-phosphate. The
following compounds were added to the standard D-fructose 6-phosphate
kinase assay at a concentration of 10 mM to observe if any would
cause inhibition as did D-tagatose 6-phosphate: L-sorbose
1—phosphate, L-galactitol l-phosphate, D-galactose 6—phosphate,
D-glucose 6—phosphate, D-mannose 6-phosphate, D-fructose l-phosphate,
and D-fructose. None of these compounds affected the rate of
D-fructose 1,6-diphosphate formation, indicating that they are not
substrates for the kinase. In addition, the formation of D-fructose
l-phosphate from D—fructose (measured with D-fructose 1-phosphate

kinase) could not be detected.

Phosphoryl donor specificity. The relative rates of

 

phosphorylation of D-fructose 6-phosphate by the six common nucleo-
side triphosphates has been reported previously (67). Their
efficiency in the standard assay at a concentration of 3.3 mM was
(%): ATP, 100; ITP, 93; GTP, 86: UTP, 57; CTP and TTP, 50. Three
of these were tested in the standard D-tagatose 6-phosphate kinase
assay with the following results (%): ATP, 100: ITP, 94; and GTP,

84. Thus, it is seen that the results for the phosphorylation of

66
D-tagatose 6-phosphate parallel the results obtained with

D-fructose 6-phosphatt.

Sulfhydryl reguirement. The thiol reagent NEM was incu-

 

bated with the enzyme at a final concentration of 0.2 mM for 30 min.
No loss of activity was noted. This is in contrast to the
D—tagatose 6-phosphate kinase of S. aureus, which was 85% inhibited
under identical conditions. The addition of BME or DTT to the
buffer during my attempts at purification did not enhance the
recovery over attempts without these compounds. Thus, there is no

apparent sulfhydryl requirement for the kinase.

Molecular weight. The kinase was applied to a Sephadex

 

G-150 column along with several proteins of known molecular weight
(92). The standards used, and their molecular weights, were as
follows: rabbit muscle lactate dehydrogenase, 140,000 (90,95);
yeast D-glucose 6-phosphate dehydrogenase, 128,000 (90): E. coli
alkaline phosphatase, 86,000 (91); and K. pneumoniae D-fructose
l-phosphate kinase, 75,000 (67). When the log of the molecular
weights of the standards is plotted against the fraction of each
with the greatest activity (Figure 14), a linear function is evident.
This plot indicates the kinase has a molecular weight of 92,000.
This determination assumes a spherical shape for the protein

molecule.

Product Identification

 

Production of product. To produce the product of the

 

kinase reaction for identification, a reaction mixture (15 ml) was

67

 

 
  

 

 

150,000 ' LDH
g GPDH
m D-Tagatose 6-Phosphate
.3 Kinase
€100,000 -
55; 90,000"
0 80,000 P
,9, _ FPK
0 70,000
2

60,000 -

n L 1 l l L l A

 

46 48 50 52 54 56 ‘58 60
Fraction Number

Figure 14. Molecular weight determination of D-tagatose
6-phosphate kinase. The standard proteins are lactate
dehydrogenase (LDH), D-glucose 6-phosphate dehydrogen-
ase (GPDH), E. coli alkaline phosphatase (AP), and
D-fructose l-phosphate kinase (FPK).

68
prepared containing 1.0 mmole of glycylglycine buffer (pH 7.5),

100 umoles of MgCl 50 umoles of ATP, 50 umoles of D-tagatose

2’
6-phosphate, and D-tagatose 6-phosphate kinase. The reaction was
monitored by removing samples at timed intervals, and assaying for

the amount of ADP present. When the formation of ADP had stopped,

the reaction mixture was adjusted to pH 8.5 with 6 M NH4OH.

Isolation of thepproduct. The reaction mixture was

 

immediately applied to a Dowex l-X8 chloride column (1.1 x 18 cm).
The column was washed successively with the following solutions as

described by Khym and Cohn (96): 0.001 M HN4OH; 0.025 M HN4

0.01 M K2B4O7; 0.025 M NH4C1, 0.0025 M NH4OH, 0.00001 M KZB4O7;

0.01 M HCl, 0.02 M KCl; and 0.02 M HCl, 0.2 M KCl. Fractions of the

Cl,

0.02 M HCl, 0.02 M KCl eluate that contained ketohexose were com-
bined, lyophilized, dissolved in 10 ml of water, and the pH adjusted

to neutrality with NaOH.

Identification of the product. The isolated ketohexose

 

product was identified as D-tagatose 1,6-diphosphate by the following
criteria: (i) it eluted from the Dowex l-X8 column in the fractions
where sugar diphosphates normally elute, (ii) the assayed phosphate
to ketohexose ratio was 1.93:1, (iii) the product was cleaved by
D-tagatose 1,6—diphosphate aldolase, as was authentic D-tagatose
1,6-diphosphate produced in the aldolase assay mixture from
D-tagatose 6-phosphate as described in Materials and Methods, (iv)
cleavage of the product with D-tagatose 1,6-diphosphate aldolase in
the presence of a-glycerolphosphate dehydrogenase resulted in 0.48

moles of NADH oxidized for each mole of organic phosphate in the

69
product, indicating it reacted through the pathway: kinase
product-——+ dihydroxyacetone phosphate -—+ a-glycerol phosphate,
(v) cleavage of the product with D-tagatose 1,6-diphosphate aldolase
in the presence of triose phosphate isomerase and a-glycerolphosphate
dehydrogenase resulted in 1.01 moles of NADH oxidized for each mole
of organic phosphate in the product, indicating it reacted through
the pathway: kinase product-——+ dihydroxyacetone phosphate +
D—glyceraldehyde 3-phosphate-——+ 2 dihydroxyacetone phosphate ——+
Za-glycerol phosphate, and (vi) the product was not cleaved by
rabbit muscle aldolase, as is known not to happen with D-tagatose
1,6-diphosphate (61,94).

The product was enzymatically dephosphorylated for further
analysis. When subjected to paper chromatography, the free keto-
hexose migrated identically with tagatose, but separated from
fructose (R = 1.14), sorbose (R = 0.96), and

tagatose tagatose

p51cose (Rtagatose = 1.22). When the dephosphorylated product was

subjected to the cysteine-H2804 reaction (89), the half-time required
for color development was 6.7 min, as it was for tagatose, while
psicose took 14.5 min and fructose 45.7 min. If this reaction

were allowed to proceed for 20 hours (88), the A412 to A605 ratio

was 0.52 for sorbose, 3.48-3.61 for the other Z-ketohexoses, and

3.58 for the dephosphorylated product. All the evidence therefore

indicates that the dephosphorylated product was tagatose.

Reaction stoichiometry. The stoichiometry of the kinase

 

reaction was determined by the simultaneous assay for ADP and

D-tagatose 1,6—diphosphate formed in separate cuvettes, as shown

70
in Table 7. These data demonstrate that for each mole of ATP and
D-tagatose 6-phosphate, one mole each of ADP and D-tagatose
1,6-diphosphate were formed. This, combined with the product
identification data, substantiates that D-fructose 6-phosphate
kinase phosphorylates D-tagatose 6-phosphate with ATP, generating

D—tagatose 1,6-diphosphate and ADP.

Discussion. The data presented in this section clearly

 

establish that the product of the L-galactitol l-phosphate
dehydrogenase reaction (D-tagatose 6-phosphate) is enzymatically
phosphorylated with ATP to yield D-tagatose 1,6-diphosphate.
Furthermore, determination of the reaction stoichiometry has
revealed that for each mole of D-tagatose 1,6-diphosphate formed,
one mole of ATP is converted to ADP.

The isolated product of the enzymatic reaction was identified
as D-tagatose 1,6-diphosphate on the basis of the following criteria:
it eluted from Dowex l-X8 as expected for a ketohexose diphosphate;
the phosphate to ketohexose molar ratio was 2:1: the product was
cleaved quantitatively by K. pneumoniae D-tagatose 1,6-diphosphate
aldolase, but was not cleaved by rabbit muscle D-fructose
1,6-diphosphate aldolase; and finally, when cleaved by the appro-
priate aldolase, the two triose phosphates formed were those
expected for D-tagatose 1,6-diphosphate. Dephosphorylation of the
reaction product released a ketohexose that was proven to be tagatose.
Paper chromatography showed identical migration of the dephosphorylated
product and tagatose. In addition, when subjected to the cysteine-

H2504 reaction, the dephosphorylated product developed its chromophore

71

Table 7. Determination of the product stoichiometry of D-tagatose
6-phosphate kinase

Both cuvettes contained 9.4 nmoles of ATP and 9.4 nmoles of D—
tagatose 6-phosphate. Reaction A, which measured ADP formed, also
contained 0.05 umole of PEP, non-limiting amounts of pyruvate
kinase and lactate dehydrogenase, and D—tagatose 6-phosphate
kinase. Reaction B, which measured D-tagatose 1,6-diphosphate,
also contained the standard components of the D-tagatose 1,6—
diphosphate aldolase assay.

D-Tagatose 1,6—

 

. a ADP formed in diphosphate formed
Time reaction A in reaction B

10 1.95 2.34

20 3.90 4.39

30 5.65 6.09

40 6.92 7.31

50 8.00 8.29

60 8.68 8.87

70 9.07 9.17

80 9.36 9.26

90 9.46 9.36
100 9.46 9.36

 

a . . . .
Time 15 1n minutes.

bnmoles.

72
at the same rate as authentic D-tagatose, and after 20 hours had

the same A /

412 ratio as the tagatose.

A605

That the D—tagatose 6-phosphate kinase is the same protein as
the D—fructose 6-phosphate kinase is demonstrated by the following
data: (i) the activities have a constant ratio of specific activity
during the purification and elute coincidentally from DEAR-cellulose;
(ii) both activities have the same rate of inactivation at 64°:

(iii) the pH optima in Tris buffer are identical; (iv) the relative
rates of phosphorylation of the two ketohexose phosphates by ATP,
ITP, and GTP were essentially identical; and (v) D-tagatose
6—phosphate inhibits the phosphorylation of D-fructose 6-phosphate.
Thus, all the evidence is consistent with the two kinase activities
being the result of the dual specificity of a single protein. This
will be confirmed by the mutant analysis in Section 3. There is
also precedence in the literature for mammalian D-fructose
6—phosphate kinases being able to phosphorylate D-tagatose
6—phosphate (68).

Regulation of the enzyme activity in vivo may be due, at least
in part, to the effects of ADP. ADP lowered the Km of the enzyme
for D-tagatose 6-phosphate, while apparently not changing the Vmax'
This effect has already been reported to occur with the D-fructose
6-phosphate substrate (67). The net result would be an activation

of kinase activity at ketohexose diphosphate concentrations which

are less than saturating in the absence of ADP.

73

C. D—Tagatose 1,6-Diphosphate Aldolase (D—Tagatose 1,6-
Bisphosphate D—Glyceraldehyde 3-Phosphate-Lyase)

 

 

Purification. During preliminary purification attempts, a

 

few techniques were not successful either because of low purifica—
tion or low recovery. Precipitation by the addition of solid
ammonium sulfate to the cell extract resulted in poor recovery of
the activity. Batchwise treatment of the cell extract of DEAE-
cellulose-chromatographed fractions with hydroxyapatite in 0.005 M
sodium phosphate buffer (pH 7.5) failed to cause adsorption of the
aldolase activity. The addition of small amounts of bentonite to
the cell extract provided up to a 1.7—fold purification, but if
enough were added to remove the contaminating NADH oxidase, the
loss of significant amounts of the aldolase activity occurred.
Precipitation of the activity at pH values less than 6 resulted in
irrecoverable loss of the activity, while retaining almost all of

the D-fructose 1,6-diphosphate aldolase activity.

Preparation of cell extract. Cell extracts from

 

galactitol-grown cultures of K. pneumoniae were prepared in 0.02 M
sodium phosphate buffer (pH 7.5), containing 10% (v/v) glycerol.
All steps took place at 0-4°. A summary of the purification scheme

appears in Table 8.

DEAE-cellulose chromatography 1. A DEAE—cellulose column

 

(1.2 x 10 cm) was equilibrated with the cell-extract buffer. The
extract was applied to the column, and the column washed with 100 ml
of the same buffer. The adsorbed protein was eluted with a lSO-ml

linear gradient of 0 to 0.45 M KCl in 0.02 M sodium phosphate buffer

74

w CH ma >um>oommp

.oom um CHE mom Um>mmao mumuumfldm mo mace: H u been Ho

mE ca ma :Hmuowmn

.HE CH ma mEsHo>m

 

 

 

vm no.m o.mm mm.N mm HH mmoHsHHoOIm<mo

mm 84 m .3 m .mm 3 03-0 xmcmsmmm

NHH 065.0 m.ov m.ow we H uncasflamolmama

ooH nmoo.o m.H¢ mmo mv uocuuxm Hamo

U>Hm>oomm camuoum ma owufl>fiuom ncwmuoum mEDHo> cofluomum
mic: H309 H309

 

wmmHoch wumnmmosmHUI©.H mmoummwulo mo coHumoHMstm .m magma

75
(pH 7.5), containing 10% (v/v) glycerol. Fractions of 2 ml were
collected (Figure 15). The fractions containing the bulk of the

activity (numbers 35-66) were pooled.

Sephadex G-150 chromatography. The combined DEAE-cellulose

 

fractions were concentrated to approximately 15 ml by pressure fil-
tration in an Amicon Diaflo apparatus. This concentrate was applied
to a Sephadex G-150 column (3.2 x 81 cm) equilibrated with 0.02 M
sodium phosphate buffer (pH 7.5) containing 10% glycerol. The
column was eluted with the same buffer, and fractions of approxi-
mately 4 ml were collected (Figure 16). Fractions 50-67 were pooled

for further purification.

DEAE-cellulose chromatography II. The combined activity

 

from the Sephadex G-150 column was applied to a DEAR-cellulose
column (1.2 x 5 cm) previously equilibrated with 0.02 M sodium
phosphate buffer (pH 7.5), containing 10% (v/v) glycerol. The
column was then washed with 100 ml of the same buffer. The adsorbed
protein was then eluted with a 200—ml linear gradient of 0 to 0.4 M
KCl in 0.02 M sodium phosphate buffer (pH 7.0), containing 10% (v/v)
glycerol. Fractions of approximately 2 ml were collected (Figure
17). DEAF-cellulose II fractions 57—75, containing the majority

of the activity, were combined. The aldolase was 48—fold purified
with a 54% yield. The activities of L-galactitol l-phosphate
dehydrogenase and D-tagatose 6-phosphate kinase could not be
detected (less than 0.0001 unit/ml). The preparation was tested

for purity by electrophoresis in 7.5% polyacrylamide gels at pH 8.9.

76

(mg/ml)

Protein

ADV can ammmHOUHm mumsmmonmaploJ mmoummmulo ADV "30853
mumcmmonmflolm.a mmoummmula mo H unmmumoumsousu mmoasaamolmsma

umnEDZ coauUMHm

.cflmuowm
.wmmaocam
.mH musmam

 

ON 00 om 0: Oh ON OH
u q 1 [fill a 4
H I .

 

m.o

 

 

n
a)
C)

(Tm/Sllufl) GSPIOPIV enequoqua-g‘t esoqebeL—q

77

(mg/m1)

Protein

.HE : mm3VmEsHo> :ofluumum .cawuowm

AOV 0cm “ommaopam mumnmmozmaplmxa mmoummmuln ADV "30856 .omalo
xmomzmmm co mmmaopam mumnmmonmfiplo.a mmoummmula mo msmmnmoumEouno .oa madmam

n.0 I

3.0 I

m.0—.

 

HGQEDZ GOfluUMHh

om om ON 00 om 0: on
fl 1 q q q q

 

0.0

 

(Tm/situn) asetoptv aqequothQ—g‘t asoqebeL—q

 

 

78

.cfimuoum A0» «:3 .mmmaomuam mumcmmonmaolmqfl amoummmuln AOV

.HE m we mEsHo> coauumwm

umHOQE>m

mumnmmonmaUI©.H mmoummmula mo HH hzmmumoumsouno mmoasHHmUIm<mn

Protein (mg/ml)

H.o

m.o

m0

:0

wonEsz coauomnm

.mmmaopam
.NH cowuumwm

 

 

 

om om 8. oo on on on om
u q H . 4 q .
. ,. . .. «.6
. . . i To
. .. i 0.0
1 0.0
i o;

 

 

(Im/sqtun) eseIOpIV eqequothq-g‘I esoqebeL-a

79
When stained for protein, the gels showed one major, and at least

four minor, bands, indicating that the aldolase is not homogeneous.

Properties

 

Stability. The activity-containing fractions from the
second DEAF—cellulose treatment were kept on ice for five weeks,
with loss of approximately one-third of the activity. Identical
fractions were kept frozen at -20° for three months with little

loss of activity.

EH optimum. The pH optimum of the aldolase activity was

 

determined in two different buffer systems. The highest activity
occurred at pH 8.0 in HEPES buffer, and at pH 8.25 in glycylglycine

buffer (Figure 18).

Substrate specificity and kinetic constants. D-Tagatose
1,6-diphosphate and D-fructose 1,6-diphosphate are substrates for
the aldolase. Lineweaver-Burk plots show that the Km values for
D—tagatose 1,6—diphosphate and D-fructose 1,6-diphosphate (Figure
19) were 0.38 mM and 0.86 mM, respectively, while the Vmax is 25—
fold higher for cleavage of the former. The following sugar
phosphates were tested as substrates for the aldolase at concentra—
tions of 10 mM: D-fructose 6—phosphate, D-tagatose 6-phosphate,
L-sorbose 1-phosphate, D-fructose l-phosphate, D-glucose 6-phosphate,
D-galactose 6-phosphate, D-mannose 6-phosphate, and L-galactitol
l-phosphate. D-Fructose l—phosphate was also tested at a concentra-
tion of 50 mM. With a limit of detection of 0.0005 unit/mg protein,

none of these compounds was cleaved by the aldolase.

80

 

100“’ '

GOP

«t' U1 0\ ‘\1 CD
0 O O O O
T T T I t

% of Maximum Activity

\M
O
1

10"

 

 

l J l I I l l

a

6.0 6.5 7.0 7.5 8.0 8.5 9.0
pH

 

Figure 18. pH optimum of D-tagatose 1,6-diphosphate
aldolase. The buffers used were HEPES (O) or glycyl-

glycine (ll).

81

 

 
 

 

   

A

>.:T16
“a
8 £112~
”'2
(1)
>V
\\ 8
p:

4

  

 

1 l L
0.5 1.0 1.5

l/ -Fructose 1,6—Diphosphat§](mM-

B T ”—-

 

1)

 

8 P
>,,\
311’.

-
856
'4 \\

g 234 P
\V
H

2-

 

! l l l l

1 2 3 4 5

l/ ED-Tagatose 1,6-Diphosphata (mM-

 

 

 

1)

Lineweaver-Burk plots for D-tagatose 1,6-
diphosphate aldolase. Determination A was done with
the assay utilizing D-fructose 1,6-diphosphate. The
data in B were obtained with the standard assay util-
izing D-tagatose 1,6-diph05phate generation. In ad-
dition, both assays contained 0.5 nmoles of KCl and
0.03 umoles of CoC12. The specifi Vmax values are
0.65 units/mg protein and 16.8 units/mg protein for

A and B respectively.

Figure 19.

82

That the cleavage of D-fructose 1,6-diphosphate is due to the
D—tagatose 1,6—diphosphate aldolase can be shown in several ways.
First, l45-fold purified D-fructose 1,6-diphosphate aldolase from
K. pneumoniae, which does not utilize D-tagatose 1,6-diphosphate as
a substrate, has a Km of 0.14 mM for D-fructose 1,6-diphosphate
(unpublished observations), as compared to the value of 0.86 mM
obtained in the above study. And second, when the D-tagatose
1,6-diphosphate aldolase was heated in a water bath at 64°, there
was a coincident decay in the two activities (Figure 20). These
data are consistent with the cleavage of both substrates by
D-tagatose 1,6-diphosphate aldolase.

Unlike the D-tagatose 1,6-diphosphate aldolase from S. aureus
(64), the enzyme from K. pneumoniae does not cleave D-psicose
1,6-diphosphate or D-sorbose 1,6—diphosphate. To demonstrate this,
a solution containing 100 umoles of DL-glyceraldehyde 3-phosphate
was adjusted to pH 8.0 with NaOH. Triose phosphate isomerase and
D-tagatose 1,6-diphosphate aldolase were added to make the final
volume 5.0 ml. The solution was incubated at 25°. Samples were
withdrawn at timed intervals to monitor the progress of the reaction.
When the formation of ketohexose had halted, the solution was
adjusted to pH 8.5 with NaOH and applied to a 1.2 x 16 cm Dowex
l-X8 bicarbonate column. The column was washed with 300 m1 of water
and then 150 ml each of 0.15 M, 0.3 M, and 0.45 M KHCO3 (51). The
0.15 M wash contained monophosphates, which were discarded. The
diphosphates were eluted in the 0.3 M wash. The ketohexose—
containing fractions were combined and acidified by addition of

+
Dowex 50W-X8 (H form) until the pH was 2.0. The sample was then

83

100

90
80

7O
60

50

40

30

% of Untreated Activity

 

20 b

 

 

l L l l

 

5 10 15 20
Minutes at 640

Figure 20. Concomitant thermal inactivation of D-taga-
tose 1,6—diphosphate aldolase and D-fructose 1,6-di-
phosphate aldolase activities. Symbols: D-tagatose
1,6-diphosphate aldolase activity (0), and D-fructose
1,6-diphosphate aldolase activity (A ) .

84
lyophilized and dissolved in 5 ml of water. The pH was adjusted
to 7.0 with NaOH.

All of the ketohexose eluted from the Dowex column as would
be expected for a diphosphate, and had a phosphate to ketohexose
ratio of 1.92:1. For each mole of organic phosphate in the product,
1.04 moles of NADH were oxidized when incubated with D-tagatose
1,6-diphosphate aldolase, triose phosphate isomerase, and
a-glycerolphosphate dehydrogenase. The reaction sequence was as
follows: ketohexose diphosphate ——+ dihydroxyacetone phosphate +
D—glyceraldehyde 3—phosphate ——+ 2 dihydroxyacetone phosphate -+
2 a-glycerol phosphate. This implies that only the D-isomer of
the commercial DL-glyceraldehyde 3-phosphate was utilized in the
condensation reaction, and that dihydroxyacetone phosphate and
D-glyceraldehyde 3-phosphate are the products of the cleavage
reaction.

For further identification, the ketohexose diphosphate was
enzymatically dephosphorylated, and subjected to analysis by paper
chromatography. The amounts used were 2 umoles for the dephosphoryl-
ated product, and 100 nmoles for the 2-ketohexose standard, whereas
the threshold of detection for the visualizing reagents was less
than 10 nmoles. The carbohydrates were visualized with an orcinol
spray for ketohexose, and a duplicate chromatogram was developed
with a silver nitrate spray for total carbohydrates. The only
spots visible from the dephosphorylated product corresponded to
tagatose and fructose (R = 1.12). Spots corresponding to

tagatose

sorbose (R = 0.96) and psicose (R = 1.22) were not
tagatose tagatose

visible. Thus, neither sorbose nor psicose could constitute as much

85
as one percent of the dephosphorylated product. The absence of
sorbose, which migrates closely to tagatose, can be confirmed by
the cysteine-H2804 reaction (88). After 20 hours, the A412 to A605
ratio for sorbose was 0.45, whereas the ratio for the other three
2-ketohexoses was 3.46-3.6. Analysis of the dephosphorylated product

gave a ratio of 3.54, indicating the absence of sorbose in the

solution.

Effect of cations. Because the assay for D-tagatose

 

1,6-diphosphate aldolase contains D-fructose 6-phosphate kinase,
which requires magnesium ions, this assay could not be used to
assess the effect of divalent cations. Instead, D—fructose
1,6-diphosphate was used for these assays, thereby obviating the
use of D-fructose 6-phosphate kinase. The effect of nine divalent
cations on the activity of EDTA-treated aldolase was determined
(Table 9). Without the addition of divalent cations, the EDTA—
treated enzyme had no detectable activity. The divalent cations
that restored the activity to levels much higher than the untreated

++ ++ , , ,
enzyme were Co and Cd . The actiVity was restored to approx1-

++ ++ ++ ++
mately the untreated levels by Mn and Zn , whereas Fe and Ca
only partially restore the activity. The activity was not restored
++ ++ ++
by Mg , Ni , or Cu .
The addition of monovalent cations to D-tagatose 1,6—diphosphate
aldolase which has not been treated with EDTA (Table 10) shows that
. . . + +

the cations which stimulate the most are K and NH4 . The effect of

++
potassium on the kinetic constants of non-EDTA-treated, but Co

supplemented, aldolase is shown by comparing Lineweaver-Burk plots

86

Table 9. Effect of divalent cations on EDTA-treated D-tagatose
1,6-diphosphate aldolase

The standard assay using D-fructose 1,6-diphosphate was used to
obtain these results.

 

Units/mg protein

 

 

Metal useda 2 x 10:5 M 10‘3 M
None <0.002 <0.002
MgCl2 <0.002 <0.002
NiCl2 <0.002 <0.002
CuSO4 <0.002 <0.002
FeSO4 0.025 <0.002
ZnCl2 0.077 <0.002
CaCl2 0.009 0.077
MnCl2 0.090 0.201
CoCl2 0.201 0.188
CdSO4 0.225 0.233

 

a
Untreated enzyme

0.074 units/mg protein.

87

Table 10. Stimulation of D-tagatose 1,6-diphosphate aldolase by
monovalent cations

5 umoles of monovalent cation were added to the standard assay
utilizing D-fructose 1,6-diphosphate as substrate.

 

 

Addition Relative velocity
None 100
KCl 1440
NH4C1 1540
NaCl 620
RbCl 179
LiCl 116
CsCl 100

 

in the presence (Figure 19B) and absence (Figure 21A) of KCl. The
presence of KCl lowers the Km for D-fructose 1,6-diphosphate from
1.27 mM to 0.86 mM, and stimulates the Vmax approximately 2.5-fold.
Varying the potassium concentration showed that it has a Ka of
approximately 6 mM (Figure 21B).

Thus, it can be seen that both monovalent and divalent cations

have significant effects on the enzyme activity.

Effect of NaBH4. Inhibition of the aldolase activity by

 

NaBH in the presence and absence of substrate (0.5 mM D-tagatose

4i
1,6-diphosphate or 25 mM D-fructose 1,6-diphosphate), was assessed
using 10 mM or 100 mM of the reducing agent. Significant loss of

activity (greater than 2%) 15 min after addition of the ketohexose

88

 

 

 

 

A r4016 -
>. m
u 4.)
-H «4 _
8 §§O.12
a: E
950.08
E.
>V
04 -
I l 1 l l

 

0.25 0.75 1.25
l/ D-Fructose 1,6-Diphosphate (mM-l)

 

w
4:-
O
O
1—

\JJ
0
O

N
O
O

 

1/velocity
(ml/Units)

   

I"
O
O

 
 

 

 

l l l l l

0.1 0.3 0.5

1/ KCl (mM-l)

 

Figure 21. Effect of potassium ion on D-tagatose 1,6-
diphOSphate aldolase. Both A and B were obtained using
the aldolase assay utilizing D-fructose 1,6-diphosphate
as substrate with 0.03 umole of CoC12. The D-fructose
1,6-diphosphate concentration in B is 4 mM. The specific

Vmax for A is 0.24 units/mg protein.

89
diphosphate was not found. A control incubation containing rabbit

muscle aldolase was 96% inhibited by 10 mM NaBH in the presence of

4
1.5 mM D-fructose 1,6-diphosphate, after 6 min. These data indicate
that D-tagatose 1,6-diphosphate aldolase does not form a Schiff's

base intermediate as part of the reaction mechanism (97,98). The

aldolase can therefore be classified as class II type (97,98).

Molecular weight. The elution of the aldolase from a

 

1.45 x 88 cm Sephadex G-150 column, with standards of known molecular
weight, was used to determine the molecular weight of the aldolase.
The standards and their molecular weights are as follows: rabbit
muscle pyruvate kinase, 237,000 (99); beef heart lactic dehydrogenase,
140,000 (90,95); yeast glucose 6-phosphate dehydrogenase, 128,000
(90); and E. coli alkaline phosphatase, 86,000 (91). Plotting the
elution versus the log of the molecular weight of the aldolase
results in a linear relationship (Figure 22). Assuming a spherical
shape for the aldolase, the molecular weight can be estimated as

157,000.

Discussion. As demonstrated by the preceding data, an enzyme

 

can be purified from cell extracts of galactitol—grown cells that
is capable of catalyzing the proposed cleavage of D-tagatose
1,6-diphosphate to dihydroxyacetone phosphate and D-glyceraldehyde
3-phosphate. The aldolase was found to be very specific, cleaving
only D-tagatose 1,6-diphosphate and, to a lesser extent, D-fructose
1,6-diphosphate. Evidence was presented showing that neither
D-psicose 1,6-diphosphate, D-sorbose 1,6-diphosphate, D-fructose

l-phosphate, nor any of a variety of sugar phosphates was cleaved

90

 

300,000 P

 
  

D-Tagatose 1,6-Diph05phate

200,000 ' Aldolase

 

150,000 P

Molecular Weight

100,000 .
90,000 4
80,000 -
70,000 -

 

 

 

94 98 102 106 110 114 118
Fraction Number

Figure 22. Molecular weight determination of D-tagatose
1,6-dipho5phate aldolase. The protein standards used were
pyruvate kinase (PK), D-glucose 6-phosphate dehydrogenase
(GPDH), lactate dehydrogenase (LDH), and alkaline phos-
phatase (AP).

91
by the aldolase. The relatively low Vmax for the cleavage of
D-fructose 1,6—diphosphate probably implies that it could not play
a significant role in catabolism of compounds that are converted to
D-fructose 1,6-diphosphate.

The effect of cations on the aldolase was shown to be con-
siderable. Removal of divalent cations with EDTA resulted in loss
of the activity, which could be restored by a number of cations with
varying degrees of success, Co++ and Cd++ being the best. In the
presence of an activating divalent cation, monovalent cations pro-
vided further activation, the effect of K+ being to lower the Km
and raise the Vmax

The lack of inhibition of activity in the presence of NaBH4 and
ketohexose diphosphate substrates allows designating the aldolase
as a class II type. This class of aldolase has been assumed to be
typical for bacterial species. Typically they require divalent
cations, and are activated by monovalent cations. One character-
istic of the aldolase that is more characteristic of class I
(mammalian) aldolases is the molecular weight of 157,000. Most
class I aldolases have molecular weights of 150,000-160,000, whereas
class II aldolases generally have molecular weights of 70,000-
80,000 (97). That this aldolase is of bacterial origin cannot
alone be used as a criterion for classification, as a few bacterial
class I aldolases have now been identified (100—105), including the
D-tagatose 1,6-diphosphate aldolase from S. aureus (64). Except for
the discrepancy in molecular weight, the data fit the criteria for

a class II aldolase.

92

It is interesting to note the differences in the D—tagatose
1,6-diphosphate aldolase from K. pneumoniae and S. aureus, the only
two examples of this enzyme ever studied. The aldolase from S.
aureus (64) is a 40,000 molecular weight protein, utilizes a Schiff's
base intermediate, is not inhibited by EDTA or activated by mono-
valent cations, and can cleave all four D-2-ketohexose diphosphates
(D-fructose 1,6-diphosphate at 47% of the rate for D-tagatose
1,6-diphosphate). The K. pneumoniae enzyme, on the other hand, is
a 157,000 molecular weight protein, utilizing a divalent cation in
its reaction mechanism, rather than a Schiff's base intermediate,
is activated by monovalent cations, and is specific for D-tagatose
1,6-diphosphate and D-fructose 1,6-diphosphate (cleaving the latter
at 4% the rate of the former). Except for catalyzing the same
physiological reaction, the two enzymes seem to have little in
common. Perhaps as more organisms are studied with respect to
this enzyme, an evolutionary or genetic basis for these differences

will become apparent.

SECTION 3

GENETIC EVIDENCE FOR THE PHYSIOLOGICAL SIGNIFICANCE OF THE

PROPOSED PATHWAY FOR GALACTITOL CATABOLISM

93

INTRODUCTION

A pathway for the catabolism of galactitol through L-galactitol
l-phosphate, a known metabolite in Klebsiella pneumoniae (59), has
been established. The data of the first two sections of this dis-
sertation indicate potential participation of a dehydrogenase,
kinase, and aldolase, involved in oxidation of L-galactitol
l-phosphate to D-tagatose 6-phosphate, phosphorylation to D-tagatose
1,6-diphosphate, and cleavage to dihydroxyacetone phosphate and
D-glyceraldehyde 3-phosphate. These three enzymes have been demon-
strated in cell extracts of galactitol-grown cells. In addition,
these enzymes have been partially purified, and their Km values and
substrate specificities found to be consistent with their role in
this pathway.

The studies of this section show the inducibility of the
dehydrogenase, kinase, and aldolase, as well as genetic evidence for
their physiological role in galactitol catabolism. It is demonstrated
that although the kinase activity is constitutive, the dehydrogenase
and aldolase activity are specicifically induced by growth on the
cells on galactitol. Analysis of mutants missing either the dehy-
drogenase, kinase, or aldolase shows that each of these enzymes is
essential to the ability of K. pneumoniae to utilize galactitol as

a carbon source for growth.

94

MATERIALS AND METHODS

All materials and methods not presented here were described

in the Materials and Methods of Sections 1 and 2.

Organisms. The wild type organism was Klebsiella pneumoniae
PRL-R3 (formerly designated Aerobacter aerogenes PRL—R3). All the
mutants described were derived from a deletion mutant of PRL-R3

that is auxotrophic with respect to growth on uracil (U-).

Media. Routine growth of cells was done in nutrient broth.
For mineral agar plates, agar and mineral medium were made up and
sterilized separately at such concentrations that the final mixture
was 1.5% (w/v) agar. Growth of cells on specific sugars was at a
concentration of 0.5% (w/v). Growth or induction of uracil auxo-

trophs was in media supplemented with 0.005% (w/v) uracil.

Enzyme Induction. Induction of the pathway enzymes in different

 

strains was initiated by adding 7 m1 of nutrient broth containing 1%
galactitol to a 7 ml nutrient broth culture of cells. The tubes
were then incubated on a reciprocal shaker at 30°. Hourly sampling
of wild—type cells under these conditions, and subsequent assays of
the extracts for the dehydrogenase and aldolase, established that
these enzymes were maximally induced after six hours. Following

this time, the specific activity of the activities decreased,

95

96
dropping to about half of the maximal values after 10 hours. All

other strains were also assayed after six hours of incubation.

Monitoring of Cell Growth. Cell growth was monitored with a

 

Coleman Junior Spectrophotometer, model 6A, at 600 nm. These cells
were grown in 18 x 150 mm tubes containing 7 ml of mineral medium
supplemented with 0.5% (w/v) of carbohydrate. The tubes were incu-
bated at 30° on a reciprocal shaker. A plot of absorbance versus

growth was prepared using actual A values from undiluted samples,

600
and corrected absorbance values from diluted samples. This plot was

used to correct for light scattering at higher absorbance values.

Enzymatic Assays. All assays were as described in Section 2,

 

except that the D-tagatose 1,6-diphosphate aldolase assay addi-

tionally contained 5 umoles of KCl and 0.15 umoles CoCl for optimal

2

activity.

Mutant A9—1 (D-Fructose 6-Phosphate Kinase Negative). Mutant

 

A9-1 was the strain isolated as reported previously (106) and found

to be missing D-fructose 6-phosphate kinase.

Isolation of Mutants Missing L-Galactitol l-Phosphate Dehy-

 

drogenase (VP-14) or D-Tagatose 1,6-Diphosphate Aldolase (VP—8).

 

To isolate mutants selectively impaired for growth on galactitol,

an overnight culture of K. pneumoniae PRL-R3 (U-) in nutrient broth
(7 ml) was harvested by centrifugation in a sterile centrifuge tube,
and suspended in 7 m1 of mineral medium containing 0.2 M ethyl
methanesulfonate. After incubation for two hours on a shaker at

30°, the cells were harvested and washed three times with 7 ml of

97
mineral medium. The cells were grown in nutrient broth until there
was a lO-fold increase in cell number. The cells were suspended in
mineral medium containing 0.5% (w/v) galactitol and 2000 units/ml
of Penicillin G. After four hours, the cells were harvested,
washed in mineral medium, and grown for 10 generations in mineral
medium plus 0.5% (w/v) D-fructose. The penicillin-treated bacteria
were placed on uracil-supplemented mineral agar containing 0.5%
galactitol plus 0.005% D—glucose. After 48 hours at 30°, the
smallest colonies were selected and grown in nutrient broth. The
cells were screened for growth in mineral agar plates containing
0.5% (w/v) of galactitol, mannitol, D-glucose, or D-fructose.
Mutants VP—8 and VP-l4 were selected as strains which grew well on
D-mannitol, D-glucose, and D-fructose, but failed to grow on galac-

titol. Both mutants were shown to be uracil auxotrophs.

Isolation of Spontaneous Revertants (VP-8R, VP-14R, and A9—1R).

 

Revertants of the above mutants were isolated by exerting selective
pressure favoring only growth on galactitol. The reason for this
was twofold: first, to establish that the above mutations were
caused by single, reversible point mutations, and second, to show
that selective pressure from galactitol alone was enough to cause
the reversions, establishing each gene's physiological importance
in the pathway. Overnight cultures of VP-8, VP-l4 and A9-1 in
mineral medium plus 0.25% of D-fructose were plated at high density
on mineral agar containing 0.5% galactitol (w/v). After 48 hours,
revertants which appeared were isolated. It was confirmed that the

cells were still auxotrophic for uracil.

98

Source of Materials. The following materials were obtained

 

from the indicated sources: uracil, penicillin G, D-mannose,

D-fructose, D-galactose, maltose, and lactose from Sigma Chemical
Co.: ethyl methanesulfonate from Eastman Organic Chemicals; agar
from Difco Laboratories; D-glucose and sucrose from Mallinckrodt

Chemical Works; and L-arabinose from Pfanstiehl Laboratories.

RESULTS

Analysis of Enzyme Defects in the Mutants. The parental strain
(PRL-R3 U-), the mutants (A9-l, VP-8, and VP-14), and revertants
(A9-1R, VP-8R, and VP-14R) were incubated in nutrient broth contain-
ing 0.5% galactitol for 6 hours. Cell extracts were prepared and
assayed for the galactitol pathway enzymes (Table 11). Strain A9-1
was confirmed to be missing D-fructose 6-phosphate kinase and was
also shown to be deficient in D-tagatose 6-phosphate kinase, whereas
A9-lR regained both these activities. Mutant VP-8 was missing
D-tagatose 1,6-diphosphate aldolase, whereas the revertant regained
this activity. The cell extract of VP-l4 lacks L-galactitol
l-phosphate dehydrogenase, but the revertant did exhibit this
activity. Thus, a mutant deficient for each one of the pathway

enzymes (dehydrogenase, kinase, and aldolase) has been isolated.

Growth of Mutants. The growth of the parental strain (PRL-R3

 

U-) on eight different sugars, including galactitol, is shown in
Figure 23. The cells were observed to exhibit a lag before growth

on galactitol took place. Mutant A9-l was seen to be affected for
growth on a number of sugars, with no growth detected on galactitol
(Figure 24). Growth of mutants VP-8 and VP-l4-—as well as revertants
A9—1R, VP—8R, and VP-l4R--on galactitol is seen in Figure 25. VP-8

and VP—l4, like A9-l, did not grow on galactitol, but were not

99

100

.oom um CHE Hem meSmcoo mumuumndm mHOE: H n uHcs Hm

 

 

NOH.0 00H.0 mmH.0 mmo.0 mvHum>
N0m.o m0H.O h0H.o 0ho.o mmlm>
m00.0 0n0.0 an.0 n00.0 mHum<
mm0.0 000.0 00H.0 H00.0v 0H1i>
H00.0v 000.0 0MH.0 0H0.0 mum>
00H.0 m00.0v m00.0v Hm0.0 H|m<
nHm.0 m00.0 00H.0 HOH.0 ID mmIHMm
mmmHOUHm mmmcHx 0mmCHx 0mmcmmonvmsmo chwum
mHMSQmOQQHp mumcmmonm mumnmmonm mumzmmocm

I0.H mmoummmuuo I0 mmoummmulo n0 wmouUDHMIo IH HouHuomHmmIH

chuowm mE\wuHco

 

m

 

mchwum
accuse 00m Hmucmuwm 0H 0003mmozmlH HOUHuUMHmmIH mo EmHH0bdumo wsu How mmsxncm mo coHuosch .HH wHQme

101

 

 
  

 

 

L
O 8 a. D-Glucose .1
_ D-Glucitol
D-Mannose ‘ \\\‘
0.6 ' D-Mannitol W \@W\\“\\\\\ -'
O 5 _ D-Fructose \\\‘
' D-Galactose

0.1; .- L-Arabinosg’

0.3 '
90.2 *f -
G galactitol
0 (if
c>
\o
4.:
m
(no.1
8
'30.08
H
3 6
_Q0.0
“0.05

0.04

0.03

0.02

 

 

(-

l l l 1 I ,
2 4 6 8 10 12 14
Time (hours)

Figure 23. Growth characteristics of K. pneumoniae PRL-R}

(U ). All carbohydrates tested were at 0.5% (w/v) concen-

tration. The inoculum was 0.2 m1 of a nutrient broth-grown
culture.

102

 

     
 
    

 

 

 

 

0.8 P
0.6 )- . . ' ' ' ' '
0 5 _ D-Fructosew ’D-GalactoseM I
' L-Arabinose II’I“’ ‘
0.4 P -_ ‘
‘I-Glucos:
0.3 '
E0.2 -
c ’ D-Mannose
C)
8 , o '
p O
m
(”0.1 '- . D-Mannitol
o F: .
50.08. .
.Q 0
H P
$30.06*
'g O. 05% «A D-GlUCitDl
004~ :, A
C .l
0.03.}!- o o o o .
'/ ' 0
; ‘LLH-Galactitol
0.02 -
l n A l l 4
2 4 6 8 10 12 14

Time (hours)

Figure 24. Growth characteristics of strain A9-1. Carbohy-
drate concentrations were 0.5% (w/v) and the inoculum was 0.2
m1 of a nutrient broth-grown culture of A9-l.

103

 

 

 

 

 

0.8 -
0.6 " VP-14R
0. P
5 A9-1R
0.4 r-
0.) ' VP-8R
E0.2 b
C.
O
O
\O
.u
(0
$0.1 -
U I-
{€0.08-
33 )-
cn0.06 5
.0
“10.05 - x
.-
0.04 P
O 03 P
0.02 P

 

v

2 4 6 8 10 12 14
Time (hours)

Figure 25. Growth of strains VP-8, VP-l4, A9-1R, VP-8R, and
VP-14R on galactitol. An inoculum of 0.2 m1 of a nutrient
broth-grown culture was used. Growth for all strains on
0.5% (w/v) D-glucose, D-glucitol, D-mannose, D-mannitol,
D-fructose, D-galactose, and L-arabinose was the same as for
PRL-R} (U') in Figure 23.

104
affected for growth on any other carbohydrate tested. All three
revertants regained the ability to grow on galactitol, and A9—lR
no longer exhibited slow growth on D-glucitol, D-mannitol, D-glucose,

or other sugars tested.

Specificity of Induction. The parent strain (PRL-R3 U-) was

 

grown on a variety of carbohydrates to show the induction pattern

of the dehydrogenase, kinase, and aldolase (Table 12). These data
demonstrate that the kinase activity is constitutive as previously
reported (106), being present in extracts of cells grown on nutrient
broth and all the carbohydrates tested. The dehydrogenase and
aldolase, on the other hand, are specifically induced by growth on
galactitol. Growth on any of the other 12 carbon sources failed to

induce the activities.

105

Table 12. Induction of the galactitol pathway enzymes in K.

pneumoniae grown on a variety of carbohydrates

 

Units/mg proteina

 

 

Carbon source Dehydrogenaseyr Kinase Aldolased
Nutrient broth <0.002 0.112 <0.005
D-glucose <0.002 0.167 <0.005
D-mannose <0.002 0.103 <0.005
D-mannitol <0.002 0.216 <0.005
D—glucitol <0.002 0.226 <0.005
D-fructose <0.002 0.209 <0.005
D-galactose <0.002 0.150 <0.005
L-arabinose <0.002 0.133 <0.005
Maltose <0.002 0.137 <0.005
Lactose <0.002 0.119 <0.005
Sucrose <0.002 0.156 <0.005
Glycerol <0.002 0.190 <0.005
Galactitol 0.299 0.128 0.239

 

a .
1 unit

1 umole of substrate used per min at 30°.

bL—Galactitol l-phosphate dehydrogenase.

CD-Fructose 6-phosphate (D-tagatose 6-phosphate) kinase.
Assayed with D—fructose 6-phosphate.

dD—Tagatose 1,6-diphosphate aldolase.

DISCUSSION

The data of this section of the dissertation revealed that
mutant strains of K. pneumoniae missing L-galactitol l-phosphate
dehydrogenase, D—tagatose 6-phosphate kinase, or D—tagatose
1,6-diphosphate aldolase fail to grow on galactitol. Revertants of
these strains regained the ability to grow on galactitol, as well
as the enzyme that was formerly missing. Also, the aldolase and
dehydrogenase activities were Specifically induced by growth on
galactitol, while the kinase was a constitutive activity.

Failure of mutants missing any of the enzymes of the pathway
to grow on galactitol is evidence that this is the sole pathway of
physiological importance for galactitol catabolism in this organism.
The ability to obtain revertants of the mutants by plating on
galactitol-containing medium suggests that they were all single
point-mutations, and affirms the physiological importance of the
gene products. Selective induction of the dehydrogenase and aldolase
confirms the high substrate specificities observed in Section 2,
indicating that these enzymes are unique to this pathway. The growth
of mutant VP—14, missing L-galactitol 1-phosphate dehydrogenase, on
D-glucitol substantiates that this activity is not merely a manifes-
tation of D-glucitol 6—phosphate dehydrogenase (47). These data
also confirm that D-tagatose 6-phosphate kinase and D—fructose
6-phosphate kinase activities are manifestations of the same protein.

106

107
Mutant A9-l was selected to be missing D-fructose 6-phosphate kinase
without respect to its growth characteristics on galactitol, and was
found to be also deficient for D-tagatose 6-phosphate kinase. This
mutant is unable to grow on galactitol, or a number of other sugars.
When a revertant is obtained from galactitol plates, both kinase
activities are restored, and growth on all sugars is comparable to
the parent strain. Thus, data presented here and in Section 2
(co-purification, substrate competition, coincident thermal inacti-
vation, etc.) establish that the constitutive D-fructose 6-phosphate
kinase also catalyzes the phosphorylation of D-tagatose 6-phosphate
to D-tagatose 1,6-diphosphate.

This section of the dissertation has, therefore, established
the following: (i) D-tagatose 6-phosphate kinase is a constitutive
activity, (ii) L-galactitol 1-phosphate dehydrogenase and D-tagatose
1,6—diphosphate aldolase are specifically induced by growth on
galactitol, (iii) a single gene product is responsible for the
D-fructose 6-phosphate kinase and D-tagatose 6-phosphate kinase
activities, and (iv) these reactions constitute the sole pathway
of physiological importance for galactitol catabolism in this

organism.

SUMMARY

A review of the literature has shown that although many species
utilize galactitol, a pathway for its metabolism has never been
established. The purpose of this dissertation was to determine the
catabolic reaction sequence that allows Klebsiella pneumoniae to
make use of galactitol as a carbon source for growth. It was pre-
viously reported that in K. pneumoniae galactitol is phosphorylated
by a PEP:galactitol phosphotransferase, yielding L-galactitol
l-phosphate. The data presented in this dissertation demonstrate
that in this organism, L-galactitol l-phosphate is further catabolized
through a previously unreported pathway.

The evidence on which I have based these conclusions is as
follows: (i) the demonstration and partial purification of
L-galactitol l-phosphate dehydrogenase, D-tagatose 6-phosphate
(D-fructose 6—phosphate) kinase, and D-tagatose 1,6-diphosphate
aldolase; (ii) the chromatographic, enzymatic, and chemical identi—
fication of the reaction products of these reactions; (iii) the
galactitol-specific induction of the dehydrogenase and aldolase;
and (iv) the selection and isolation of mutant strains unable to
utilize galactitol and which are also missing one of these three
enzymes. That this is the sole physiological pathway of galactitol

catabolism is demonstrated by the failure of these mutants to grow

108

109
on galactitol, and the lack of a dehydrogenase, kinase, or epimerase
in wild-type cells capable of modifying the free hexitol.
Thus, in K. pneumoniae, galactitol is catabolized as follows:
galactitol ——+ L-galactitol l-phosphate -—+ D-tagatose 6—phosphate
-—+-D-tagatose 1,6-diphosphate -+ dihydroxyacetone phosphate +

D—glyceraldehyde 3-phosphate.

LI ST OF REFERENCES

10.

ll.

12.

l3.

14.

15.

LIST OF REFERENCES

Carr, C. J., and Krantz, J. C. Jr., Adv. Carb. Chem., 1,
175 (1945).

Lohmar, R. L., and Goepp, R. M. Jr., Adv. Carb. Chem., 4,
211 (1949).

Lohmar, R. L., in W. Pigman (Editor), The Carbohydrates,
Academic Press, New York, 1957, p. 241.

Furlong, J. R., and Campbell, L. E., Proc. Chem. Soc.
(London), 29, 128 (1913).

Bliss, C. A., Hamon, N. W., and Lukaszewski, T. P., Phyto-
chemistry, 11, 1695 (1972).

Bliss, C. A., Hamon, N. W., and Lukaszewski, T. P., Phyto-
chemistry, 11, 1705 (1972).

Fink, H., and Just, F., Biochem. Z., 296, 306 (1938).

Quan-Ma, R., and Wells, W. W., Biochem. Biophys. Res. Commun.,
20, 486 (1965).

Wells, H. J., and Segal, S., Fed. Europ. Biol. Soc. Letters,
5, 121 (1969).

Gitzelman, R., Curtis, H. C., and Muller, M., Biochem.
Biophys. Res. Commun., 22, 437 (1966).

Wells, W. W., Pittman, T. A., and Egan, T. J., J. Biol. Chem.,
239, 3192 (1964).

Carr, C. J., and Krantz, J. C., J. Biol. Chem., 107, 371 (1934).
Edson, N. L., Biochem. J., 30, 1862 (1936).

Smith, W. W., Finkelstein, N., and Smith, H. W., J. Biol.
Chem., 135, 231 (1940).

Weinstein, A. N., and Segal, 5., Biochim. Biophys. Acta, 156,
9 (1968).

110

16.

17.

18.

19.

20.

21.

22.

23.

24.

25.

26.

27.

28.

29.

30.

31.

32.

33.

34.

35.

36.

37.

111
Blakely, R. L., Biochem. J., 49, 257 (1951).
McCorkindale, J., and Edson, N. L., Biochem. J., 57, 518 (1954).
Smith, M. G., Biochem. J., 83, 135 (1962).

Hayman, S., and Kinoshita, J. H., J. Biol. Chem., 240, 877
(1965).

van Heyningen, R., Nature, 184, 194 (1959).
Moonsamny, G., and Stewart, M., J. Neurochem., 14, 1187 (1967).

Mano, Y., Suzuki, Y., Yamata, K., and Schimazono, N., J.
Biochem., 49, 618 (1961).

Clements, R., Weaver, J., and Winegrad, A., Biochem. Biophys.
Res. Commun., 37, 347 (1969).

Hayman, S., Lou, M., Merola, L., and Kinoshita, J. H., Biochim.
Biophys. Acta, 128, 474 (1966).

Eitel, E. H., Ind. Eng. Chem., 12, 1202 (1920).

Dozois, K. P., Carr, C. J., and Krantz, J. C., J. Bacteriol.,
36, 599 (1938).

Dozois, K. P., Hatchel, F., Carr, C. J., and Krantz, J. C.,
J. Bacteriol., 30, 189 (1935).

Richmond, M. H., and Maaloe, O., J. Gen. Microbiol., 27, 285
(1962).

Bourgain, L., Ann. Biol. Clin. (Paris), 19, 463 (1961) cf.
Chem. Abstracts, 56, 15943h.

Lulla, B. S., J. Indian Chem. Soc., 25, 355 (1948) cf.
Chem. Abstracts, 43, 2278e.

Barnett, J. A., J. Gen. Microbiol., 52, 131 (1968).
Moore, H., and Chupp, C., Mycologia, 44, 523 (1952).
Sebek, O. K., and Randles, C. I., J. Bacteriol., 63, 693 (1952).
Arcus, A. C., and Edson, N. L., Biochem. J., 64, 385 (1956).

Kersters, K., Wood, W. A., and DeLey, J., J. Biol. Chem., 240,
965 (1965).

Kersters, K., and DeLey, J., Methods Enzymol., 9, 1970 (1966).

Lin, E. C. C., J. Biol. Chem., 236, 31 (1961).

3H.

39.

40.

41.

42.

43.

44.

45.

46.

47.

48.

49.

50.

51.

52.

53.

54.

55.

56.

57.

58.

112

Chakravorty, M., Veiga, L. A., Bacila, M., and Horecker, B. L.,
'J. Biol. Chem., 237, 1014 (1962).

Chakravorty, M., and Horecker, B. L., Methods Enzymol., 9,
163 (1966).

Shaw, D. R. D., Biochem. J., 64, 394 (1956).

Ozek, 0., Bull. Faculte Med. Istanbul, 3, 1675 (1940) cf.
Chem. Abstracts, 40, 72835.

Weimberg, R., Biochem. Biophys. Res. Commun., 8, 442 (1962).

Marmur, J., and Hotchkiss, R. D., J. Biol. Chem., 214, 383
(1955).

Baddiley, J., Buchanan, J.-G., Carss, B., Mathias, A. P.,
and Sanderson, A. R., Biochem. J., 64, 599 (1956).

Liss, M., and Kaplan, N. O., J. Biol. Chem., 2l8, 849 (1956).
Liss, M., and Kaplan, N. 0., Bacterial. Proc., 147 (1960).

Liss, M., Horowitz, S. B., and Kaplan, N. O., J. Biol. Chem.,
237, 1342 (1962).

Wolff, J. B., and Kaplan, N. O., J. Bacteriol., 71, 557 (1956).

Shockley, T. C., and Pride, H. S., J. Bacteriol., 77, 695
(1959).

Kelker, N. E., and Anderson, R. L., J. Bacteriol., 105, 160
(1972).

Kelker, N. E., Simkins, R. A., and Anderson, R. L., J. Biol.
Chem., 247, 1479 (1972).

Lengeler, J., J. Bacteriol., 124, 26 (1975).

Kundig, W., Ghosh, S., and Roseman, 8., Proc. Nat. Acad. Sci.
(USA), 52, 1067 (1964).

Kundig, W., and Roseman, S., J. Biol. Chem., 246, 1407 (1971).
Roseman, S., J. Gen. Physiol., 54, 1383 (1969).

Kaback, H. R., Annu. Rev. Biochem., 39, 731 (1970).

Kaback, H. R., Biochim. Biophys. Acta, 265, 367 (1972).

Tanaka, S., and Lin, B. C. C., Proc. Nat. Acad. Sci. (USA),
57, 913 (1963).

59.

60.

61.

62.

63.

64.

65.

66.

67.

68.

69.

70.

71.

72.

73.

74.

75.

76.

77.

113

Shimamoto, G. T., M. S. Thesis, Michigan State University
(1975).

Lengeler, J., J. Bacteriol., 124, 39 (1975).

Bissett, D. L., and Anderson, R. L., Biochem. Biophys. Res.
Commun., 52, 641 (1973).

Bissett, D. L., and Anderso., R. L., J. Bacteriol., 117, 318
(1974).

Lowry, O. H., Rosebrough, N. J., Farr, A. L., and Randall,
R. J., J. Biol. Chem., 193, 265 (1951).

Bissett, D. L., Ph.D. Dissertation, Michigan State University
(1975).

Peterson, E. A., and Sober, H. A., Methods Enzymol., 5, 3
(1962).

Winkler, H. H., Biochim. Biophys. Acta, 117, 231 (1966).

Sapico, V., and Anderson, R. L., J. Biol. Chem., 244, 6280
(1969).

Tung, T.—C., Ling, K.—H., Byre, W. L., and Lardy, H. A.,
Biochim. Biophys. Acta, 14, 488 (1954).

Garen, A., and Levinthal, C., Biochim. Biophys. Acta, 38,
470 (1960).

Noltmann, E. A., Gubler, C. J., and Kuby, S. A., J. Biol.
Chem., 236, 1225 (1961).

Reeves, W. J., Jr., and Fimognari, G. M., Methods Enzymol., 9,
288 (1966).

Bucher, T., and Pfleiderer, G., Methods Enzymol., l, 435
(1955).

Sapico, V., and Anderson, R. L., J. Biol. Chem., 245, 3252
(1970).

Roe, J. H., Epstein, J. H., and Goldstein, N. P., J. Biol.
Chem., 178, 839 (1949).

Ashwell, G., Methods Enzymol., 3, 73 (1957).
Fiske, C. H., and SubbaRow, Y., J. Biol. Chem., 66, 375 (1925).

Clark, J. M., Experimental Biochemistry, W. H. Freeman and
Co., San Francisco, 1964, p. 32.

78.

79.

80.

81.

82.

83.

84.

85.

86.

87.

88.

89.

90.

91.

92.

93.

94.

95.

96.

114

Green, A. A., and Hughes, W. L., Methods Enzymol., l, 67
(1955).

Joshi, M. D., and Jagannathan, V., Methods Enzymol., 9, 371
(1966).

Tombs, M. P., Souter, F., and MacLagen, N. F., Biochem. J.,
73, 167 (1959).

Cuatrecasas, P., and Anfinsen, C. B., Methods Enzymol., 22,
345 (1971).

Lamed, R., Levin, Y., and Wilchek, M., Biochim. Biophys. Acta,
304, 231 (1973).

Gabriel, 0., Methods Enzymol., 22, 565 (1971).
Gabriel, 0., Methods Enzymol., 22, 578 (1971).

Hanson, T. E., and Anderson, R. L., J. Biol. Chem., 24l, 1644
(1966).

Benvenue, A., and Williams, K. T., Arch. Biochem. Biophys.,
34, 225 (1951).

Smith, I., Chromatographic and Electrophoretic Techniques,
Vol. 1, Interscience Publishers, Inc., New York, 1963,

p. 246.

Dische, Z., and Devi, A., Biochim. Biophys. Acta, 39, 140
(1960).

Bissett, D. L., Hanson, T. E., and Anderson, R. L., Microchem.
J., 19, 71 (1974).

Andrews, P., Biochem. J., 96, 595 (1965).

Lazdunski, C., and Lazdunski, M., Eur. J. Biochem., 7, 294
(1969).

Andrews, P., Biochem. J., 91, 222 (1964).

Monod, J., Changeux, J.-P., and Jacob, F., J. Mol. Biol., 6,
306 (1963).

Thomas, T. D., Biochem. Biophys. Res. Commun., 63, 1035 (1975).
Castellino, F. J., and Barker, R., Biochemistry, 7, 2207 (1968).

Khym, J. K., and Cohn, W. E., J. Amer. Chem. Soc., 75, 1153
(1953).

97.

98.

99.

100.

101.

102.

103.

104.

105.

106.

115
Rutter, W., Federation Proceedings, 23, 1248 (1964).

Horecker, B. L., Tsolas, O., and Lai, C. Y., The Enzymes, 7,
213 (1972).

Cottan, G. L., Hollenberg, P. F., and Coon, M. J., J. Biol.
Chem., 244, 1481 (1969).

Bai, N. J., Pai, M. R., Murphy, P. S., and Venkitasubramanian,
T. A., Arch. Biochem. Biophys., 168, 230 (1975).

Bai, N. J., Pai, M. R., Murphy, P. S., and Venkitasubramanian,
T. A., Arch. Biochem. Biophys., 168, 235 (1975).

Kaklij, G. S., and Nadkarni, G. B., Arch. Biochem. Biophys.,
160, 47 (1974).

Kaklij, G. S., and Nadkarni, G. B., Arch. Biochem. Biophys.,
160, 52 (1974).

Lebherz, H. G., and Rutter, W. J., J. Biol. Chem., 248, 1650
(1973).

Lebherz, H. G., Bradshaw, R. A., and Rutter, W. J., J. Biol.
Chem., 248, 1660 (1973).

Sapico, V., Hanson, T. E., Walter, R. W., and Anderson, R. L.,
J. Bacteriol., 96, 51 (1968).

II(llHWIIIHIIHIIHHIIIIHWIIII(”INN“)!IMIIIID