ABSTRACT

METABOLISM OF D-FUCOSE AND
L-ARABINOSE IN A PSEUDOMONAD

by A. Stephen Dahms Jr.

The first known degradative pathway by which an
organism utilizes D-fucose (6-deoxy-D-galactose) has been
elucidated. Two unique, pyridine nucleotide-linked dehyd-
rogenases have been found which are distinguishable, in
part, by their ability to oxidize either the furanose or
the pyranose hemiacetal of D-fuoose to the y- or the 6-
lactone reSpectively. The NAB—linked D-aldohexose dehyd-
rogenase oxidizes B-D-fucopyranose to D-fucono-é-lactone
which Spontaneously hydrolyzes to D-fuconate. The NAD(P)—
linked L-arabino-aldose dehydrogenase oxidizes B-D-fuco-
furanose to D—fucono-Y-lactone which is hydrolyzed to
D-fuconate by a y-lactonase. D-Fuconate is dehydrated by
a Specific dehydratase to form 3,6-dideoxy-D-Ehgegg
hexulosonic acid (2-keto-3-deoxy-D-fuconate) which, in
turn, is cleaved by a Specific aldolase, 2-keto-3-deoxy-
D-fuconate aldolase, to yield pyruvate and D-lactaldehyde.
No enzyme activity which modified D-fucose by means of
isomerization, phoSphorylation, epimerization, or reduc-
tion could be demonstrated. It was concluded that D-fucose
was not degraded by a pathway analogous to those which

occur in other organisms for the degradation of L-fucose

A. Stephen Dahms Jr.
or L-rhamnose, which are the two other common 6-deoxy
aldohexoses.

The five enzymes of the D-fucose pathway were
purified and some of their prOperties determined and
compared to similar enzymes in the literature. The
Specific activities of the enzymes in cell extracts and
the Km values are in physiologically significant ranges.
The intermediates of the Dbfucose pathway were also iso-
lated and were identified by derivatization and chemical
synthesis.

The NAB-dependent D-aldohexose dehydrogenase has
been purified 327-fold, is NAB-Specific, is not affected
by thiols, thiol inhibitors, or metal ion activators, is
induced by growth on D-fucose and D-glucose. and oxidizes
D-fucose. D-galactose, 3,6-dideoxy-D—galactose, 2-deoxy-
D-galactose, Dbglucose, 2-deoxy-D-glucose, 6-deoxy—D—
glucose, D-allose, D-altrose, and D-mannose to the cor-
responding 6-1actones.

The NAD(P)-dependent dehydrogenase has been puri-
fied 276-fold, is Operative equally well with either NAD
or NADP, is not affected by thiols, thiol inhibitors, or
metal ion activators, is induced by growth on D-fucose,
D-galactose, L-arabinose. and 6-iodo-6-deoxy—Dbga1actose,
and oxidizes D-fucose, Dbgalactose, 3,6-dideoxy—D-
galactose, 2-daoxy-D-galactose, L-arabinose. L-mannose,
and 6-iodo-6-deoxy-D-ga1actose to the correSponding y-

lactones.

A. Stephen Dahms Jr.

The lactonase has been purified 16-fold, hydro-
lyzes y-D-lactones, is not activated by thiols or metal
ions,is not inactivated by thiol inhibitors or EDTA, and
is induced by growth on D-fucose, L-arabinose, and 6-iodo-
6-deoxy-D-galactose.

D-Fuconate dehydratase has been purified 30-fold,
catalyzes the irreversible dehydration at C—2 and C-3
32522 hydroxyl groups of its substrates, is Specific for
D-fuconate and L-arabonate out of 20 sugar acids tested,
has an absolute requirement for a divalent cation, is
activated by thiols and is inactivated by thiol inhibitors,
and is induced by growth on D-fucose, L-arabinose, and D-
galactose.

2-Keto-3-deoxy-D-fuconate aldolase has been puri-
fied 50-fold, catalyzes the C-3, C-4 cleavage of its sub-
strate to form pyruvate and an aldehyde, is Specific for
Z—keto-3-deoxy-Dbfuconate and 2-keto-3-deoxy-L-arabonate
out of 9 deoxy sugar acids tested, has an absolute
requirement for a divalent cation, is stabilized by
thiols and is induced by growth on D-fucose and L-arabinose.

With the exception of the NAB-linked D-aldohexose
dehydrogenase, all the enzymes of the D-fucose pathway act
upon L-arabinose or its corresponding intermediates.
Mutants which were lacking D-fuconate dehydratase and
2-keto-3-deoxy-D-fuconate aldolase showed defective
growth on both D-fucose and L-arabinose and support the

hypothesis that the enzymes of the D-fucose pathway also

A. Stephen Dahms Jr.
function in the degradation of L-arabinose. 2—Keto-3—
deoxy-L-arabonate, the product resulting from the D-
fuconate dehydratase-catalyzed dehydration of L-arabonate,
has been found to be degraded in other microorganisms by
Z-keto-3-deoxy-L-arabonate dehydratase. The latter
dehydratase could not be demonstrated in the pseudomonad
under investigation, and it has been concluded that a new
L-arabinose pathway has been found through which 2-keto-
3-deoxy-L-arabonate is cleaved by Z-keto-B-deoxy-D-

fuconate aldolase to yield pyruvate and glycolaldehyde.

METABOLISM OF D-FUCOSE
AND L-ARABINOSE IN A PSEUDOMONAD

By

\\

A; Stephen Dahms Jr.

A THESIS

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

DOCTOR OF PHILOSOPHY

Department of Biochemistry

1969

ACKNOWLEDGMENT

The author wishes to eXpress his sincere appre-
ciation to his mentor, Dr. Richard L. Anderson, for his
inestimable guidance, encouragement, and patience
throughout the graduate eXperience. He would also like
to thank his committee members, Dr. w. w. Wells, Dr. 8.
Aust, and Dr. w. Frantz and eSpecially Dr. L. L. Bieber,
for his counseling and guidance. The provocative dis-
putations with Dr. Joseph w. Mayo, Richard E. Palmer,
Thomas E. Hanson, and Virginia L. Sapico are deeply
appreciated.

The author is eSpecially grateful to his wife,
Judy, for her love and encouragement and for her hard
work as a chemist to aid with the finances throughout
the duration of this work. The support of the National
Institutes of Health predoctoral fellowship is appreci-
ated.

ii

VITA

A. Stephen L. Dahms Jr. was born on September 12,
19#3 in Mankato, Minnesota. He graduated from Central
High School in Saint Paul, Minnesota in June, 1961, and
then attended the College of Saint Thomas where his
interests in chemistry were further enkindled by the
beneficent goading of Drs. Morath, Ryan, and Allen, in
that order. In the winter of his 3rd year, Mr. Dahms
obtained employment as a chemist in the basic research
laboratory of the Division of Internal Chemicals at
Minnesota Mining and Manufacturing. During his 1% years
at 3M, Mr. Dahms became directly involved with research
in the catalysis of photochemical reactions of organo-
metallics, new syntheses of organometallics, development
of unique analytical techniques for the quantitative
analysis of organic peroxides and hydroperoxides, and
formulation and analysis of new elastomers. Deepite the
Spurious and malicious rumors that his research tech-
niques and phiIOSOphy of chemistry were irreversibly coag-
ulated while in the industrial environment, Mr. Dahms
considers his duration at 3M as being one of his most
didactic experiences. In June, 1965,1uaterminated his
physical affiliations with 3M, received a B.S. in chem-

istry, and departed for East Lansing, Michigan to commence

iii

graduate work in Biochemistry under the guidance of Dr.
R. L. Anderson. On December 31, 1966, to gain a needed
tax advantage, he married a fellow student, Judith Claire
Daneault, and a year later, 3 days before his Prelimin-
ary Examinations, a daughter, Jacquelyne Kristin, was
born.

Mr. Dahms was a National Institutes of Health
predoctoral trainee for the first two years of his
graduate experience and was awarded a National Institutes
of Health predoctoral fellowship for the latter two years.
Upon completion of his work at Michigan State University
in May, 1969, the author will accept a postdoctoral posi-
tion under Dr. Paul D. Boyer at UCLA. The research will
concern the mechanism of action and site Specificity of
uncouplers of energy-linked phoSphorylation. Mr. Dahms
has been awarded National Science Foundation and Atomic
Energy Commission postdoctoral fellowships for his dura-
tion at UCLA.

Mr. Dahms is a member of the American Chemical
Society, the American.Association for the Advancement of
Science, Sigma Xi, Phi Kappa Phi, the American Philatelic

Society, and the Faraday-Woodward Society.

iv

TABLE

ACKNONLEDGMENTS o o o o '

VITA . . . . . . . . .
LIST OF TABLES . . . .
LIST OF FIGURES . . . .
ABBREVIATIONS . . . . .
INTRODUCTION . . . . .
EXPERIMENTAL PROCEDURES

Cultivation of
tion of Cell Extracts . .

OF CONTENTS

0 O O O O O

Bacteria and Prepara-

Analytical Procedures . .

Reagents . .

Assay for Aldose-Ketose Isomerization.

Assay for Kinase Activity

Assay for Aldose Reductase Activity

NAD(P)-Dependent Dehydrogenase Assay

MAD-Dependent Dehydrogenase Assay

Lactonase Assay‘ . . . . .

Dehydratase Assay . . . .

2-Keto-3-deoxy-D-fuconate Aldolase

Endpoint Assay . . . . . .

2-Keto—3-deoxy-D-fuconate Aldolase

Standard Assay . . . . . .

Preparation of D-Fucose .

Preparation of Potassium D-Fuconate

Page

ii

iii
xi
xiv

xix

N

000 O\\x) N

10
10
11
12

13

13
13
19

RESULTS

A.

Selection of Mutant Strains . . . . .

C O O O O O O O O O O O C O O O O O O O

INVESTIGATION OF VARIOUS POSSIBLE ENZY-
MATIC REACTIONS INVOLVING D-FUCOSE . . .

Isomerization or Eprmerization . . .
PhOSphorylation . . . . . . . . . . .
Reduction . . . . . . . . . . . . . .
Dehydrogenation . . . . . . . . . . .

CHARACTERIZATION OF THE NAD(P)-DEPENDENT
DEHYDROGENASE . . . . . . . . . . . . . .

1. Purification . . . . . . . . . . . .
Protamine Sulfate Fractionation . . .
Heat Step . . . . . . . . . . . . . .
Ammonium Sulfate Fractionation . . .
Sephadex G-200 Chromatography . . . .
Calcium.PhQSphate . . . . . . . . . .

2. PrOperties . . . . . . . . . . . . .
pH Optimum . . . . . . . . . . . . .
Substrate Specificity . . . . . . . .
Nucleotide Specificity . . . . . . .
Reversal of the Dehydrogenation . . .
Stability . . . .r. . . . . . . . . .
Induction . . . . . . . . . . . . . .

3. Product Identification . . . . . . .

Enzymatic Preparation of D-Fucono-
Y-laCtone o o o o o o o o o o enema/o

Enzymatic Preparation of Potassium
D-Fuconate O O O O O O O I O O O O 0

vi

Page
20
22

22
22
25
25
25

31
31
31
31
33
33
33
3b
34
34
37
37
54
57
57

57

59

C.

D.

CHARACTERIZATION OF THE NAD—DEPENDENT
DEHYDROGENASE O O O O O O O O O O O O O O

1. Purification . . . . . . . . . . . .
Protamine Sulfate Fractionation . . .
Ammonium Sulfate Fractionation . . .
Sephadex G-200 Chromatography . . . .
DEAE-cellulose Chromatography . . . .
Calcium PhoSphate Gel . . . . . . . .

2. PrOperties . . . . . . . . . . .'. .
pH Optima . . . . . . . . . . . . . .
Substrate Specificity . . . . . . . .
Nucleotide Specificity . . . . . . .
Reversal of the Dehydrogenation . . .
Anomer Preference . . . . . . . . . .
Stability . . . . . . . . . . . . . .
Induction . . . . . . . . . . . . . .

3. Product Identification . . . . . . .
Product Isolation . . . . . . . . . .

CHARACTERIZATION OF D-FUCONO-Y-LACTONASE

1. Preliminary Experiments . . . . . . .

2. Purification. . . . . . . . . . . . .
Protamine Sulfate Fractionation . . .

Ammonium Sulfate Fractionation . . .

Sephadex G—200 Chromatography
3. Properties . . . . . . . .1. . . . .
pH Optimum . . . . . . . . . . . . .
Substrate Specificity . . . . . . . .

vii

Page

64
64
64
64
66
66
67
67
67
67
71
71
91
99
105
105
107
108
108
108
108
112
112
112

. 112

113

4.

Stability o o o o o o o o o o o o o o 0
Induction .q. . . . . . . . . . . . . .
Product Identification . . . . . . . .

Preparation of the Reaction Product . .

CHARACTERIZATION OF D-FUCONATE DEHYDRATASE.

1.
2.

3.

Preliminary EXperiments . . . . . . . .
Purification . . . . . . . . . . . . .
Protamine Sulfate Fractionation . . . .
Ammonium Sulfate Fractionation . . . .
Sephadex G-200 Chromatography . . . . .
DEAE-cellulose Chromatography . . . . .
PrOperties . . . . . . . . . . . . . .
pH Optimum . . . . . . . . . . . . . .
Substrate Specificity . . . . . . . . .
Metal Ion Activators . . . . . . . . .
Sulfhydryl Activation and Inhibition .
Reversibility of Dehydration . . . . .
Stability . . . . . . . . . . . . . . .
Induction . . . . . . . . . . . . . . .

Identification of the Dehydration
PrOdUCtoooooooooooooooo

Enzymatic Synthesis of the Dehydration
PrOd-uct O O O O 0 O O O O O O O O O I 0

Absorption Spectra of the Z-Thiobar-
bituric Acid Chromogen . . . . . . . .

Ceric Sulfate Decarboxylation . . . . .
a-Keto Acid Derivatives . . . . . . . .

Periodate Oxidation . . . . . . . . . .

viii

Page
113
122
122
122
125
126
127
128
128
128
130
130
130
137
137
143
143
147
147

151

151

152
155
161
168

Cleavage by 2-Keto-3-deoxy-D-fuconate
AldOlaSe O O O O O O O O O O O O O O

5. Chemical Synthesis of 2-Keto-3-deoxy-
D-fuconate.............

CHARACTERIZATION OF Z-KETO-B-DEOXY-D-
FUCONATE ALDOLASE . . . . . . . . . . .

1. Preliminary EXperiments . . . . . .
2. Purification . . . . . . . . . . . .
Protamine Sulfate Fractionation . .
Ammonium Sulfate Fractionation . . .
Sephadex G-200 Chromatography . . .
Heat Step . . . . . . . . . . . . .
3. Properties . . . . . . . . . . . . .
pH Optimum . . . . . . . . . . . . .
Substrate Specificity . . . . . . .
Metal Ion Activation . . . . . . . .
Sulfhydryl Activation . . . . . . .
Equilibrium Constant . . . . . . . .
Stability . . . . . . . . . . . . .
Induction . . . . . . . . . . . . .

4. Identification of the Cleavage
PrOdUCtSoooooooooooooo

ANALYSIS OF MUTANTS LACKING D-FUCONATE
DEHYDRATASE AND Z-KETO-3-DEOXY-D-FUCONATE
ALDOLASE, AND THE RELATIONSHIPS AMONG THE
DEGRADATIVE PATHWAYS FOR D-FUCOSE, L-
ARABINOSE, AND D-GALACTOSE . . . . . . .

1. Mutant Strain 5-1-10-1 (D-Fuconate
DehydrataseleSS) o o o o o o o o o o

2. Mutant Strain 73-1-2 (Lacking 2-Keto-
3-deoxy-D-fuoonate Aldolase) . . . .

ix

Page

173

176

187
187
195
195
197
197
197
198
198
198
198
209
209
209
216

216

226

226

232

Page
DISCUSSION . . . . . . . . . . . . . . . . . . . . 238
SUMMARY . . . . . . . . . . . . . . . . . . . . . 269
BIBLIOGRAPHY . . . . . . . . . . . . . . . . . . . 270
APPENDIX A . . . . . . . . . . . . . . . . . . . . 284
APPENDIX B . . . . . . . . . . . . . . . . . . . . 291

Table
I.

II.

III.

IV.

VI.

VII.

VIII.

XI.

XII.

XIII.

XIV.

XV.

LIST OF TABLES

Chromatographic analysis for epimeriza-

tion of D-fucose . . . . . . . . .

Spectrophotometric analysis for phos-
phorylation of D-fucose . . . . .

Spectrophotometric analysis for reduc-
tion of D-fucose . . . . . . . . .

Spectrophotometric analysis for NAD-
linked oxidation of D-fucose . . .

Correlation between D—fucose oxidation,
NAD reduction, and lactone formation

Purification of the NAD(P)-dependent
dehydrogenase . . . . . . . . . .

Substrate Specificity of the NAD(P)-
dependent dehydrogenase . . . . .

Tabulation of Km and V ax values for
the NAD(P)-dependen¥ dehydrogenase

Effect of mixed substrates on NAD(P)-
dependent dehydrogenase activity .

Reversibility of the NAD(P)-dependent
dehydrogenase. . . . . . . . . . .

Effects of various reagents on NAD(P)-
dependent dehydrogenase activity .

Inducibility of the NAD(P)-dependent
dehydrogenase. .7. . . . . . . . .

Physical prOperties of potassium D-
fuconate and D-fucono-Y-lactone .

Purification of the NAD-dependent
dehydrogenase. o o o o o o o o o o

Tabulation of Km and Vmax values for
the NAD-dependent dehydrogenase .

xi

Page

23

26

27

28

30

32

38

48

49

55

56

58

6O

65

7O

Table
XVI.

XVII.

XVIII.

XIX.

XXVI.

XXVII.
XXVIII.

XXXI.

XXXII.

XXXIII.

XXXIV.

Effect of mixed substrates on NAD-
dependent dehydrogenase activity

Reversibility of the NAD-dependent
dehydrogenase. o o o o o o o o 0

Effects of various reagents on NAD-
dependent dehydrogenase activity

Inducibility of the NAD-dependent
dehydrogenase. . . . . . . . . .

Purification of the y-lactonase,. .

Substrate Specificity of the y-lactonase.

Effects of various reagents on y-
lactonase activity . . . . . . .

Inducibility of the y-lactonase . .

Purification of D-fuconate dehydratase

Effects of metal ions on D—fuconate
dehydratase activity . . . . . .

Sulfhydryl activation and inhibition of

D-fuconate dehydratase . . . . . . .

Inducibility of D-fuconate dehydratase

Ceric sulfate decarboxylation of 2-keto-

3-deoxy-D-fuconate . . . . . . .

Characterization of the dehydration

product as 2-keto-3-deoxy-D-fuconate

Purification of 2-keto-3-deoxy-D-
fuconate aldolase . . . . . . .

Effect of metal ions on 2-keto-3-deoxy-

D-fuconate aldolase . . . . . .

Effects of thiols and thiol group
inhibitors on aldolase activity

Equilibrium constants for substrates of

2-keto-3-deoxy-D-fuconate aldolase .

Inducibility of 2-keto-3-deoxy-D-
fuconate aldolase. . . . . . . .

xii

Page

88

92

104

106
111
116

121
123
129

142

146
150

160

175

196

205

210

213

217

Table

XXXVI.

XXXVII.
XXXVIII.

XXXIXO

XL.

XLI .

XLII.

XLIII.

Page

Identification of the cleavage products
of 2-keto-3-deoxy-D-fuconate . . . . . . 220

Identification of the cleavage products
of 2-keto-3-deoxy-L-arabonate . . . . . 225

Enzyme levels in mutant strain 5-1-10-1 . . 227

Comparison of the prOperties of D-fuconate
and D-galactonate dehydratases . . . . . 233

Enzyme levels in mutant strain 73-1-2 . . . 236

Comparison of various soluble bacterial
dehydrogenases Specific for non-phos-
phorylated monosacbharides . . . . . . . 245

Comparison of various aldonolactonases . . 254

Comparison of various aldonic acid
dehydratases . . . . . . . . . . . . . . 259

Comparison of various aldolases which
cleave keto deoxy acids . . . . . . . . 264

xifl.

Figure

9.

10.

11.

12.

LIST OF FIGURES

Reactions in the synthesis of D-fucose

pH Optimum of the NAD(P)-dependent
dehydrogenase . . . . . . . . . . .

Lineweaver-Burk plot relating NAD(P)-
dependent dehydrogenase activity to
D—galactose concentration . . . . .

Lineweaver-Burk plot relating NAD(P)-
dependent dehydrogenase activity to
D-fucose concentration . . . . . .

Lineweaver-Burk plot relating NAD(P)-
dependent dehydrogenase activity to
L-arabinose concentration .'. . . .

Lineweaver-Burk plot relating NAD(P)-
dependent dehydrogenase activity to
NAD+ concentration. . . . . . . . .

Lineweaver-Burk plot relating NAD(P)-
dependent dehydrogenase activity to
NADP+ concentration . . . . . . . .

General outline for the preparation of
D-fucose, Defuconate, and.D-fucono-

Y-laCtone o o o o o ,0 o I o o o o o 0

pH Optimum of the NAD-dependent dehydro-

genase O O O O O O O O O O O O O O

Lineweaver-Burk plot relating NAD-
dependent dehydrogenase activity to
D-fuoose concentration . . . . . .

Lineweaver-Burk plot relating NAD-
dependent dehydrogenase activity to
D-glucose concentration . . . . . .

Lineweaver-Burk plot relating NAD-

dependent dehydrogenase activity to
D-galactose concentration . . . . .

xiv

Page

19

36

40,41

43 .44

46,47

51

53

63-

69

73

75

77

Figure Page

13. Lineweaver-Burk plot relating NAD-
dependent dehydrogenase activity to
D-mannose concentration . . . . . . . . 79

14. Lineweaver-Burk plot relating NAD-
dependent dehydrogenase activity to
2—deoxy-D-glucose concentration . . . . 81

15. Lineweaver—Burk plot relating NAD-
dependent dehydrogenase activity to

2-deoxy-D-galactose concentration . . . 83
16. Lineweaver-Burk plot relating NAD-

dependent dehydrogenase activity to

D-altrose concentration . . . . . . . . 85
17. -Lineweaver-Burk plot relating NAD-

dependent dehydrogenase activity to

D-allose concentration . . .~. . . . . 87
18. Lineweaver-Burk plot relating NAD-

dependent dehydrogenase activity to

NAD+ concentration. . . . . . . . . . . 9O

19. Lineweaver-Burk plot relating NAD-
dependent dehydrogenase activity to
D-glucono-b-lactone concentration . . . 94

20. Lineweaver-Burk plot relating NAD-
dependent dehydrogenase activity to

NADH concentration . . . . . . . . . . 96
21. Identification of D-glucose as the reduc-

tion product of D-glucono-5-lactone

by the NAD-dependent dehydrogenase . . 98
22. Comparison of the rate of oxidation of

0.- axld a,B-D-glucose o o o o o o o o o 101

23. Thermal denaturation of the NAD-
dependent dehydrogenase relative to
the NAD(P)-dependent dehydrogenase . . 103

24. Participation of a lactonase in the
hydrolysis of Dbfucono-Y-lactone . . . 110

25. pH Optimum Of the y-lBCtomseo o o o o o o 115

26. Lineweaver-Burk plot relating y-lactonase
activity to D-galactono-Y-lactone
concentration 0 o o o o o o o o o o o o 118

27. Thermal denaturation of the y-lactonase . 120

XV

Figure Page

28. Fractionation of D-galactonate and D-
fuconate dehydratases on Sephadex
G-ZOOooooooooooooooooo 1.32

29. Fractionation of dehydratases on DEAE-
cellulose . . . . . . . . . . . . . . . 134

30. pH Optimum of D-fuconate dehydratase . . . 136

31. Lineweaver-Burk plot relating D—fuconate
dehydratase activity to D-fuconate
concentration 0 o o o o o o o o o o o o 139

32. Lineweaver-Burk plot relating D—fuconate
dehydratase activity to L-arabonate
concentration 0 o o o o o o o o o o o o 141

33. Effect of Mg2+ concentration on D-fuconate
dehydratase activity . . . . . . . . . 145

34. Thermal denaturation of D-fuconate
dehydratase relative to D-galactonate
dehydratase . . . . . . . . . . . . . . 149

35. Dowex—l formate chromatography of the
enzymatically prepared dehydration
pI‘OduCt O O O O O O O O O O I O O O O 0 151+

36. Absorption Spectra of the TBA chromogen
resulting from the periodate oxidation
of the dehydration product . . . . . . 157

37. Stability of the B-formyl pyruvate-TBA
complex under the conditions of the
assay 0 O O O O O .0 O O O 0 O O O O O O 159

38. Absorption Spectra of the TBA chromogen
resulting from the periodate oxidation
of the sodium borohydride-reduced and
ceric sulfate-oxidized dehydration
product 0 O L O O O O O O O O O O O O 0 1 63

39. Absorption Spectrum of the 3-methyl-2-
benzothiazolinone hydrazone azine of
the dehydration product . . . . . . . . 165

40. Absorption Spectrum of the semicarbazone
Of the dehydration prOdUCt o o o o o o 167

41. Absorption spectrum of the quinoxaline
derivative of the dehydration product . 170

xvi

Figure Page

42, Comparison of the rates of release of
B‘fOrmyl pyruvate o o o o o o o o o o o 172

43. Reaction rate of the chemical synthesis
of 3,6-dideoxy-DL-hexulosonic acid . . 179

44. Elution of 3,6-dideoxy-DL-hexulosonic
acid from Dowex-i formate using a
0-5.5 molar formic acid gradient . . . 182

45. Elution of 3,6-dideoxy-DL—hexulosonic
acid from Dowex-l formate using a
0-0.06 molar formic acid gradient . . . 184

46. Sephadex G—200 profile of 2-keto-3-deoxy-
D-fuconate aldOlaSG . o o o o g o o o o 190

47. Sephadex G—200 profile of 2-keto-3-deoxy-
D-fuconate.aldolase, 2-keto-3-deoxy-
D-galactonate kinase, and 2-keto-3-
deoxy-6-phoSpho-D-galactonate aldolase. 194

48. pH Optimum of 2-keto-3-deoxy-D-fuconate
aldOlase. . C O C O O O O O O O O I O O 200

49. Lineweaver-Burk plot relating 2-keto-3-
deoxy-D-fuconate aldolase activity to
2-keto-3-deoxy-D-fuconate concentra-
tionooooooooooo‘oooooo 202

50. Lineweaver-Burk plot relating 2-keto-3-
deoxy-D-fuconate aldolase activity to
2-keto-3-deoxy-L-arabonate concentra-
tion 0 O O O O O O O O O O O 0 O I O O 204

51. Analysis of the cleavage reactions by
Sephadex G-200 fractions in the
absence and presence of MnClz . . . . . 208

52. Time-dependent thiol renaturation of
2-keto-3-deoxy-D-fuconate aldolase . . 212

53. Thermal denaturation of 2-keto-3-deoxy-
D-fuconate aldolase . . . . . . . . . . 215

54. Absorption Spectra of the 2,4-dinitro-
phenylhydrazones of pyruvic acid and
SPOt #1 O O C O O O O '0 O C O O O O O O 222

55. Absorption Spectra of the 2,4-dinitro-

phenylhydrazones of D-lactaldehyde
andsp0t#zoooooooooooooo 22""

xvii

Figure Page

56. Growth of wild-type cells on D-fucose,
L-arabinose, D-galactose, and D-glu-
cose . . . . . . . . . . . . . . . . . 229

57. Growth of mutant strain 5-1-10-1 on
D-fucose, L-arabinose, D—galactose,
and D-glucose . . . . . . . . . . . . 231

58. Growth of mutant strain 73-1-2 on
D-fucose, L-arabinose, D-galactose,

and. D-glucose o o o .‘ o o o o o o o o 235
59. Pathway of D-fucose degradation . . . . . 239
60. LnArabinose degradation in pseudomonads . 240
61. All trans configuration of B-furanose
forms of D-fucose, D-galactose, and
L-arablnose............. 251
B-l. Electron micrographs of the pseudomonad . 295

xviii

ATP

Bicine
DEAE-cellulose
EDTA

Hepes

KDA
KDF
NAD
NADH

NADP
NADPH

TBA
Tris

ABBREVIATIONS

adenosine triphoSphate
N,N-bis(2-hydroxyethyl)glycine
diethylaminoethyl-cellulose
ethylenediamine tetraacetate

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

2-keto-3—deoxy-Lparabonate
2-keto-3-deoxy-D—fuconate
nicotinamide adenine dinucleotide

reduced nicotinamide adenine
dinucleotide

nicotinamide adenine trinucleotide

reduced nicotinamide adenine
trinucleotide

2—thiobarbituric acid

2-amino-2-(hydroxymethyl)-1.3-
propanediol

xix

INTRODUCTION

D-Fucose (6—deoxy-D-galactose) is generally con-
sidered to be a rare carbohydrate whereas the L-enantio-
morph is considered to be relatively abundant and wide-
Spread in nature. However, many published reports on
the natural occurrence of B-D-fucosides (see Literature
Survey in the appendix) indicate that the D-enantiomorph
is more prevalent than is generally appreciated. Further,
Specific B-D-fucosidases have been found in several
organisms, and a microbial membrane tranSport system
active on D-fucose has been thoroughly documented. How-
ever, deSpite the plethora of information concerning
D-fucose, its biodegradative pathway has not been
described for any organism.

The purpose of this investigation was to eluci-
date the biodegradative pathway of D-fucose in a bacter-
ium and to characterize the participating enzymes. The
pathway was determined to be: D—fucose.o D-fucono-Y-
lactone + D-fucono-b-lactone 4'D-fuconate-e 2-ket043-
deoxy-D-fuconate e1pyruvate + D-lactaldehyde. In addi-
tion, it was found that the enzymes instrumental in D-
fucose degradation also function in a new pathway for
L-arabinose degradation. An abstract on aSpects of this

work has been published (1).

EXPERIMENTAL PROCEDURES

Cultivation of Bacteria and Preparation of Cell
ExtractS- The organism used in this study was selected
for its ability to utilize D-fucose as a sole carbon
and energy source and was classified as a pseudomonad.
Details of the isolation and identification are given
in the Appendix B. The organism was grown aerobically
in Fernbach flasks on a rotatory shaker at 32°C in a
one liter medium consisting of 1.35% NaZHP04-7H20, 0.15%
KHZPOLP, 0.3% (NHstou. 0.02% MgSOu'7H20, 0.000% FeSOu,
and 3.5% carbohydrate (autoclaved separately). Step-
wise addition of the carbohydrate was necessary since
1% carbohydrate markedly reduced the growth rate. After
inoculation, 20 ml 25% w/v of the carbohydrate was intro-
duced per liter of medium; 6 additional 20 ml 25% w/v
portions were added to the flasks at six hour intervals
before harvesting. Very thick cell suSpensions (absor-
bance at 600 nm, 18-22) were obtained using this tech-
nique. Turbidity measurements were made with a Coleman
Junior Spectrophotometer on appropriate dilutions in 18
mm test tubes. The cells were harvested 4 hours after
the final addition of the carbohydrate and approximately
40 hours after innoculation. The cells were resuSpended

in glass-distilled water and recentrifuged prior to

sonic disruption.

Extracts were prepared from washed cells in ice-
cold 0.10 molar sodium phoSphate buffer (pH 7.0) or 0.10
molar Bicine buffer (pH 7.4) by a 13 minute eXposure to
10-kc sonic oscillation in a Raytheon sonic oscillator
equipped with an ice-water cooling jacket. The superna-
tant fluid resulting from a 10 minute centrifugation at
40,000 x g was used as the cell extract.

Analytical Procedures- Reducing sugars were deter-
mined by the method of Sumner and Howell (1A). Aldonic
acids were determined after conversion to the correSpond-
ing lactones by heating in 1 N HCl for 5 minutes. Lactones
were determined as the hydroxamic acids by the method of
Hestrin (2). Descending paper chromatography was per-
formed using Whatman #1 filter paper in the following
six solvent systems:

(1) water saturated 2—butanone

(2) n-propanol-formic acid-water (6:3:1)
(3) water saturated phenol

(4) pyridine-0.10 N HCl- n-butanol (3:2:5)
(5) n-butanol-water-95% ethanol (5:4:1)
(6) n-butanol-pyridine-water (6:4:3)

Carbohydrates were visualized by benzidine-HCl
and periodate (3), N,N-dimethylfip-phenylenediamine (4)
and alkaline silver nitrate (5). Lactones were visual-
ized by their formation of hydroxamic acids. 2-Keto-3-

deoxy aldonic acids were visualized with the periodate-

1+
thiobarbituric acid reagents of Warren (6). a-Keto acids
were located by Spraying with 0.1% 2,4—dinitrOphenyl-
hydrazine in 2 N,HC1 followed by 10% KOH (18).

Protein was determined SpectrOphotometrically with
the aid of a nomgraph (courtesy of Calbiochem) based on
the data of Warburg and Christian (7). In crude extracts
or preparations high in nucleic acid content, protein
concentration was estimated by the biuret method (8),
with bovine serum albumin as the standard. Light measure-
ments were performed on a Gilford 2400 absorbance-record-
ing Spectrophotometer thermostated at 25.0°C using micro-
cuvettes with a 1 cm light path or on a Gilford 300
digital Spectrophotometer also with a 1 cm light path.
Spectra of the quinoxaline derivatives, the B-formyl
pyruvate-thiobarbituric acid chromogen, the malondialde-
hyde-thiobarbituric acid chromogen, the 3-methy1-2-
benzothiazolinone hydrazone azine, and the tetraazopenta-
methine cyanine derivatives of 3-methyl-2—benzothiazolinone
hydrazone were recorded with a Cary model 15 Spectrophoto-
meter.

L-Ribulose and D-fuculose were determined by the
cysteine-carbazole method (9) with Lnrhamulose as the
standard. The cysteine-carbazole reagents were added to
aliquots containing the ketoses and were incubated at
35°C for 10 minutes. Under the above conditions, using
an 18 mm path length, 0.1 umole of L-rhamulose in 7.4 ml
of solution gave an absorbance of 0.28 at 540 nm (36).

5

Standard L-rhamulose was the gift of Dr. J. W. Mayo.

Pyruvate was determined with lactic acid dehydro-
genase and NADH and by the method of Friedman and Haugen
(10) as modified by Sayre and Greenberg (11). Acetalde-
hyde was determined by the method of Barker and Summerson (268)
using periodate removal techniques of Itagaki and Susuki
(12). Carbon dioxide from ceric sulfate decarboxylations
of a-keto acids was measured manometrically according to
the procedure of Meister (13). Sodium borohydride reduc-
tions were carried out according to the method of
Ghalambor, Levine, and Heath (14). Carbonyl compounds
were determined as the tetraazopentamethine cyanine
derivatives with 3-methyl-2-benzothiazolinone hydrazone
(15). Periodate consumption was determined by the method
of Dixon and Lipkin as modified by Sugimoto and Okazaki
(16.269).

2-Keto-3-deoxy-D-fuconate and 2—keto-3-deoxy-D-
galactonate were determined using modifications in the
2-thiobarbituric acid (TBA) assay of Weissbach and
Hurwitz (17). The TBA assay for B-formyl pyruvate,
which is liberated upon periodate oxidation of 2-keto-3-
deoxy sugar acids, is influenced by unknown factors (18)
and necessitated the utilization of different molar
absorptivities for the reSpective 2-keto-3-deoxy sugar
acids. The molar absorptivities for KDF and 2-keto-3-
deoxy-D—galactonate were found to be 27,900 and 60,500

6

reSpectively by determining c-keto acid content in the
semicarbazide assay (19) and by determining the amount
of C02 released upon ceric sulfate decarboxylation.
Microdetermination of c-keto acids was also carried out
on the azines of the 3-methyl-2-benzothiazolinone hydra-
zones (20).
Thin-layer chromatography was performed in the
following three solvent systems:
(i) pyridine-0.10 N HCl-n-butanol (3:2:5)
(ii) n-butanol-ethanol-O.5 N NHAOH (70:10:20)
(111) benzene-tetrahydrofuran-glacial acetic acid (57:35:8)
Reagents- D-Fucose was synthesized by modifications
of the method of Freundenberg §t_gl. (21) and Schmid.g§
21' (22). Details for the preparation of this compound
are given below. 6—Iodo-6-deoxy-D-galactose was prepared
by modifications of the method of Raymond and Schroeder
(23). Potassium salts of D-galactonic, L-arabonic,
D-arabonic, D-xylonic, and D-lyxonic acids were prepared
by the method of Moore and Link (24). Potassium salts
of D- and L-fuconic and 6-iodo-6-deoxy-D-galactonic
acids were prepared by hypoiodide oxidation of the cor-
reSponding aldoses. Potassium 3-deoxy-Dgxylg-hexonate
was prepared by KOH hydrolysis of D-galacto-a-metasac-
charinic acid lactone synthesized by the method of Evans
.gt‘gl. (25). D-Glucarate and D-galactarate were obtained
from Sigma and were crystallized as the dicyclohex-

<ammonium salts according to the procedure of Fish and

7
Blumenthal (26). Potassium salts of D-mannonic, L-
rhamnonic and cellobionic acids were prepared by hypoio-
dide oxidation but were not crystallized. D-Ribonate and
D-gluconate were prepared by alkaline hydrolysis (pH 8.5)
of the correSponding y- and é-lactones, the gifts of Dr.
W. A. Wood, L-Galactonate and L-gluconate were prepared
by the alkaline hydrolysis of the correSponding y- and
6-lactones, the gifts of Dr. J. W. Mayo. The free acids
were generated by treatment with Ac50w-X8-H+, 50-100
mesh, and lyophilization. Pyridine nucleotides were
obtained from P-L Biochemicals. Protamine sulfate and
2-amino—2-(hydroxymethyl)-1,3-propanediol-HCl buffer
(Tris-HCl) were obtained from Sigma Chemical Co.
D-Galactono-y-lactone and standard D-fucose were obtained
from General Biochemicals. LiAlH4 was obtained from K
and K Laboratories and was finely powdered under an
anhydrous atmOSphere before use. Thin-layer chromatog-
raphy plates were Silica Gel F254 (Brinkman). N-2-
Hydroxyethylpiperazine-N-Z'-ethanesulfonic acid buffer
(Hepes), N,N-bis(2-hydroxyethyl)glycine buffer (Bicine),
N-tris(hydroxymethyl)methyl glycine buffer (Tricine),
N-tris(hydroxymethyl)methyl-Z-aminoethanesulfonic acid
buffer (TES), DL-threonine and L-threonine were obtained
from Calbiochem. Ethyl methanesulfonate and 2-thio-
barbituric acid were obtained from Eastman: the latter
compound was recrystallized by the procedure of

Waravdekar and Sasalaw (27). 3-Methyl-2-benzothiazolinone

8
hydrazone-H01 was obtained from Aldrich Chemicals and
was used without further purification.

Barium salts of 2-keto-3-deoxy-6-phoSpho-D-
galactonate and 2-keto-3-deoxy-6-phoSpho-D-gluconate
were the gifts of Dr. W. A. Wood. The barium salts
were converted to the sodium form by treatment with
AG50W-X8-Na+, and portions were dephoSphorylated with
intestinal alkaline phOSphatase (obtained from General
Biochemicals). The phosphatase reaction mixture con-
sisted of 2 umoles of phOSphate ester, 1 umole MgClg,
and 5 mg alkaline phOSphatase in a volume of 5 ml. The
reaction was followed titrimetrically with a Sargent
recording pH stat at pH 8.3. The reaction was carried
out at 28°C until proton release had ceased; the reac-
tion mixture was then neutralized with a cation exchange
resin and filtered.

Mzgpinositol and dulcitol were the gifts of Dr.
W. W. Wells. Glucose-free 2—deoxy-Deglucose, glucose-
free D-mannose, galactose-free 2-deoxy-D-galactose, D-
altrose, D-allose, adonitol, and xylitol were the gifts
of Dr. W. A. Wood. L~Mannose, L-glucose, L-galactose,
and L-fructose were the gifts of Dr. J. W. Mayo.

2-Keto-4-hydroxy glutarate was the gift of Dr.
Eugene Dekker (University of Michigan). 2.Acetamido-
2-deoxy-D-allose and 2-acetamido-2-deoxy—D—altrose were
provided by Dr. M. B. Perry (National Research Council

of Canada, Ottawa). 3,6-Dideoxy-D-galactose was the

9
gift of Dr. Otto Lfideritz (Max Planck Institute, Frei-
burg). Nchetylneuraminic acid was provided by Dr. w.
Jourdian (University of Michigan). 6-Deoxy—D-allose
and 6-deoxy-D-glucose were the generous gifts of Dr. T.
Reichstein (University of Basel, Basel, Switzerland).

PHOSphate buffers were prepared from the sodium
salts, and all melting points are uncorrected.

.ASSay for.Aldose-Ketose Isgmerization- Isomerase
activity was measured colorimetrically by the cysteine-
carbazole method. The reaction mixture consisted of 20
umoles of the aldose, 2 umoles MgClz, 5 umoles phOSphate
buffer (pH 7.0), and 1-5 mg of the cell extract. The
mixture was incubated at 30°C and aliquots were with-
drawn at time intervals and assayed for ketose.

.Assay for Kinasefgptivity- Kinase activity was
determined Spectrophotometrically at 340 nm and 25.0°C.
The reaction mixture contained 10 umoles of the aldose,
0.5 umole ATP (pH 7.2), 1 umole MgClZ, 0.5 “mole phos-
phoenolpyIUvate (pH 7.2), 0.07 umole NADH, 5 umoles
glycylglycine buffer (pH 7.2), 13 pg lactate dehydro-
genase-pyruvate kinase, and 0.05-0.1 mg of cell extract
in a total volume of 0.15 ml. .A control to correct for
NADH oxidase and.ATPase contained all the reaction com-
ponents except the aldose. _

Agggy for.Aldose Reductase Activity-.Activity was
measured Spectrophotometrically at 340 nm by following
the oxidation of NADH or NADPH. The assays consisted of

10

0.07 umole NAD(P)H, 5 umoles of glycylglycine buffer
(pH 7.2), and 10 umoles of the aldose in a volume of
0.15 ml. (A control to correct for NADH oxidase con-
tained all the reaction components except the aldose.

NAD(P):Dependent Dehydrogenase Assay- Activity
was measured SpectrOphotometrically at 340 nm and 25.0°C.
The reaction mixture (0.15 ml) consisted of 2.5 umoles
D-galactose, 0.3 umoles NADP+, 15 umoles Tris-HCl buf-
fer (pH 8.1), and a limiting amount of the enzyme. The
assay was linear with reSpect to time and enzyme con-
centration. All carbohydrate solutions used in the
analysis of the NAD(P)-dependent and the NAD-dependent
dehydrogenases were allowed to approach mutarotational
equilibrium by incubation at room temperature for
periods not less than two hours; this was done to mini-
mize the possible effects resulting from the preferential
enzymatic oxidation of one anomer at a rate greater than
the rate of anomerization. A unit of NAD(P)-dependent'
dehydrogenase was defined as the amount of enzyme which
catalyzed the reduction of 1 umole of NADP+ per hour.

NAD-Dependent Dehydrogenase Assay- Activity was
measured SpectrOphotometrically at 340 nm and 25.0°C.
The reaction mixture (0.15 ml) consisted of 2.5 umoles
D-glucose, 0.3 umole NAD+, 15 umoles Tris-H01 buffer
(pH 8.1), and a limiting amount of the NAD—dependent

dehydrogenase. The assay was linear with reSpect to

11
enzyme concentration and time. A control to correct for
NADH oxidase was necessary in the early steps of purifi-
cation and contained all the components except D-glucose.
NADH oxidase could be completely eliminated by preparing
sonic extracts in a solution consisting of 0.10 molar
Bicine and 0.147 mM 2-thioethanol (pH 7.4). A.unit of
NAD-dependent dehydrogenase activity was defined as the
amount of enZyme which catalyzed the reduction of 1 umole
of NAD+ per hour.

Lactonase Assay? Enzyme activity was measured
Spectrophotometrically at 540 nm using the hydroxamic
acid procedure of Hestrin (2). The standard curve was
prepared using D-galactono-Y-lactone. The assay con-
sisted of 10 umoles D-galactono-y-lactone, 40 umoles
Tris-HCl buffer (pH 7.5), and a limiting amount of the
lactonase in a total volume of 0.15 ml. The reaction
mixture was incubated at 30°C before quenching with 1.0
ml of the alkaline hydroxylamine reagent. The assay was
linear with reSpect to enzyme concentration and time.

A control to correct for the non-enzymatic hydrolysis

of lactones was necessary and contained all the compo-
nents except the lactonase. All lactone solutions were
freshly prepared before each assay. A unit of lactonase
activity was defined as the amount of enzyme which
catalyzed the hydrolysis of 1 umole of D-galactono-Y-

lactone per hour. All lactonase preparations were free

12
of contaminating NAD-dependent dehydrogenase, NAD(P)-
dependent dehydrogenase, D-galactonate dehydratase, and
D-fuconate dehydratase.

Dehydratase Assaye Two assays were used routinely
for the detection of deoxy hexulosonic acids. The TBA
assay was restricted to qualitative detection of KDF,
2-keto-3-deoxy-D-ga1actonate, and 2-keto-3-deoxy-L-
arabonate. (All enzyme-catalyzed dehydration reactions
were assayed for a-keto acids by the semicarbazide
method. The assay consisted of 20 umoles Bicine buffer
(pH 7.4), 0.147 umole 2-thioethanol, 5 umoles D-fuconic
of L-arabonic acid (pH 7.4), 5 umoles MgClZ, and a limit-
ing amount of the enzyme in a total volume of 0.15 ml.
The reaction mixture was incubated at 30°C for 30 minutes
and was quenched by the addition of 1.0 ml of the semi-
carbazide reagent. The mixture was incubated at 30°C for
15 minutes and diluted to 5.0 ml with water; the absor-
bance was determined at 250 nm in a 1 cm quartz cuvette
against a diluted reagent blank. Controls for contribu-
tions to absorbance by protein, buffer, thiol, and carbo-
hydrate were necessary. Under the above conditions, 0.5
umole of the c-keto acid gave an absorbance of 1.02. The
assay was linear with reSpect to enzyme concentration and
time. A.unit of dehydratase activity was defined as the
amount of enZyme which catalyzed the formation of 1 umole
of a-keto acid per hour. D-Galactonate dehydratase was

also assayed with the above procedure with the exception

13

that the 2—thioethanol was omitted.

2-Keto-3-deoxy-D—fuconateAldolase Endpgint Assayr
Endpoint assays were used for the determination of the
amount of pyruvate liberated upon enzymatic cleavage of
2-keto-3-deoxy-D-fuconate (KDF). Enzyme activity was
measured Spectrophotometrically at 340 nm and 25.0°C.
The reaction mixture (0.15 ml) consisted of 13 ug lactic
acid dehydrogenase, 0.07 umole NADH, 1 umole MgClz, 0.143
umole 2-thioethanol, 15 umoles Hepes buffer (pH 7.4),
0.02-0.04 umole KDF, and a limiting amount of the aldolase.
Levels of NADH oxidase and endogenous pyruvate were also
determined and appropriate corrections made.

2-Keto-3-deoxy-D-fuconate Aldolase Standard Assay-
Pyruvate liberated upon cleavage of KDF or KDA was mea-
sured in a coupled-assay using lactic acid dehydrogenase
and NADH. EnZyme activity was measured SpectrOphotomet-
rically at 340 nm and 25.0°C. The reaction mixture
(0.15 ml) consisted of 13 ug lactic acid dehydrogenase,
0.07 umole NADH, 1 umole MnClz, 0.147 umole 2-thioethanol,
15 umoles Hepes buffer (pH 7.8), 5 umoles KDF or KDA, and
a limiting amount of the enzyme. The assay was linear
with reSpect to enzyme concentration and time. Controls
to correct for NADH oxidase and possible endogenous pyru-
vate were run.

Preparation of D-Fucose- Anhydrous zinc chloride

(360 g) was added to a 6-liter Erlenmeyer flask equipped

14

with a polyethylene mixing blade and a mechanical stirrer.
Acetone (3.75 liters) was rapidly added and the suSpension
stirred until the zinc chloride was dissolved. Concen-
trated sulfuric acid (12.0 ml) was added dropwise from a
pipet, such that no acid would touch the inside of the
neck. dSD-Galactose (200 g), doubly recrystallized
according to the procedure of Wolfrom and Thompson (28)
and chromatographically homogeneous on Whatman #1 paper
in solvent systems 3, 4, and 5, was quickly added and the
suSpension stirred for 6 hours. A suSpension of anhydrous
sodium carbonate (600 g) in water (1.05 liters) was added
slowly over a period of 30 minutes. The zinc carbonate
suSpension was collected by suction filtration and was
washed several times by resuSpending in acetone and
refiltering. The filtrates were combined and reduced in
volume to a syrup under reduced pressure. The acetal was
extracted with three 400-ml portions of diethyl ether
which were combined, dried over anhydrous sodium sulfate,
and reduced in volume under vacuum to a syrup (bath tem-
perature 30°C). The acetal was freed from condensation
products of acetone by heating at 100°C and 0.05 mm Hg
for 1 hour: yield 271.3 g (93.5%) of a crude, undistilled
syrup, 1,2—3,4-di-O-iSOpropylidene-a-D-galactOpyranose.

Di-O-isapropylidene-D-galactose (270 g) was stirred
into anhydrous pyridine (520 ml) in a 1-liter, 3-necked
flask fitted with a mechanical stirrer and a drying tube.

- 1'5
peToluenesulfonyl chloride (285 g), recrystallized from
petroleum ether:ligroin (10:3), was added and the reac-
tion stirred for 5.5 hours during which time the tempera-
ture was maintained below 60°C. Pyridine-HCl was immedi-
ately filtered out, following completion of the reaction,
and was washed with pyridine and diethyl ether. Water
(10 ml) was added to the filtrate to destroy excess tosyl
chloride and the mixture was concentrated under reduced
pressure (bath 25°C) to approximately 250 ml. The solu-
tion was immediately poured into cold water (2.5 liters,
15°C) while being stirred moderately fast by a mechanical
stirrer. .After 10 minutes a precipitate developed which
solidified upon further standing. The solid residue was
broken up manually into chunks, filtered, and washed with
water. The product was suSpended in distilled water by
mixing in a large Waring blender at maximum Speed for a
period of 5 minutes after which time it was again filtered
and washed: this procedure was repeated until both pyri-
dine and colored impurities were removed. The product was
dried over anhydrous calcium sulfate before dissolving in
anhydrous methyl alcohol (1 liter) and treating with Norit
(20 g). Norit treatment was repeated until the solution
was free of all coloration. The filtrate (1400 ml) was
left for approximately 12 hours at 4°C: crystallization
began within 20 minutes. The product was collected by
suction filtration, washed with cold water (1.5 liter)

16
and dried in a vacuum desiccator over anhydrous calcium
sulfate. The product, 1,2-3,4-di-O-iSOpr0pylidene-6fip-
toluenesulfonyl—a-D—galactOpyranose, was obtained inwa yield of
88.2% (280 8): mp. 101-02°C; reported, 102-03°C (22).

The tosyl product (64.3 g) was dissolved in
anhydrous benzene (100 ml) followed by distillation to a
final volume of 70 ml. A suSpension of anhydrous diethyl
ether (140 ml) and 6.2 g lithium aluminum hydride were
added through a dropping funnel, and the solution was
stirred magnetically for 30 hours while refluxing under
anhydrous conditions. After 12 hours of the 30 hour
period had passed, additional lithium aluminum hydride
(3.0 g) was added. Following completion of the reaction,
the suspension was allowed to settle, and the supernatant
was quickly decanted into a flask containing ethyl ace-
tate (10 ml) to destroy excess hydride. Anhydrous diethyl
ether was added to the reflux.flask, the suSpension was
allowed to settle, and the solution was quickly decanted
as above. This extraction procedure was repeated three
times. The ether extracts were combined and extracted
with 10% sodium chloride (2.5 liters) until all the
aluminum hydroxide was removed. Following filtration
through Hyflo Super-Cel,the sodium chloride-aluminum
hydroxide solution was back-extracted with diethyl ether
(two-350 ml portions). The ether extracts were combined

and concentrated under vacuum to a syrup (50 ml) which

17

was distilled under vacuum (0.45 mm Hg); the 81-8300
fraction was collected. Extreme care had to be taken
not to exceed 85°C at this pressure or else an unknown,
highly exothermic reaction occurred. 1,2-3,4-Di-0-iso-
prOpylidene-a-D-fuccpyranose distilled as a colorless
oil which solidified after standing to form a solid
melt; the yield was 23.0 g (58%): melting point 36-3700,
reported 36-3700 (22).

Di-O-isoprOpylidene-a-D-fucOpyranose (23.0 g) was
refluxed for 3 hours in 350 m1 1% sulfuric acid. The
clear solution was neutralized while hot with a warm
barium carbonate suSpension. Following filtration of
the mixed barium salts, the neutral solution was decolor-
ized with Norit and was concentrated under vacuum. The
resulting syrup was dehydrated by three treatments with
absolute ethanol and was crystallized from 95% ethanol;
yield 12.2 g (78%), overall yield 36.2%; m.p. 140-4500,
reported 140-45°C (21). No depression of the melting
point was observed upon addition of authentic D-fucose.
The product co-chromatographed with standard D-fucose in
solvent systems 3, 4, and 5. The melting point of the
phenylhydrazone was 171-72°c; reported 172°C (29, 30).
Further evidence is supplied below where synthetic D-
fucose is converted into four known derivatives of the
free acid, D-fuconic acid. A summary of the reactions

for the synthesis of D-fucose is presented in Figure 1.

18

Figure 1. Reactions in the synthesis of D-fucose

 

       

C\__

0113 CH 3

1,2-3,4-Di-O-isoprOpylidene-6-petoluenesulfonyl-
a-D-galactopyranose

 

1,2-3,4-Di-O-isoprOpylidene-a-D-fucopyranose

on i

 

D-Fucose

19

Prgparation of Potassium D-Fuconate- Anhydrous
D-fucose (2.0 g) was dissolved in 4 ml of distilled water
and was added to 25 ml of absolute methanol. Resublimed
iodine (6.3 8) was dissolved in absolute methanol (70 ml)
in a 500 ml. 3-necked flask fitted with a mechanical
stirrer, a drying tube, and a dropping funnel. The tem-
perature was maintained at 40°C throughout the course of
the reaction with an electronic thermoregulator. The
methanolic D-fucose solution was quickly added to the
methanolic iodine solution prior to the dropwise addi-
tion of methanolic KOH (115 ml, 4%). The resultant
straw-yellow solution was set aside at room temperature
and allowed to crystallize; it was complete within 24
hours. The product crystallizes as the monohydrate (31)
in long, fine needles, m.p. 166-6700. The product was
recrystallized from methanol-water (97:3), m.p. 169-7000,
which remained unchanged upon further crystallization.
The product was chromatographically homogeneous in solvent
systems 1, 2, 3, and 5. D-Fucono-Y-lactone was prepared
by treatment of the potassium salt with Dowex-50W-H+,
lyOphilization, and crystallization from diethyl ether-
ethanol (10:1); m.p. 104-05°C, reported 104-06°C (32).
The amide was prepared through the lactone: m.p. 175-76°C,
reported 180°C (33): further recrystallization did not
affect the melting point of the amide. The benzimidazole

derivative was prepared by the method of Moore and Link

20

(24); m.p. 252-5300, reported 248-4900 (34).

Selection of Mutant Strains— An overnight culture
of the pseudomonad in D-glucose-mineral medium was har-
vested by centrifugation in sterile tubes, and the result-
ing pellet was suSpended in 7 ml mineral medium containing
1.4 mmole of ethyl methanesulfonate (35). After incuba-
tion for 3.5 hours at room temperature, 0.02 ml of the
suSpension was inoculated into 7 ml D-glucose-medium and
incubated for 24 hours at 32°C. The cell mass increased
in absorbency at 600 nm from 0.050 to 0.60 (1.8 cm light
path), as measured with a Coleman Junior Spectrophotometer.
The cells were plated on D-glucose-mineral-agar after
dilution by a factor of 110,000:1 (250 cells per plate).
The resulting colonies were replicated on L-arabinose-
mineral-agar and D-glucose-mineral-agar sequentially.

Mutant strain 5-1-10-1 was selected as a strain
which grew well on D-glucose but failed to grow on
L-arabinose. Further characterization indicated that the
mutant possessed the following phenotype: D-fucose’,
L-arabinose', D-glucose+, D-galactose+, D-xylose+,
Demannose', D-fructose', lactose" which was identical to
the parental strain with the exception that the mutant
was negative on L-arabinose and D—fucose. A revertant
was obtained by incubation of mutant strain 5-1-10-1 with
0.5% D-fucose-mineral medium for 48 hours at 32°C. The
revertant possessed the same phenotype as the parental

strain.

21
Mutant strain 73-1-2 was selected as a strain
which grew on L-arabinose and D-fucose at about 40% the
wild-type rate but grew normally on D-glucose and D-
galactose. This strain was identical to the parental

strain in other reSpects.

RESULTS

A. INVESTIGATIONS OF VARIOUS POSSIBLE ENZYMATIC REACTIONS
INVOLVING D-FUCOSE

The initial step in the metabolic pathway of an m-
deoxy aldose can potentially involve any of several enzy-
matic reactions. These include aldose-ketose isomeriza-
tion, epimerization, phOSphorylation at C-1, reduction at
C-1, or oxidation. The results of a systematic investiga-
tion of these possible reactions are given below.

Isomerization or Epimerization- Using crude extracts
of D-fucose-grown cells, attempts were made to chromato-
graphically demonstrate either the epimerization of D-
fucose to another deoxyaldose or the isomerization of
D-fucose to D-fuculose. Using the procedure outlined in
Table I, it was found that D-fucose was not modified to
a detectable extent (<0.03 umole per hour per mg protein).
D-Fucose was not isomerized to a ketose which would have
yielded an orcinol positive Spot nor was it epimerized to
a deoxyaldose which exhibited partition coefficients
differing from those of D-fucose in the two solvent sys-
tems employed. 6-Deoxy-D-gulose, 6-deoxy-D-talose, 6-
deoxy-Dbglucose, and 6-deoxy-L-idose, the possible epi-
merization products of D-fucose, were not available for

use as standards, so it is not definitely known that their

22

Table I.

23

Chromatographic analysis for epimerization or
isomerization of D—fucose. Cell-free extracts

of D-fucose-grown cells were prepared as des-
cribed in EXperimental Procedures. The reaction
mixture consisted of 150 umoles D-fucose, 10
umoles MgClz, 10 umoles sodium phOSphate buffer
(pH 7.0), and cell extract (15 mg of protein) in
a volume of 1.0 ml. The mixture was incubated
at 30°C, and 50 ul aliquots were withdrawn after
4 hours. The aliquots were deionized with Dowex-
50W-H+, and the eluate was Spotted on Whatman #1
paper. The chromatograms were deve10ped in
solvent systems 3 and 6 and were visualized with
alkaline silver nitrate for reducing compounds
and with an orcinol Spray for ketoses. A control
consisted of a identical reaction mixture with
the exceptions that L-arabinose-grown Aerobacter
aerogenes (known to possess L-arabinose isomerase)
was used as the enzyme source, and L-arabinose
was used as the substrate.

 

 

 

Rf Values
Substrate Enzyme Source System #3 (System #6
D-Fucose * .62 .44
LmArabinose * .54 .40
D-Fucose Pseudomonad .62 .44
ImArabinose A. aerogenes .54, .66** .40, .50**

 

* No enzyme added

**Orcinol positive Spot

24
chromatographic separation from D-fucose would have been
achieved; however, the probability that D-fucose and its
possible isomerization product possessed the same parti-
tion coefficients in the two solvent systems is very
small, Since other hexoses have Rf values ranging from
.36 to .72 in solvent system 3 and from .37 to .55 in
solvent system 6.

,Aerobacter aerogenes is known to possess an
inducible L-arabinose isomerase which converts L-arabinose
to L-ribulose (37). In the chromatographic analysis a
control experiment was run in which an identical reaction
mixture was prepared with the exceptions that L-arabinose-
grown Aerobacter aerogenes was used as the enzyme source
and L—arabinose was used as the substrate. The cell
extract effected the partial conversion of L-arabinose
to a compound which exhibited the same chromatographic
characteristics as ribulose and was orcinol positive
(Table I). Thus, L-arabinose isomerase in Aerobacter
aerogenes was detected using the same conditions employed
for the investigation of possible D-fucose isomerization
in extracts of the pseudomonad; this control provides an
indication of the reliability of the techniques used in
the detection of possible D-fucose isomerization.

A colorimetric assay was also used for the detec-
tion of possible isomerization of D-fucose to D-fuculose
in extracts of the pseudomonad. The basic reaction mix-

ture used in the colorimetric assay was identical to that

25

described in Table I. No ketose formation was detected
((0.004'umole per hour per mg) either in the basic reac-
tion mixture or in reaction mixtures supplemented with 2
mM CoClz, 10 mM ATP, 10 mM NAD+, or 20 mM EDTA. The
results of the chromatographic and colorimetric analyses
indicated that cell-free extracts of D-fucose-grown cells
did not isomerize D-fucose to D-fuculose.

ghggphorylation- The possible phoSphorylation of
D-fucose with ATP by extracts of D-fucose-grown cells was
investigated using the assay described in Experimental
Procedures. No ATP-dependent phOSphorylation ((0.001
umole per hour per mg) could be detected (Table II).

Reduction- The poSsible reduction of D-fucose by
extracts of D-fucose-grown cells was investigated using
the SpectrOphotometric assay described in EXperimental
Procedures. No D-fucose reductase activity ((0.002 umoles
per hour per mg) could be detected (Table III).

Dehydrogenation- In attempts to detect an initial
phOSphorylation reaction for D-fucose, it was observed
that if NAD+ were added to the complete assay in the
presence of D-fucose and cell extract, the endogenous
rate due to NADH oxidase drastically decreased, thereby
suggesting some reduction of NAD+ was occurring. Further
investigation Showed that a dehydrogenase was present
which could oxidize ZD-fucose, as well as a number of
other carbohydrates, to aldonolactones (Table IV). It

was also shown that NADP+ could serve equally well in the

26

Table II. PhoSphorylation of D-fucose. The standard assay
was employed. The pH of all added components was

W

Experiment Rate of NADH Oxidation
(umoles per hour per mg protein)

 

Complete* 0.116
Complete minus D-fucose 0.116

Cogplete plus 1 umole Mn2+,
Co , or EDTA 0.115

NADH oxidase
(cell extract plus NADH) 0.051

Endogenous.ADP
(PEP-coupling system plus cell

extract) 0.000
Complete plus 0.1 mmole ADP .>O.6

Complete minus cell extract 0.000
Complete plus 0.1 umole NAD+ 0.012

 

*See EXperimental.Procedures

27

Table III. Spectrophotometric assay for D-fucose reduc-
tase activity. The standard assay was employed
in which both NADH and NADPH were tested as
electron donors. Cell-free extracts of D-fucose-
grown cells were used as the source of the

 

 

 

enzyme.
EXperiment Specific Activity
(umoles/hr/mg protein)
Complete*, using NADH 0.185
Complete, using NADPH 0.000
NADH oxidase (complete minus D—fucose) 0.183

Complete + 1 umole MgClZ, EDTA, or
2-thioethanol, 0.185

 

*See Experimental Procedures.

28

Table IV. NAD-dependent dehydrogenase activity. Dehydro-
genase activity was measured SpectrOphotometrically
at 340 nm and 25.0°C. The reaction mixture con-
sisted of 5 umoles sodium phOSphate buffer (pH 7.0),
0.1 umole NAD+, 10 umoles aldose, and a limiting
amount of the dehydrogenase in a total volume of
0.15 ml. Crude extract of D-fucose grown cells
was used as the source of the enzyme. The control
to correct for NADH oxidase contained all the
above components with the exceptions that the
aldose was eliminated and NADH was substituted for
NAD+. Those carbohydrates which were negative
((0.002 umoles per hour per mg protein) were:
L-galactose, D-xylose, L-fucose, D-lyxose, and

 

 

 

D-arabinose.
Carbohydrate Specific Activity
(umoles/hr/mg protein)
D-Galactose 2.1
D-Fucose 1.8
LqArabinose 0.8
D-Glucose 0.5
D-Mannose 0.3
6-Iodo-6-deoxy-D-galactose 1.2

D-Galactose .(NADP+ substituted for NAD+) 1-9

29
role as an electron acceptor in the oxidation of galac-
tose and D-fucose. Crude purification involving ammonium
sulfate fractionations and Sephadex G-200 chromatography
resulted in NAD-dependent D-fucose dehydrogenase activity
devoid of NAD-dependent D-glucose and Demannose dehydro-
genase activities. In addition, re-examination of crude
extracts indicated that NADP+ could not serve as an elec-
tron acceptor for D-mannose or D-glucose oxidation. These
data suggested a minimum of two pyridine nucleotide-
linked dehydrogenases. Further examination and purifica-
tion described below revealed that there are only two
dehydrogenases in Duglucose-, D-galactose-, L-arabinose-,
and D-fucose-grown cells; they are, a NAD-dependent
dehydrogenase and a NAD(P)-dependent dehydrogenase.

Using the NAD(P)-dependent dehydrogenase, a direct
correlation was obtained between the amount of D-fucose
oxidized, the amount of lactone formed, and the amount of
NADH produced in the oxidation (Table V). Some hydrolysis
of the lactone had occurred as evidenced by the higher
lactone concentration after lactonization. Since all the
D-fucose oxidized can be accounted for as lactone and
since the previous experiments concerning alternative
metabolic reactions for D-fucose were negative, it was
concluded that D-fucose degradation by any other means
in this microbe except by an initial oxidation was improb-
able.

Table V.

Component

30

Correlation between D-fucose oxidation, NADH pro-
duction, and lactone formation. The reaction mix-
ture consisted of 50.0 umoles pyruvic acid (pH
7.0), 25.0 umoles D-fucose, 50 ug lactic acid
dehydrogenase-pyruvate kinase, 1 umole NAD+, 50
umoles sodium cacodylate buffer (pH 7.0), and

12.5 mg cell extract of D—fucose-grown cells in a
total volume of 1.00 ml. The reaction was incu-
bated at 25°C and the pyruvate concentration
monitored by the method of Friedman and Haugen.
The pyruvate concentration became constant after
one hour, and the lactone and fucose concentra-
tions were then determined. An aliquot was lac-
tonized to convert any free aldonic acids to the
corresponding lactone. The cell extract contained
no NAD-dependent dehydrogenase activity (<0.0002
umoles per hour per mg of protein), as determined
with glucose and NAD+.

Initial umoles Final umoles Change

 

D-Fucose

25.0* 0.00 25.0

Pyruvic Acid 50.0* 25.4 24.6

Lactone
Lactone

00.0 20.8 20.8

(after lactonizing) 00.0 24.1 . 24.1

*Zero time aliquots were assayed.

B. CHARACTERIZATION OF THE NAD(P)-DEPENDENT DEHYDROGENASE
1. Purification

Cell extracts of D-galactose-grown cells were pre-
pared as described in EXperimental Procedures. Except
where indicated otherwise, the fractionation procedures
were carried out at 0—4°C. A summary of the purification
is given in Table VI.

Protamine Sulfate Fractionation- The protein con-
centration of the cell extract was adjusted to 15 mg per
ml by dilution with 0.10 molar sodium phOSphate buffer
(pH 7.0). Ammonium sulfate (13.7 s) was dissolved in 525
m1 cell extract to a final concentration of 0.20 molar,
and then 105 m1 of a 2% protamine sulfate solution in
0.10 molar sodium phoSphate buffer (pH 7.0) was added
with stirring to a final concentration of 0.33%. After
30 minutes the suSpension was centrifuged at 40,000 x g
for 10 minutes, and the resulting precipitate was dis-
carded.

Heat Step- The NAD-dependent dehydrogenase is heat
labile with a half-life of 42 seconds at 55°C whereas the
NAD(P)-dependent enzyme is somewhat more stable. Thus,
the contaminating NAD-dependent dehydrogenase activity
could be removed by a carefully controlled heating proce-
dure. The 40,000 x g supernatant (630 ml) resulting from

the protamine sulfate step was immersed in a 60°C bath

31

32

Table VI. Purification of the NAD(P)-dependent dehydrogen—

 

 

 

 

 

ase.
Fraction Units Specific Fold 280/260
Activity Ratio
Cell extract 101,000 11.8 1.0 0.620
Protamine sulfate 99,500 12.7 1.1 0.895
Heat step 82,500 23.4 1.89 0.960
Ammonium sulfate 55,500 69.5 5.90 1.22
Sephadex G-200 30,100 540. 45.7 1.31
Calcium phosphate gel 14,500* 3340. 276. 1.49

 

*Only a portion of the pooled Sephadex G-200 fractions was
purified by calcium phoSphate gel. This value has been

corrected for the total volume of the pooled Sephadex frac-
tions.

33
and stirred gently until the temperature reached 55°C
(in about 20 minutes). The protein solution was heated
for an additional 2 minutes at 55°C, cooled in an ice-
water bath, and centrifuged at 40,000 x g. The conse-
quent precipitate was discarded.

Ammonium Sulfate Fractionation- The 40,000 x g
supernatant (620 ml) was brought to 40% of saturation by
the addition of 122 g of ammonium sulfate and the result-
ing precipitate was centrifuged down and discarded. The
supernatant was then brought to 60% of saturation with
81.3 g of ammonium sulfate and centrifuged, and the
resulting precipitatewas collected by centrifugation and
dissolved in 70 ml 0.10 molar sodium phOSphate buffer
(pH 7.0). The protein concentration at this stage was
28 mg per ml.

Sephadex G-200 Chromatography— The above 40-60%
fraction was placed on a column (6 x 60 cm) of Sephadex
G-200 equilibrated with 0.01 molar sodium phOSphate buf-
fer (pH 7.0). The enzyme was eluted with the same buffer.
Fractions (15 ml) were collected, and those which contained
the most activity were pooled (90 m1 total).

Calcium.PhoSphate'Gel- The pooled Sephadex G—200
fractions containing the NAD(P)-dependent dehydrogenase
activity were adjusted to pH 6.5 with 0.05 molar HCl and
dialyzed for 24 hours against 0.01 molar sodium cacodylate
buffer (pH 6.5). A portion of the dialyzed protein (5.0
ml) was then treated with 20% v/v of calcium phoSphate

34
gel prepared according to the method of Wood (49). The
gel suSpension was centrifuged for one minute in a
clinical centrifuge and then successively eluted with 1
ml each of 0.01, 0.02, 0.03, 0.04, and 0.05 molar sodium
phOSphate buffer (pH 7.0). Approximately 50% of the
activity eluted in the 0.01-0.02 molar range whereas
100% of the activity was recoverable in the 0.01-0.05

molar range.
2. Properties

pH Optimum- NAD(P)-dependent dehydrogenase activ-
ity as a function of pH was maximal at pH 9.4 in Tris-HCl
and glycine buffers (Figure 2). However, maintenance of
the dehydrogenase at pH 9.4 in an ice-water bath in the
presence Of 0.10 molar Tris-HCl buffer resulted in an
irreversible loss of 10% of the starting activity within
12 minutes. In addition, at pH values )9, reaction rates
were not linear beyond 10 minutes and were observed to
decrease eXponentially. Further, the Km values for L-
arabinose, D-fucose, and D-galactose at pH 9.4 were
found to be 5-14 fold higher than at pH 8.1 (see below).
For these reasons, the standard assay employed a buffer
at pH 8.1 rather than at the pH Optimum.

Substrate Specificity- A large number of carbo-
hydrates were examined as possible substrates for the

enzyme. The carbohydrates which were found to be oxidized

35

Figure 2. pH Optimum of the NAD(P)-dependent
dehydrogenase. The standard assay was employed
except that the pH and the buffer composition were
varied with the dehydrogenase concentration con-
stant. The calcium phOSphate fraction with the
highest SpeCific activity was used. Each buffer
was 0.10 molar. The pH measurements were deter-
mined on duplicate reaction mixtures. The pH of
the reaction mixture did not vary during the 10

minute reaction period.

% RELATIVE ACTIVITY

100

75

50

25

36

Figure 2

 

GLYCINE

_ .

TRIS-HCl

 

 

 

pH

37
by the enZyme were D-fucose, D-abequose (3,6-dideoxy-D-

galactose), D—galactose, 2—deoxy-D-galactose, L-arabinose,
6-iodO-6-deoxy-D-galactose, and L-mannose (Table VII).

The carbohydrates which did not serve as substrates also
did not reduce the rate of oxidation of 33.3 mM D-galac-
tose when added at equimolar concentrations.

Lineweaver-Burk plots for D-galactose, D-fucose,
and L-arabinose are presented in Figures 3, 4, and 5.

The reactions were performed at two pH values: (1) the
standard assay (pH 8.1); (ii) the pH optimum (pH 9.4).

A tabulation of the Km and Vmax values derived from the
plots in Figures 3-5 is presented in Table VIII. All
rates with mixed substrates were non-additive (Table IX),
suggesting that the activity in the Sephadex G-200 frac-
tion represented one enzyme.

Nggleotide Specificity- NAD and NADP served
equally well as cofactors for the dehydrogenase reaction
at pH 8.1. From the Lineweaver-Burk plots shown in
Figures 6 and 7, the Km values for NAD+ andNADP+ were
found to be 14.8 cm and 67.6 uM reSpectively.

Reversal of the Dehydrogenation- The product of
D-fucose oxidation by this enzyme has been identified as
D—fuoonO-Y-lactone (see Product Identification, this
section). The dehydrogenation reaction for similar
enzymes has been found by other investigators to be
reversible when using a high concentration of the reSpec-

tive lactone and a low pH (38-45). The reversibility of

38

Table VII. Substrate Specificity of the NAD(P)-dependent
dehydrogenase. The standard assay was employed
with the exception that the carbohydrate concen-
tration was 33.3 mM.

1

 

 

 

 

 

 

 

 

Carbohydrate Relative Rate %
2-Deoxy-D-galactose 142
D-Galactose 110
D-Fucose 100
LnArabinose 47
3,6-Dideoxy-D-galactose (abequose) 42
6-Iodo-6-deoxy-D-galactose 37
LnMannose* 34

 

Carbohydrates which were not active as substrates
(i.e.. less than 1% the activity on D-fucose) are: galactose-
6-P, glucose-6-P, fructose-6-P, D-xylose, lactose, melezitose,
maltose, cellobiose, DL—glyceraldehyde, D-ribose, D-lyxose,
2-deoxy-D-ribose, L-rhamnose, sucrose, D-arabinose, L-xylose,
N-acetyl-D—glucosamine, D-glucosamine, D-mannitol, D-sorbitol,
myO-inositol, L-sorbose, raffinose, turanose, melibiose,
'EFEhalose, L-fucose, L-arabitol, a-methyl-glucoside, inulin
(8.3 mM), L-galactose, L-glucose, L-mannose, L-fructose, D-
fructose, xylitol, D-arabitol, adonitol, glucuronic acid,
galacturonic acid, D-xylulose, galactaric acid, glucaric acid,
6-deoxy-Dbglucose, D-glucose, 2-deoxy-D-glucose, D-allose,
D-altrose, 2—acetamido-D-allose, 2-acetamidooD-altrose, 6-
deoxy-D-allose, and D-mannose.

*L-Mannose was not saturating at 33.3 mM. The K value was
later determined to be about 100 mM. L-Mannose at 250 mM
was oxidized at the same rate as Dt-galactose at 33.3 mM.

39

Figure 3. Lineweaver-Burk plots relating NAD(P)-
dependent dehydrogenase activity to D-galactose
concentration at two pH values. With the excep-
tion of the differences in buffers, the routine
assay was employed. The calcium phOSphate frac-

tion with the highest Specific activity was used.

40

Figure 3A

mu

Ca: mmoaoflwouo
a

om mm

 

 

Cm ma

 

HI:-

o.m

 

a euoa w a; u mmoaofiaouo mom as A

_

 

 

41

Figure 3B

o.m

Ea: mmoeofimouo

 

 

 

3.0 ma

rfl>

[Com

 

S
a mica w o.m u mmoeofimouo mom a

 

 

42

Figure 4. Lineweaver-Burk plots relating NAD(P)-
dependent dehydrogenase activity to D-fucose con-
centration at two pH values. Other conditions

were the same as described in Figure 3.

43

Azsv mmoosmao
a

oo.H om.o

 

Figure 4A

 

_ _

H.m ma

HI:>

l. CO“

A

m_suoa w o.m u mmooomuo mom as

_

 

 

 

Figure 4B

Azsv mmooaeuo

 

oo.a

3.6

H

om.o

mm.o

 

 

3.0 ma

A

_

14>

.Ilo.m

 

I. a
z mica w 5.6 n amooomuo mom a

r

 

 

45

Figure 5. Lineweaver-Burk plots relating NAD(P)-
dependent dehydrogenase activity to L-arabinose
concentration at two pH values. Other conditions

were the same as described in Figure 3.

46

F igure 5A

£5 mmofimameua

 

 

saw me

 

\

o
b
HI

L 0.3

 

I a
z +1.0." N +4..“ u mmOZHm<m<IA mom M

 

 

47

Figure 5B

Azsv mmozamammtq

 

 

 

s.a ma

vfl>

ll 00m

 

(

E
I . u a mom a
a euoa N N n : mmOZHmmm4_a

 

 

48

Table VIII. A tabulation of the KIn and Vmax values for the
NAD(P)-dependent dehydrogenase. The data were
derived from the data depicted in Figures 3-5.

1 —1-.
r r—r *1 ’ —

ll

pH 8.1 pH 9.4

 

Carbohydrate Km (mM) Relative Vmax Km (mM) Relative Vmax

 

D-Galactose 0.17 110 2.0 72
D-Fucose 0.50 100 6.6 100
ImArabinose 0.14 47 0.52 12

 

Table IX. Effect of mixed substrates on NAD(P)-dependent
dehydrogenase activity. The standard assay was
employed with the exception that the carbohydrate
concentration was 6.7 mM. When both substrates
were mixed, the concentration of each substrate
was the same as that used above. A Sephadex G-200
fraction was used.

W
Specific Activity

 

Substrate (umoles/hour/mg)
D-Galactose 310
D-Fucose 285
IpArabinose 132
D-Galactose + D-Fucose 291
D-Galactose + LnArabinose 215

D-Fucose + LmArabinose 198

 

50

.OOHHOQ soapomOH

Canada OH on» you Hmosaa was open Scapomon 0:9 .uouaaaps mos hpabapom
caeaooan Sherman esp seas soapowwe opmsamosa soaoamo one .za m.mm

was sodpmnpsOoSoo Omopomamoto .psmpmSOO SoapmapsOosoo omdSOmpohsob

on» Spas .oopooaOSH mm .ooahmb mos ScapmeSoosoo +Q¢z on» Sosa Sodpaoowm
0:» Spas Ochoaaao mos mmmmm endpsOa 089 .SOHSOHpSooSOO +Q¢z op hpabdpom

ommSmwonomsoo SSOOSOQOOIAmvmdz wSapmHon SOHQ Masmlaoboososdq .w oaswam

51

EB +642

 

o.mu 0.0m o.mm

 

rd>

In o.m

 

I. a
z mica w m.a : +m<z mom a

_

 

Figure 6

 

 

52

.b shaman SH confluence
we came on» ones msoapdosoo Honpo .SOHpmapSm0SOo +mm<z op mpabapom

commowonomsoo pneumoaoOIAmvm<z wsapmHOH poaa MhsmahobmosoSaq .5 OHSMam

53

Figure 7

Azav +moaz

 

H

m.m

 

 

rfl>

ii o.m

 

a
as muoa w m.b n +madz mom a

 

 

54
the NAD(P)-dependent dehydrogenase was tested with both
6- and y-lactones. The data, shown in Table X, indicate
that the reaction is reversible with y-lactones of
D-fuconic and D-galactonic acids but not with D-glucono-
6-lactone; further, the reaction was not inhibited by
DbgluconO-é-lactone. The b-lactones of D-fuconic and
D-galactonic acids could not be tested because of their
instability (46-48). Since the product was identified
as D—fucono-Y-lactone and since the reversibility of the
reaction was demonstrated with the y-lactones of the
reSpective substrates, it may be inferred that the dehyd-
rogenase is Operative on the furanose form of the carbo-
hydrate.

Stability- The half-life of the dehydrogenase has
been found to be 13 minutes at 55°C. The enzyme, in
Sephadex G-200 fractions, is stable to freezing at -20°C
in 0.01 molar sodium phOSphate buffer (pH 7.0) for periods
up to two months. LyOphilization of pooled Sephadex G-200
fractions containing the enzyme in 0.01 molar sodium phos-
phate buffer (pH 7.0) resulted in a 60-80% loss of activ-
ity.

The effects of various metal ions, thiols, and
thiol group inhibitors are shown in Table XI. The enzyme
is not affected by 1 mM 2-thioethano1 or 1 mM dithio-
threitol. Similarly, 0.5 mM iodoacetate or 0.5 mM pr

chloromercuribenzoate is not inhibitory to enzyme activ-

55

Table X. Reversibility of the NAD(P)-dependent dehydrogenase.
The assay consisted of 0.26 umole NADPH, 10 umoles
of the reSpective lactone, 30 umoles Tris-maleate
buffer (pH 6.5) and a limiting amount of the dehyd-
rogenase in a total volume of 0.15 ml. A Sephadex
G-200 fraction was used. The pH of the reaction
mixture did not vary during the 15 minute reaction
period. The lactone solutions were prepared before
each assay in 0.20 molar Trisémaleate (pH 6.5).

 

Lactone Specific Activity
(umoles/hour/mg)
D-GalactonO-Y-lactone 13.5
D-Fucono-Y-lactone 8.4
D-Glucono-O -lac tone < 0 . 02

D-GalactonO-Y-lactone +
D-glucono-O-lactone 13.4

D-FuconO-Y-lactone +
D-glucono-O-lactone 8.2

 

56

Table XI. Effect of various reagents on NAD(P)-dependent
dehydrogenase activity.
utilized in which the enzyme was added to the

cuvette containing the reagent.

The standard assay was

The Sephadex c-200

fraction with the highest Specific activity was
used. The pH of the reagents was 7.0.

 

 

% of Control

 

Reagent Concentration Activity
p—Chloromercuribenzoate 0.5 mM 100
Iodoacetic acid 0.5 mM 100
2-Thioethanol 1.0 mM 100
Dithiothreitol 1.0 mM 100
EDTA 6.6 mM 100
MgClg 6.6 mM 100
CoC12 6.6 mM 95
(N34)2804 6.6 mM 95
MnClz 6.6 mM 90
NiClz 6.6 mM 70
F6804 6.6 mM 10
ZnC12 6.6 mM 10
CaC12 6.6 mM 5

 

57
ity. There was no observed inhibition by 6.6 mM EDTA or

activation by metal ions, also at 6.6 mM.

Induction- The inducibility of the NAD(P)-dependent
dehydrogenase was tested by growth on various substrates.
The results (Table XII) indicate that Defucose, D-galac-
tose, L-arabinose, and 6-iodo-6-deoxy-D-ga1actose induce
the NAD(P)-dependent dehydrogenase to a level 35-120 fold
over the non-induced level present in nutrient broth-
grown cells. The data suggest that the enzyme is instru-
mental in the metabolism of D-fucose, D-galactose, L-arab-

inose, and 6-iodo-6-deoxy-D-galactose.
3. Product Identification

The enZymatic reaction product resulting from the
NAD(P)-dependent oxidation of D-fucose, D-fucono-y-
lactone, was prepared on a small scale and was identified
by chromatography, derivatization, and by comparison with
authentic, chemically prepared D—fucono-Y-lactone, as
described below. Also, a large scale reaction was
carried out at a high pH to hydrolyze the D—fuconO-Y-
lactone, formed from the oxidation of D-fucose, to the
free acid, which was subsequently crystallized as the
potassium salt.

Enzymatic Preparation Of DeFucono-Y-lactone- The
reaction mixture consisted of 200 umoles D-fucose, 250

umoles pyruvic acid (pH 6.4), 125 ug lactic acid dehydro-

58

Table XII. Induction of the NAD(P)-dependent dehydrogenase
by various growth substrates. The standard
assay was employed. Cell extracts were pre-
pared by sonic disruption in 0.01 molar Bicine
buffer and 0.143 mM 2-thioethanol (pH 7.4).
Protein was estimated by the biuret assay.

 

 

 

 

 

 

 

Growth Substrate Specific Activity
(umoles/hour/mg)
D—Fucose 11.8
LqArabinose 15.7
D-Glucose 0.250
D-Galactose 11.8
Nutrient Broth 0.130

6-Iodo-6-deoxy-D-galactose 4.69

59

genase, 1 umole NAD+, and 2 ml of a Sephadex G-200 frac-
tion (3.5 mg protein). The NAD(P)-dependent dehydrogenase
contained no NAD-dependent dehydrogenase, y-lactonase,
D-fuconate dehydratase, on 2-ketO-3-deoxy-thuconate
aldolase, and was prepared in the same manner described
in the purification procedure with the exception that the
enzyme was isolated from L-arabinose-grown cells. The pH
was maintained at 6.4 by automatic titration with a
Sargent recording pH stat using 0.10 molar NaOH as the
titrant or by incorporating 400 umoles Tris-maleate
buffer (pH 6.4) into the reaction mixture. The reaction
mixture was maintained at 25°C and was Judged to be com-
plete after 2 hours as measured by the complete loss of
reducing sugar. The reaction mixture was then deionized
by passage through a mixed bed resin (50:50, Dowex-50W-H+,
Dowex-i-OH') and chromatographed on Whatman #1 paper in
solvent systems 1, 2, 3, and 5; the lactone co-chromato-
graphed with authentic D-fucono-y-lactone prepared by
lactonization of D-fuconic acid. The yield of the enzy-
matically prepared lactone was 150 umoles (75%). A com-
parison of the prOperties of the enzymatically and chem-
ically prepared lactones is presented in Table XIII.

‘Eggymatic Preparation of Potassium DAFuconate-
The reaction mixture (80 ml) consisted of 20 mmoles of
D-fucose, 25 mmoles of pyruvic acid (pH 8.9), 500 ug
lactic acid dehydrogenase, 1 umole NAD+, and 20 ml Of a

60

Table XIII. A comparison of the physical properties
potassium D-fuconate and D-fucono-y-lactone
prepared by chemical and enzymatic routes.

 

 

 

 

 

   

Preparation
Derivatives* Enzymatic Chemical Lit. Ref.
m.p. K'I-D-Fuconate 169-70°C 169-7000 169-70 °C (31)
m.p. Lactone 105-06°C 104-05°C 105-06°C (32)
m.p. Amide 176-77°C 175-76°C 180°C (33)
m.p. Benzimidazole 251-53°C 252-5300 248-49°C (34)
Chromatography (Whatman #1) Rf Values

Enzyfmati n ChemLcaJ_

Solvent System #1:

D-Fuconic acid .10 .098

D-Fucono-Y-lactone .61 .62
Solvent System #2:

D-Fuconic acid .47 .47

D-Fucono-Y-lactone .61 .61
Solvent System #3:

D-Fuconic acid .70 .68

D-FuconO-v-lactone - -
Solvent System #5:

D-Fuconic acid .26 .25

D-FuconO-y-lactone .56 .55

4

*NO depressions of the melting points were observed upon
mixing.

61
Sephadex G-200 fraction (35 mg prOtein) containing the
NAD(P)-dependent dehydrogenase prepared in the same
manner as described in the purification procedure with
the exception that the enzyme was isolated from L-arab—
inose-grown cells. The enzyme preparation was devoid of
D-fuconate dehydratase, 2-keto-3-deoxy-D-fuconate aldolase,
v-lactonase, and NAD-dependent dehydrogenase activity.
The pH was maintained at 8.9 by automatic titration with
a Sargent recording pH stat, using 0.50 molar KOH as the
titrant. The temperature was maintained at 25°C for 6
hours after which time the reaction was judged to be
complete as measured by the complete loss of reducing
sugar and the constancy of the pyruvate concentration.
The reaction mixture (100 ml) was then passed through a
Diaflow ultrafiltration cell equipped with a 10,000
molecular weight cut-off ultrafiltration membrane to
remove protein. The eluate from the ultrafiltration
cell was reduced in volume under vacuum to 15 m1, applied
to a AG50W-XB-K+ column (1 x 12 cm) and eluted with
deionized water (350 ml). The eluate from the column
was reduced in volume under vacuum to 4.0 ml. The salt-
syrup slurry was then triturated with two portions of
absolute methanol (110 ml, 45°C) and filtered. The fil-
trate was seeded with authentic potassium D-fuconate
(see Experimental Procedures); crystallization commenced

immediately. The long, fine needles characteristic of

62
this salt were collected by suction filtration and washed
with diethyl ether (25 ml); yield 4.21 g. The superna-
tant yielded an additional 0.56 g after repeating the
above procedure. Total yield was 4.67 g (97%) after
recrystallizing from methanol-water (100:2). A compari-
son Of the properties of the enzymatically and chemically
prepared sugar acids is presented in Table XIII. Enzy-
matically prepared potassium D-fuconate was found to be
identical in all reSpects to the chemically prepared
material. .A general outline for the chemical and enzy-
matic preparations for D-fucose, potassium D-fuconate,

and D-fucono-y-lactone is presented in Figure 8.

. 63

Figure 8. A general outline for the preparation of

D-fucose, D-fuconic acid, and D-fuconolactone.

D-GALACTOSE

(Chemical) 1 (Enzymatic)
' D-FUCOSE -—————a-d>-——-—- deionization

(Enzymatic)

‘ ultraf iltra ti on 1

 

 

 

Dowex-50-K+
trituration
l D-FUCONO-Y-LACTON E

 

 

1—p K+-D-FUC,ONATE F

(Chemical)

Dowex-50-H+
lactonization

| an

 

 

 

C. CHARACTERIZATION OF THE NAD-DEPENDENT DEHYDROGENASE
1. Purification

Cell extracts of D-glucose-grown cells were pre-
pared as described in Experimental Procedures. Except
where indicated otherwise, the fractionation procedures
were carried out at 0-4°C. A summary of the purification
is given in Table XIV.

Protamine Sulfate Fractionation- The protein con-
centration was adjusted to 6.5 mg per ml by dilution with
0.10 molar sodium phOSphate buffer (pH 7.0). Ammonium
sulfate (17.2 g) was dissolved in 650 m1 cell extract to
a final concentration of 0.20 molar, and then 130 m1 of
a 2% protamine sulfate solution in 0.10 molar sodium
phOSphate buffer (pH 7.0) was added with stirring to a
final concentration of 0.33%. After 30 minutes the sus-
pension was centrifuged at 40,000 x g for 10 minutes,
and the resulting precipitate was discarded.

Ammonium Sulfate Fractionation- The 40,000 x g
supernatant (800 ml) from the protamine sulfate step was
brought to 30% of saturation by the addition of 110 g
of ammonium.sulfate and centrifuged. The supernatant
was then brought to 40% of saturation with 48.1 g Of
ammonium sulfate. The resulting precipitate was collected
by centrifugation and was dissolved in 53 ml 0.01 molar
sodium phOSphate (pH 7.0). The protein concentration was
20 mg per ml. It might be noted that the 0-30% fraction,

64

65

Table XIV. Purification of NAD-dependent dehydrogenase.

 

Fraction Units Specific Fold 280/260
Activity Ratio
Cell extract 50,140 12.2 1.0 .633
Protamine sulfate 45,130 11.1 1.0 .859
Ammonium sulfate 27,500 27.2 2.24 1.17
Sephadex G-200 14,600 134 11.0 1.25
DEAE-cellulose 11,700* 830 68 1.48
Calcium phOSphate gel 6,750 3980 327 1.58

 

*Only a portion Of the pooled Sephadex G-200 fractions was
purified by DEAE-cellulose. This value and the units
recovered from the calcium phOSphate gel step have been
corrected for the total volume of the pooled Sephadex
fractions.

66
which contains 30-40% of the total soluble protein present
in crude extracts of the pseudomonad, contains the rust-
colored protein which bestows the characteristic pigmen-
tation to the crude extract and the microorganism.

Sephadex G-200 Chromatography- The 30-40% ammonium
sulfate fraction (53 ml) was placed on a column (6 x 60
cm) of Sephadex G-200 equilibrated with 0.01 molar sodium
phOSphate buffer (pH 7.0). The enzyme was eluted with
the same buffer. Fractions (15 ml) were collected, and
those which contained the most activity were pooled
(135 ml total).

DEAR-Cellulose Chromatography- DEAE-cellulose
(Sigma, exchange capacity = 0.9 meq per g) was pretreated
as recommended by Sober 23 a}. (50) and was equilibrated
with 0.02 molar sodium phOSphate buffer (pH 7.0). The
pooled Sephadex G-200 fractions were reduced in volume
to 15 m1 using a Diaflow ultrafiltration cell equipped
with a 10,000 molecular weight cut-Off ultrafiltration
membrane (Amicon Corporation). .A portion of the Sephadex
G—200 concentrate (2 ml) was applied to a DEAR—cellulose
column (3 x 5 cm) which was then washed with 60 m1 of the
above buffer, and then successively eluted with a step-
wise gradient composed Of 60 ml each of 0.10, 0.20, 0.30,
0.40, and 0.80 molar sodium chloride in 0.02 molar sodium
phOSphate buffer (pH 7.0). Of the two dehydrogenases,

only the NAD-dependent dehydrogenase remains active fol-

67

lowing the above DEAE-cellulose fractionation, and thus,
contaminating NAD(P)-dependent dehydrogenase activity was
removed. The NAD-dependent enzyme elutes in the 0.20-
0.30 molar sodium chloride range. The fractions contain-
ing the most activity were pooled (44 ml total).

gaggium.PhOSDhate Gel- The procedures used were
the same as described for the calcium phOSphate purifi-
cation of the NAD(P)-dependent dehydrogenase. The NAD-
dependent dehydrogenase exhibited the same elution char-

acteristics as the NAD(P)-dependent dehydrogenase.
2. PrOperties

pH Optima- NAD-Dependent dehydrogenase activity
as a function Of pH was maximal at pH 8-8.5 in Tris-HCl
buffer and at pH 9-10 in glycine buffer (Figure 9).

Substrate Specificity— A large number of carbo-
hydrates were examined as possible substrates for the
dehydrogenase. The following carbohydrates were oxidized
by the enzyme: D-glucose, 2-deoxy-D-glucose, 6-deoxy-D-
glucose, D-galactose, 2-deoxy-D-ga1actose, D-fucose,
3,6-dideoxy-D-galactose, D-altrose, D-allose, and D-
mannose (Table XV). The carbohydrates which did not
serve as substrates also did not reduce the rate of oxi-
dation of 33.3 mM D-glucose when added at equimolar con-
centrations.

The Lineweaver-Burk plots for eight of the ten

68

.bodaoa composes opSSaa

OH on» mafinso owsm£o no: cap ma one .mmadpxaa Soapomoh OQSOHHQSO

so oosfiahopoo Ones mpSmEopSmmoa we on» use .amaoa oa.o mm: Seaman
comm .boms mm: mpdbapom onHooam amonwas map Spa: Soapomam opmsamosa
azaoamo one .psmpmsoo SoapmapSoosoo ommaowoaemSOO can and: define»
who: Soapamoaaoo Meagan on» one mm on» was» paoowo oohOHaao was amend

bamoampm one .ommaowOHUmsoo SamoaoQCOImdz esp mo madpao mg .m ohswam

Figure 9

69

 

 

GLYC INE

 

    
 

 

 

100

In
(\

IIIAICLO‘V HALLV'IHH %

50

10.0

9-0

8.0

pH

70

Table XV. A tabulation of the Km and Vmax values for the
NAD-dependent dehydrogenase. The data were
derived from the data depicted in Figures 10-17.

 

Carbohydrate Relatige Vmax Km (mM)
D-Glucose 188 0.86
D-Galactose 142 1.6
D-Mannose 124 4.5
2-Deoxy-D-glucose 107 1.6
D-Fucose 100 5.8
2-Deoxy-D-galactose 94 6.3
Dqutrose 79 2.4
Dqulose 47 13.0
6-Deoxy-D-glucose 61 *

3,6-Dideoxy-D-galactose 23 *

 

The carbohydrates which were not active as substrates
(i.e. <1% the activity on Defucose) are: galactose-6-P,
glucose-6-P, fructose-6AP, D-xylose, lactose, melezitose,
maltose, cellobiose, DL_glyceraldehyde, D-ribose, D-lyxose,
2-deoxy-D—ribose, L-rhamnose, sucrose, D-arabinose, L-xylose,
N-acetyl-D-glucosamine, Déglucosamine, D-mannitol, D-sorbitol,
m O-inositol, L-sorbose, 6-iodo-6-deoxy-D—galactose, raffin-
ose, turanose, melibiose, trehalose, L-fucose, L-arabitol,
d-methyl-glucoside, inulin (1.25 umoles), L-galactose, L-
glucose, L-mannose, L-fructose, D-fructose, xylitol, D-
arabitol, adonitol, glucuronic acid, galacturonic acid, D-
xylulose, galactaric acid, glucaric acid, 6-deoxy-D-allose,
2-acetamido-D-allose, and 2-acetamidO-Dbaltrose.

*Not determined.

71
carbohydrates active as substrates are presented in
Figures 10 through 17. A tabulation of the Km and Vmax
values derived from these kinetic plots is also presented
in Table XV. .All rates with mixed substrates were non-
additive thereby suggesting that the activity in the
DEAE-cellulose fraction represented one enzyme (Table
XVI).

Nucleotide Specificity- From the Lineweaver-Burk
plot presented in Figure 18, the Km value forNAD+ was
found to be 7.7 x 10"5 pl. NADP+ was absolutely ineffec-
tive as a cofactor for the dehydrogenase reaction using
concentrations up to 20 mM. In addition, NADP+ (4 mM)
was not a competitive inhibitor of the standard assay.

Reversa;_of the Dehygrggenatiggg The fact that
the isolation of a lactone resulting from D-fucose oxi-
dation by the NAD-dependent dehydrogenase was consis-
tently unsuccessful (see Product Identification, this
section) suggested that the product was the unstable
b-lactone rather than the stable y-lactone. The ring
size of the lactone resulting from the oxidation of
D—galactose, D-glucose, and D-fucose by the NAD-dependent
dehydrogenase was determined by Observing the reversal of
the dehydrogenation reaction using 0- and y-lactones (38-
45). The reduction Of D-fucono-y-lactone, D-galactono-
y-lactone, or D-gluconO-O-lactone in the presence of the

purified enzyme and.NADH was tested, and the results are

72

.psmpmsOo SodpcapSmoSoo ommaomOSOhcoo on»
spas .Oopmoaosd mm .oOdam>.mma Soapmapsoocoo openehsophmo one .mpa
shaped camaooam beeswan on» Spas soapomhm opmsamosa endedmo on» wadms
bomoaaao mm: memos oncosmpm was .sOapmhpsoosoo omoofimum op hpabapom

omdSowoaommoo pneumoaoulmdz maapdfloa poaa masmlaobmosSSaq .oa chamHm

73

Figure 1 0

om.o

Ass. mmoobmao
H

m~.o

 

 

alt:-

Ilo.o.n

 

.a mica w m.m u mmooomno mom as

a

 

 

74

came on» once msoapaosoo H0390

.oa Oaswah SH confluence mm

.aodpmapsoosoo mmoosawlm op mpabapom

omdSowOHoasoo unmoSOQOOIde wSapmHoH SOHQ masmuhmbmosoaaq .Ha shaman

75

Figure 11

Aaav mmooeaono
a

om.H 00.

H om.o

 

 

:1 Com

vd>

o.m

0.:

 

m egos w mgr n mmoopqouo mom ea

_

 

 

76

“OH maswam SH monanomov we

been esp cams mSOHpHOSOO Hospo .SOHpmeaeosoo cmopomHmwtm Op thbapom

cacaowoaomsoo peooaoaconmmz wsHpmHoa pOHa mastHobmososHA

.NH enemas

77

Figure 12

0N.o

Ea: 868338
a

0H.o

 

 

 

m.m

o.m

Ilo.OH

 

a 9.3 w a...“ u mmoeofldouo mom as

 

 

78

.OH oastm SH OonHaomoo we

came on» cams WSOHPHUSOO scape .SOHpmeSoosoo omossmalm op thprom

mnemowoaumSoO SSoUSOQOOImdz maHpmHOH SOHQ MhsmlnobmoSOsHH

.ma shaman

79

Figure 13

om.o

case mmozzazuo
a

 

m~.o

 

 

rfl>

ll o.m

 

.m mica w m.: n mmozzmano mom as

_

 

 

80

.OH oasmam SH OmpHaomoO we came 0:9

one: mSOHpHUSOO Hospo .soHpmeaooaoo omooSHmIQImwooUIN op thPHpom

cmmaowoaomnoo pneumoaOOIde waHpmHea pOHa xhsmlhobOOSOaHH

.sd enemas

81

Figure 14

om.o

Aasv mmoosaouo-wxomoum

 

H
mm.o

 

 

_

rd>

:1 o.m

 

.m nice w b.a n amoopaeuouaxomoum mom ea

a

 

 

82

.oH oastm SH OonHaomoO no Oahu 0:»

one: maoHpHosoo scape .soHpmapsOoaoo omopomHmwlalhwooulm op thbapom

ommSowoaomcoo pneumoaoonmdz wchmHoh SOHa masmlaoboo3osHH

.MH whfime

83

Figure 15

Ca: mmoaofidouonaxomoum

 

om.o

H
mm.o

 

 

(J

   

 

A

o.m

vfl>

m4.

0.0H

a muoa on m6 4. mmoeoflmoaonaxomonm mom Ba

 

 

84

each on» ones wsoHpHoaoo Henna

.oH ohstm SH oopdnomoo mm

.aoapoapSooaoo omOHpHmso op apa>apoc

omosowOHOhcco SSOOSOQOOIQ<Z wsHpmHoa SOHQ RasmuaobmosoSHA .mH ohswdm

85

Figure 16

Azsv mmomgqmro

 

 

 

 

a
o.a ma.o om.o mm.o
_ _ H
J m.
It o.m
S
a
ll mom.
I: 0.0a
.m muoa a cum u amomqurn mom as
O

 

 

86

.OH oaame SH oonHHOmOO we
came on» one: mSoHuHomoo scape .sOHpmhpSooaoo omOHHOIQ op thbdpoo

omoaowoacmsoe SSOOSOQOOIQSZ wSHpoHOH SOHQ masm:ao>mozosHH .mH ohawdm

87

Figure 17

mao.o

£5 83445

 

H
omo.o

36.0

 

 

H

_

 

o.m

ed>

o.OH

a msoa w T? n amended mom as

 

 

88

Table XVI. Effect of mixed substrates on NAD-dependent
dehydrogenase activity. The standard assay was
employed with the exception that the carbohyd-
rate concentration was 33.3 mM. When both sub-
strates were mixed, the concentration of each
substrate was the same as that used above. A
DEAE-cellulose fraction which contained no
NAD(P)-dependent dehydrogenase was used.

 

 

Substrate Specific Activity
(umoles/hour/mg protein)

 

DeGlucose 410
D-Galactose 306
D-Mannose 271
D-Fucose 216
D-Mannose + D-Fucose 243
D-Mannose + D-Glucose 342
D-Mannose + D-Galactose 287
D-Glucose + D-Galactose 350
D-Glucose + D-Fucose 303

D-Fucose + D-Galactose 259

b

89

.OONHHHSS

mos thbdpow oamaooam newsman on» SSH: Soapooam cpmnamosa_aSHOHoo one
.28 m.mm was SOHpmesOoaoo omooSHOIQ .psmpmaoo SoHpmapsOoaoo ommaow
nonemSOO on» and: .oopcOHOSH no .OOHHmb mes SOHpmapsoosoo +942 on» Swap
paoowo OOHOHQSO was mamas oSHpsOa one .soapmausOoaoo +Q¢z op avaproc

ommSOwOHomSOo SambSOQOOIsz wSHpmHoa SOHQ masmlaobmososHH .mH onswam

90

Figure 18

0.0N

Azsv +m<z

 

o.mH

0.0H

 

 

A

A

o.H

III] CON

rfl>

ll 0.:

 

m.muoa w a.a n +o<z mom as

r

 

 

91
presented in Table XVII. It may be seen that the reaction
is reversible only with the 0-lactone which suggests, in
turn, that the enzyme is operative on the pyranose form
of the carbohydrate. The data also indicate that the
y-lactones are not competitive inhibitors of the reverse
reaction under the conditions described in Table XVII.
The O-lactones of D-fuconic and D-galactonic acids could
not be tested because of their instability (46-48).

From the Lineweaver-Burk plot presented in Figure
19, the Km value for D-glucono-b-lactone was found to be
about 10 mM; The half-life for D-glucono-O-lactone under
the conditions Of the assay (pH 6.5) is approximately 10
minutes (51, 52), and for this reason the Km value must
be considered to be an approximation. Using conditions
under which the lactone was saturating (66.6 mM) and vary-
ing the NADH concentration, a Km value for NADH was
ObtainedfiFrom the Lineweaver-Burk plot presented in
Figure 20, the Km value was found to be 39 uM.

Glucose has been identified as the product of the
reduction of D-glucono-O-lactone by trapping with hexo-
kinase and measuring the resulting glucose—6-phOSphate
with g1ucose-6-phOSphate dehydrogenase. The results are
presented in Figure 21.

Anomer Preference- Experiments were carried out
to determine whether both pyranose forms were equally

effective as substrates. The rate of oxidation of an

92

Table XVII. Reversibility of the NAD-dependent dehydro-
genase. The assay consisted of 10 umoles of
the reSpective lactone, 0.26 umole NADH, 30
umoles sodium phOSphate buffer (pH 6.5) and
a limiting amount of the dehydrogenase in a
volume of 0.15 ml. A DEAE-cellulose fraction
was used. The pH of the reaction mixture did
not change during the 15 minute reaction
period. The lactones solutions were prepared
before each assay in 0.10 molar phosphate
buffer (pH 6.5).

 

Lactone Specific Activity

 

(umoles/hr/mg)
D-Glucono-é-lactone 18.6
D-Galactono-Y-lactone (0.02
D-Fucono-Y-lactone (0.02

D-Glucono-O-lactone +
D-galactono-Y-lactone 18.4

D-GluconO-O-lactone +
D-fuconO-Y-lactone 18.4

93

.SOHpomoh one no Chance on» SH hump no: OHO OHSSHHE SOHvaoH

,OSS no ma one .Am.w mav endgamosa asdbom HmHoa 0H.o SH humus comm
oaomon Ochoaoaa one: mSOHpSHOm mmopomH smcnm .OOHHmb was SOHpmeSoosoo
oaopooH 0:» pus» SOHpaOoHO can 89H: HHSN OHnt SH OONHHde pump SSH:
HOOHSSOOH was momma one .SOHpmapSooaoo esopOOHuoloaoousln op thbapOm

omdSOmOHOASSO pneumoaoolmwz wadpmHoa SOHa masmrnobaosoaaq .mH csanm

94

ES mzoaoflueuozooaaouo

03.0 om.o 0N.o 0H.o

 

Figure 19

 

m.m

vfl>

ll 0.0H

 

.2. ~13 w o; u azoaoflueuozoooqouo mom ea

_

 

 

95

.aoHpomoa on» mo

menace can SH hamb no: UHO ohfipNHa SOHpomOH can mo mm was .UeHadb was
Soapmnp:monoo mndz on» was» Soapaoowc map Spa: HH>X OHQSB SH Oopaom
tong pan» on HmOHpSoUH mm: ammmm one .SoapmnpsOosoo mmdz op thprom

omSSOwoaOhSOO pneumoaoonndz mSHpmHoa pOHa masmlnobcozosHA .om «Human

96

Figure 20

00H

Assv mesa

o.mm

0.0m

 

 

   

m.m

rd>

o.m

 

I. a
a mica w a.m n moaz mom a

 

 

97

Figure 21. Identification of D-glucose as the product
of D-glucono-b-lactone reduction by the NAD-dependent
dehydrogenase. The complete assay consisted of 1.25
umoles D-allose; 0.75 mmole ATP. 1.5 mmole MgClZ, 20
umoles sodium cacodylate buffer (pH 6.5), 0.001 umole
NAD+, 0.3 umole NADP+, 10 umoles D-glucono-O-lactone
(freshly prepared in 0.10 molar sodium cacodylate buf-
fer, pH 6.5), excess hexokinase and glucose-6-phOSphate
dehydrogenase, and a limiting amount of the NAD-depen-
dent dehydrogenase. A calcium phOSphate fraction was
used. Dqulose and NAD+ were used as a NADH regenerat-

ing system.

Curve A: Complete
Curve B: Complete minus NADP+
Curve C: Complete minus any one Of the following
components:
NAD-dependent dehydrogenase
hexokinase
glugose-6-phOSphate dehydrogenase
NAD

D-allose
D-gluconO-O-lactone

ABSORBANCE (340 NM)

98

 

 

 

 

 

Figure 21
I I T |
0.150 r-
A
\.
0.100 __
B
./
00050 F'—
NADP+
ADDED \
o I 4L !
5.0 10.0 15.0 20.0

MINUTES

 

99
equilibrium solution of d,B-D-glucose was compared to
the rate of oxidation of freshly prepared a-D-glucose
at 15.600 and pH 7.5. Under the above conditions, in
which the mutarotational step is the rate limiting step
in the oxidation (38, 53), d,B-D-glucose was preferen-
tially Oxidized at a rate 5-fold that of a-D-glucose
(Figure 22). It may be inferred that B-D-glucose is
probably the true substrate for the enzyme.

Stability- The half-life of the NAD-dependent
dehydrogenase at 55°C was found to be about 40 seconds
(Figure 23). The NAD-dependent enzyme is not heat
stable relative to the NAD(P)-dependent enzyme as shown
also in Figure 23. Inactivation profiles of the enzymes
with their reSpective substrates were superimposable
and were linear, suggesting that the activities were
due to single enzymes.

The enzyme, in Sephadex G-200 fractions, is
completely stable to freezing at -20°C in 0.01 molar
sodium phOSphate buffer (pH 7.0) for six months.
LyOphilization of pooled Sephadex G-200 fractions con-
taining the enzyme in 0.02 molar sodium phOSphate buf-
fer (pH 7.0) resulted in a 50% loss of activity.

The effects of various metal ions, thiols, and
thiol group inhibitors are shown in Table XVIII. The
enzyme is not affected by 1 mM 2-thioethanol or 1 mM

reduced glutathione. Similarly, 0.5 mM iodoacetate of

100

Figure 22. Comparison of the rate of oxidation of

d- and a,B-D-glucose by the NAD-dependent dehydrogen-
ase. The standard assay was employed with the excep-
tions that 0.026 umoles D—glucose and 15 umoles Tris-HC1
buffer (pH 7.5) were used. The cuvettes containing
the reaction mixture minus glucose were equilibrated
at 15.6°C before adding the equilibrated d,B-D-
glucose or the freshly prepared a-D-glucose solution.
The reaction was maintained at 15.600. ExPeriments
were identical in all reSpectS with the exception of
the state of mutarotational equilibrium of the D-

glucose solutions.

ABSORBANCE ( 340 NM)

0.750

.0.500

0.250

Figure 22

101

 

 

a.B-D-GLUCOSE

c-D-GLUCOSE

 

 

MINUTES

 

102

Figure 23. Line A: Thermal denaturation of the NAD(P)-
dependent dehydrogenase at 55°C. The protein solution
(1.08 mg per ml) was heated in 0.01 molar sodium phos-
phate buffer (pH 7.0). Aliquots were withdrawn at the
indicated times and assayed with the standard assay.
The thermal denaturation profiles were superimposable
using D-fucose, L-arabinose, and Degalactose with
either NAD or NADP as the electron acceptor. A calcium
phOSphate fraction was used. Line B: Thermal denatur-
ation of the NAD-dependent dehydrogenase at 55°C. A
DEAE-cellulose fraction was used. The protein solu-
tion (1.8 mg per ml) was heated in 0.01 molar sodium
phOSphate (pH 7.0). Aliquots were withdrawn at the
indicated times and assayed with the standard assay.
~The heat inactivation profiles using D-mannose, D—
galactose, and D-fucose, in addition to D-glucose, were

superimposable.

%.ACTIVITY.AFTER HEATING

 

 

103

Figure 23

 

 

 

I l I |

 

0.5 1.0 1.5 2.0
MINUTES AT 55°C

104

Table XVIII. Effect of various reagents on NAD-dependent
dehydrogenase activity. The standard assay
was utilized in which the enzyme was added to
the cuvette containing the reagent. The cal-
cium phOSphate gel fraction with the highest
Specific activity was used. The pH of all
reagents was adjusted to 7.0.

 

 

% of Control

 

Reagent Concentration Activity
p—Chloromercuribenzoate 0.5 mM 100
Iodoacetic acid 0.5 mM 100
2-Thioethanol 1.0 mM 100
Reduced glutathione 1.0 mM 100
EDTA 6.6 mM 100
MgClZ 6.6 mM 100
MnClz 6.6 mM 97
NHhcl, (NH4)ZSOu 6.6 mM 86
CoClz 6.6 mM 51
NiClz 6.6 mM 35
CuSOu . 6.6 mM 16
FeSOu 6.6 mM 16
CaC12 6.6 mM 12

 

105

0.5 mM prchloromercuribenzoate were not inhibitory to
enzyme activity. There was no observed inhibition by

6.6 mM EDTA or activation by metal ions at 6.6 mM.

Induction- The inducibility of the NAD-dependent

dehydrogenase was tested by growth on various substrates.
The data (Table XIX) indicate that D-glucose and D-fucose
induced the dehydrogenase to a level 48-146 fold over

the non-induced level as present in L-arabinose-grown
cells and suggest that the enzyme is instrumental in the

metabolism of D-fucose and D-glucose.”
3. Product Identification

The isolation of a lactone resulting from D-fucose
oxidation by the NAD-dependent dehydrogenase, using the
same techniques employed for the isolation of D-fucono-
y-lactone resulting from D-fucose oxidation by the NAD(P)-
dependent dehydrogenase, was consistently unsuccessful.
This was concluded as being due to the instability of
the b-lactone as compared to the y-lactone (46-48). In
support of this conclusion the reversibility studies indi-
cated that the product of the NAD-dependent dehydrogenase
was a b-lactone rather than a y-lactone. Thus, due to
the instability of D-fuconO-O-lactone, the apparent
product Of D-fucose oxidation by the NAD-dependent
dehydrogenase is D-fuconate which was identified by co-

chromatography and derivatization.

106

 

 

 

Table XIX. Induction of the NAD-dependent dehydrogenase
by growth on various substrates. The cell
extracts were prepared by sonic disruption
in 0.01 molar Bicine buffer and 0.143 mM 2-
thioethanol (pH 7.4) to eliminate NADH oxi-
dase activity. The standard assay was
employed. Protein was estimated by the
biuret assay.

Growth Substrates Specific Activity

(umoles/hr/mg)

D-Fucose 5-55

LnArabinose 0.115

D-Glucose 16.3

DaGalactose 0.690

Nutrient Broth 0.204

6-IodO—6-deoxy-D-galactose 0.330

107

Product Isolation- The reaction mixture consisted
of 100 umoles D-fucose, 125 umoles pyruvic acid (pH 6.5).
400 umoles Tris-maleate buffer (pH 6.5), 50 ug lactic
acid dehydrogenase, 1 umole NAD+, and 2 ml of a DEAE-
cellulose fraction containing the NAD-dependent dehydro-
genase (1.5 mg protein) in a total volume of 2.5 ml.
The NAD-dependent dehydrogenase fraction contained no
NAD(P)-dependent dehydrogenase, D-fuconate dehydratase,
Y-lactonase, or 2-keto-3-deoxy-D-fuconate aldolase. No
lactone was detectable after one hour although the D-fucose
concentration, as measured by reducing sugar, was zero
and the pyruvate concentration had become constant. The
reaction mixture was then deionized by passage through a
Dowex—SOW-H+ column (0.5 x 2 cm) and chromatographed on
paper in solvent systems 1 and 5. The product co-chromato-
graphed with authentic D-fuconate; no D-fucono-y-lactone
was detected. A portion of the Dowex-50W-H+ eluate was
lactonized and chromatographed in solvent systems 1 and
5. .The lactonization product was found to co—chromato-
graph with authentic D-fuconO-Y-lactone in both systems.
It was concluded that D-fuconate was the apparent product
of the NAD-dependent dehydrogenation of D-fucose and
that the b-lactone was probably the immediate product

which Spontaneously hydrolyzed to form D-fuconate.

D. CHARACTERIZATION OF A D-FUCONO-Y-LACTONASE
1. Preliminary EXperiments

D-Fucono-Y-lactone is a stable lactone which
hydrolyzes slowly, even at pH 9 (48). Since this y-
lactone is stable, since D-fuconO-Y-lactone was iden-
tified as the product of D-fucose oxidation by one Of
the dehydrogenases (Section B), and since hydrolysis
of D-fucono-Y-lactone was noted in the initial experi-
ments (Table V), the presence of a lactonase was postu-
lated. The presence of a lactonase was later confirmed
by assaying the enzymatic enhancement of the rate of
proton release in the hydrolysis of various aldonolac-
tones. Evidence is presented in Figure 24 which shows
the participation of a lactonase in the hydrolysis of

D-fuconO-y-lactone.
2. Purification

Cell extracts of D-galactose-grown cells were
prepared as described in EXperimental Procedures.
Except where indicated otherwise, the fractionation
procedures were carried out at 0-4°C. A summary of the
purification is given in Table XX.

PrOtamine Sulfate Fractionation- The protein con-
centration was adjusted to 17.5 mg per ml by dilution

with 0.10 molar sodium phOSphate buffer (pH 7.0).

108

109

Figure 24. The participation of a lactonase in the
hydrolysis of D-fucono-Y-lactone determined by auto-
matic titration using a Sargent recording pH stat.
The assay consisted of cell extract (5 mg of protein)
and 40 umoles D-fucono-Y-lactone in a volume of 1.0
ml. The rate of proton release was monitored in an
unbuffered solution by automatic titration at pH 7.0
with 0.005 M NaOH. The controls consisted of deter-
mining the rate Of proton release in the separate
cell extract and lactone solutions. The cell extracts
were from D—fucose-grown cells.

Curve A: Complete

Curve B: Complete minus fuconO-Y-lactone

Curve C: Complete minus enzyme

110

Figure 24

 

 

 

 

 

 

mooz a mood as

10.0

 

MINUTES

111

Table XX. Purification of the y-lactonase.

W

 

Fraction Units Specific Activity Fold
Cell extract 178,000 20.9 1
Protamine sulfate 142,000 13.9 0.69
Ammonium sulfate 85,500 80.1 4

Sephadex G-200 17,000 333 16

 

112

Ammonium sulfate (13.7 g) was dissolved in 525 ml cell
extract to a final concentration of 0.20 molar, and then
105 ml of a 2% protamine sulfate solution in 0.10 molar
sodium phosphate buffer (pH 7.0) was added with stirring
to a final concentration of 0.33%. After 30 minutes the
suSpension was centrifuged at 40,000 x g for 10 minutes,
and the resulting precipitate was discarded.

Ammonium Sulfate Fractionation- The 40,000 x g
supernatant (620 ml) was brought to 40% of saturation by
the addition of 122 g of ammonium sulfate and the result-
ing precipitate was centrifuged down and discarded. The
supernatant was then brought to 60% of saturation with
81.3 g of ammonium sulfate and centrifuged, and the
resulting precipitate was collected by centrifugation
and dissolved in 35 ml 0.05 molar Bicine buffer (pH 7.5).
The protein concentration at this stage was approximately
42 mg per ml.

Sephadex G-200 Chromatography- The 40-60% ammonium
sulfate fraction was placed on a column (6 x 60 cm) of
Sephadex G-200 equilibrated with 0.05 molar Bicine buffer
(pH 7.4). The enzyme was eluted with the same buffer.
Fractions (15 ml) were collected, and those which con-

tained the most activity were pooled (60 ml total).
3. Properties

pH Optimum- Lactonase activity as a function of
pH was maximal at pH 7.6 in Tris-H01 and sodium phOSphate

113
buffers (Figure 25).

Substrate Specificity- The lactonase catalyzed
the hydrolysis of y-D-lactones but failed to catalyze the
hydrolysis of L-galactono-y-lactone or D-glucono-é-
lactone as indicated in Table XXI.

The data in Table XXI also indicate that the Km
values for D-fucono-Y-lactone, D-galactono-Y—lactone,
and D-ribono-y-lactone are approximately equal to or
less than 20 mM since the lactones were saturating at
40 mM. From the Lineweaver-Burk plot presented in
Figure 26, the Km value for D-galactono-Y-lactone was

determined to be 2.1 x 10"2

M.

Stability- The halfelife at 60°C was determined
to be 3 minutes (Figure 27). The thermal denaturation
profile of the pooled Sephadex G-200 fractions was
linear, suggesting that the observed activity was due
to a single enzyme.

The effects of various metal ions, thiols, and
thiol group inhibitors are shown in Table XXII. The
enzyme is not affected by 1 mM 2-thioethanol, 1 mM
reduced glutathione, or 1 mM sodium sulfide. Similarly,
0.50 mM iododacetate or 0.50 mM pgchloromercuribenzoate
was not inhibitory to enzyme activity. There was no

observed activation or inhibition with 5 mM EDTA or 5 mM

metal ion.

114

.oaonomH can no
mammHoaomn OHnmsaNsOlsOs hon oonooaaoo macs money HHd .Ooms mas anabnnom
OHmnooam nmoswna can nnns SOHnOme oomlo Noomsaow one .ooaaoa aOnnOmoa
can menace hymn no: end moasana acnnomoh can no ma oSn one .moHSnNHa
soHnomoH onoonHasc no oosnaaonoo who: oncoaoaammoa mm 029 .oonmoHOSH mm

.OOnHmb one: Sonnmans®osoo Seaman can one ma can nmnn SOHnaoowo can and:

OONHHHns mm: mmmmw bamusmnm one .ommaonomH can no asaHnao ma .mN Shawna

Figure 25

115

 

 

PHOSPHATE
./

 

 

100

75.—

XIIAICLOV HALLV'IHH %

50

8.5

8.0

7.5

7.0

pH

116

 

o
o
ma.a
em.m
om.m

H.0N
05.0
Hon.o
mmo.o
mmo.o

m.ma
mb.o
mo.m
os.m
ma.s

ozonomHlosoSooanlm
ozonOmHl>loaonomHmolq
economHl>IosonHm|Q
economHI>loSoosmIQ

osonomHi>aosonomHooam

 

ASHE m\mOHoanv
oncm OHnmaaNSm

ASHE m\mOHoanv
ondm oHnmamusmnaoz

ASHE m\mOHoanv
onmm bobaomno

ozonodq

g

monounmnnm HHd

.SOnnmnnaoosoo mnsn nm wannmnsnmm one:
.Ha o.H no oaSHop m SH economH obnnooamcn

onn no mcHoan 0: one Ame my nomanwo HHOO no OoanmsOo means
one .o.u ma nm sonnmannn Oanaonsm an bosOHHon we: OmSOHOH

Sonona no onma one

.omdSonomH osn no annOHnHooam onmhnmnsm .HXN OHnma

117

.ocmOHaao mas annbnnom OHM
Inooam nmoawn: man ans SOHnOmHm oomlo Novenaom one .nsonmsoo Scanmanaooaoo
OhmsonomH can ands .oonSOHOSH mm .oonamn mes acnnmnnaoosoo osonomH can nmsn
aOnnaooNo onn and: OcaOHasO was ammmw bamosmnm one .aonnmhnSeocoo economHi>

:oSonOSHmeQ on annbnnom ommaonomH wananmH nOHa masmnnobmosoan .om Owswnm

118

Figure 26

ON.o

2e: 208317020833:

 

 

 

_

H , .
mHoo OH.o mo.o
_ D _
nl.o.H
0&0 .Ilo.N
b
H
Io.m
I03:

 

a ~13 w Sm 4. 26831688385 won as

_

 

 

119

.mmmmm andcmnm can gnaw crammed 0cm moann

bcnmoHpaH can nm azmhdsnns who: mnosuHHd .Am.m mav oaHOHm anoa

no.0 SH was AHa ace we m.H .H8 H0 SonnSHom anonOHa one .Oom:
mos enhances oaeaooam Sherman or» seas nonsense oomuo Monsanto

one .0000 nm commonomH osn no SoHnmaSnmSoo Hmsaosa .mm onswam

120

 

 

 

 

0N1 NIVNHH IIIAIIDV 24

F igure 2 7
I I I I I I
_ O
I I I I l I
8 5’. E S 8 3

3.0

1.5 2.0 2.5
MINUTES AT 60°C

1.0

0.5

121

Table XXII. Effect of various reagents on lactonase activ-
ity. The standard assay was employed in which
the enzyme was added to the cuvette containing
the reagent. The Sephadex G-200 fraction with
the highest Specific activity was used. The
pH of all reagents was 7.0.

W
% of Control

 

Reagents Concentration Activity
pHChloromercuribenzoate 1 mM 100
‘Iodoacetic acid 1 mM 100
2—Thioethanol 1 mM 100
Reduced glutathione 1 mM 100
Sodium sulfide 1 mM 100
EDTA 5 mM 100
Mg012 5 mM 100
MnC12 5 mM 100
CoClz 5 mM 100
NHucl 5 mM 100
FeSOn 5 mM 103
NiClz 5 mM 103
CuSOu 5 mM 90
CaClZ 5 mM 100
ZnClZ 5 mM 97
Sodium hOSphate buffer 50 mM 100
(pH 7.0? '

Boiled enzyme -- 0

 

122
Induction- The inducibility of the y-lactonase was
tested by growth on various substrates. The data (Table
XXIII) indicate that D-galactose, D-fucose, L-arabinose,
and 6-iodo-6-deoxy-D-galactose induced the enzyme 2—5
fold over the non-induced level present in nutrient broth-

grown cells.
4. Product Identification

The reaction product resulting from y-lactonase
action of D-fuconO-y-lactone, 1,2. D-fuconate, was
identified by chromatography and derivatization.

Preparation of the Reaction Product- The reaction
mixture consisted of 20 umoles of D-fucono-y-lactone and
0.5 mg of a Sephadex G—200 fraction in a total volume of
1.0 ml. The rate of proton release, 1.3. the reaction
rate, was followed by automatic titration at pH 7.0 using
a Sargent recording pH stat and 0.01 molar NaOH. The
reaction was complete within 20 minutes as indicated by
the total absence of lactone and by the negligible rate
Of proton release. The non-enzymatic rate, as measured
in the absence of protein, was less than 0.2 mmole/20
minutes.) The reaction product was deproteinized by
passage through a column of Dowex-50W-H+ (0.5 x 1 cm)
and was Spotted on paper. The chromatograms were developed
in solvent systems 1 and 5. The reaction product was
found to co-chromatograph with authentic D-fuconate in

both solvent systems.

123

Table XXIII. Inducibility of the y-lactonase by growth on
various substrates. The standard assay was
employed. Cell extracts were prepared by
sonic disruption in 0.01 molar Bicine buffer
and 0.143 mM 2-thioethanol (pH 7.4). Protein
was estimated by the biuret assay.

 

 

Growth Substrates Specific Activity
(umoles/hr/mg)
D-Fucose 19.5
LqArabinose 20.6
D-Glucose 4.86
D-Galactose 20.9
Nutrient Broth 7.74

6-Iodo-6-deoxy-D-galactose 24.2

 

_ 124

A portion of the Dowex-50W-H+ eluate was relacton-
ized and chromatOgraphed on paper which was then develOped
in solvent systems 1 and 5. The relactionization product
co-chromatographed with authentic D-fucono-y-lactone. It
was concluded that Dufuconate was the sole reaction
product of y-lactonase action on DefuconO-Y-lactone.

It is felt that, although the lactonase was not
purified beyond Sephadex G-200 chromatography, the enzy—
matic activity observed was due to a single enzyme on the
basis of the following: (i) the thermal denaturation
profile was first order; (11) only one symmetrical peak
was Observed when ammonium sulfate fractions of galactose,
fucose, or arabinose cell extracts were chromatographed
on Sephadex G-200: (iii) a single, symmetrical pH Optimum

was Observed.

E.: CHARACTERIZATION OF D-FUCONATE DEHYDRATASE

The proceeding three sections described the
properties of the enzymes reSponsible for the formation
of Defuconate from D-fucose. The present section is con-
cerned with the purification and properties of a Specific
dehydratase which catalyzes the irreverSible dehydration
of D-fuconate and the consequent formation Of 3,6-dideoxy-
Dgthggg-hexulosonic acid (2-keto-3-deoxy-D-fuconate,
D-KDF). It will also be shown that the enzyme which
dehydrates D-fuconate also dehydrates L-arabonate: L-
arabonate dehydratase has been reported in the literature
but was not characterizable due to its extreme lability
(54-56).

Another dehydratase, D-galactonate dehydratase,
similar to D-fuconate dehydratase, is also present in
this pseudomonad. Dataarejpresented in this section
which distinguishes D—fuconate dehydratase from D-galac-
tonate dehydratase. Due to the structural similarity
between D-galactonate and D-fuconate, D-galactonate
dehydratase might also be eXpeoted to dehydrate D-fuoonate,
however, data will be given in this section and later
sections which SUpport the conclusion that Degalactonate

dehydratase doeSunot dehydrate D-fuconate in vivO.

125

126

1. Preliminary EXperiments

When cell extracts of D-fucose-grown cells were
incubated with D-fuconate, followed by oxidation with
periodate, a 2-thiobarbituric acid (TBA)-positive com-
pound was formed. The physical properties of the TBA
chromogen necessitated that the parent compound possess
the structure of a 2-keto-3-deoxy sugar acid (see
Product Identification, this section) which consequently
required the postulation of a dehydration step. Thin-
layer chromatography indicated that only one TBA-positive
compound accumulated in the presence of D-fuconate and
cell extract; analysis of fractions resulting from
Sephadex G-200 chromatography of the protamine sulfate-
treated cell extract indicated that probably one enzyme
was responsible for D-fuconate dehydration. Preliminary
studies also indicated that the enzyme possessed an abso-
lute requirement for a divalent cation and was activated
by thiols.

Due to the structural similarity between D-galac-
tonate and Dufuconate (6-deoxy-D-galactonate), it was
initially postulated that the enzyme which dehydrated
D-fuconate was the same enzyme which dehydrated D-galac-
tonate, D-galactonate dehydratase (E.C. 4.2.1.6). Evi-
dence to the contrary was obtained in the preliminary
studies which indicated that the two activities were

distinguishable in the following analyses: (i) Sephadex

127
G-200 and DEAE-cellulose chromatography: (ii) pH Optimum;
(iii) thermal denaturation; (iv) sulfhydryl inhibitor
reSponses; and (v) mixed substrate studies.

D-Galactonate dehydratase was later found to
dehydrate Dufuconate; however, the Km value was about
0.12 M. The Specificity of the dehydratases can be
summarized as (i) D-Fuconate dehydratase dehydrates
L-arabonate and, at low concentration; D-fuconate, but
will not dehydrate D-galactonate; (ii) D-Galactonate
dehydratase dehydrates D-galactonate and, at high con-
centrations, D-fuconate, but will not dehydrate L-arabon-
ate. In the following experimenta,L-arabonate was used
as the substrate for Dafuconate dehydratase whenever
D—galactonate dehydratase was present to prevent possible
interference from D-galactonate dehydratase.

The results of the preliminary experiments bas-
ically indicated that an enzyme had been found which
dehydrated D—fuconate at physiological concentrations,
that the enzyme was not D-galactonate dehydratase, and
that the resultant product was a 2-ketO-3-deoxy sugar
acid.

2. Purification

Cell extracts Of L-arabinose-grown cells were pre-
pared as in Experimental Procedures with the exception

that the cells were disrupted in 0.10 molar Bicine buffer

128
(pH 7.4). Except where indicated otherwise, fractiona-
tion procedures were carried out at 0-4°C. A summary of
the purification is given in Table XXIV.

Protamine Sulfate Fractionation- The protein con-
centration was adjusted to 14.5 mg per ml by dilution
with 0.10 molar Bicine buffer (pH 7.4). Ammonium sulfate
(10.0 g) was dissolved in 380 ml cell extract to a final
concentration of 0.20 molar, and then 75 m1 of a 2%
protamine sulfate solution in 0.10 molar Bicine (pH 7.0)
was added with stirring to a final concentration of 0.33%.
After 30 minutes the suSpension was centrifuged at 40,000
x g for 10 minutes, and the resulting precipitate was dis-
carded.

.Ammonium Sulfate Fractionation- The 40,000 x g
supernatant (455 ml) was brought to 40% Of saturation by
the addition Of 89.5 g of ammonium sulfate and the
resulting precipitate was centrifuged down and discarded.
The supernatant was then brought to 60% Of saturation
with 60.0 g of ammonium sulfate and centrifuged, and the
resulting precipitate was collected by centrifugation
and dissolved in 75 ml 0.10 molar Bicine buffer (pH 7.4).
The protein concentration was 21 mg per ml.

Sephadex G-200 Chromatography- The above 40-60%
fraction was placed on a column (6 x 60 cm) of Sephadex
G-200 equilibrated with 0.10 molar Bicine buffer (pH 7.40.
The enzyme was eluted with the same buffer. Fractions

(15 ml) were collected, and those which contained the

129

Table XXIV. Purification of the D-Fuconate Dehydratase*

 

:— 1
I

 

 

 

Fraction Units Specific Fold 280/260
"a. Activity

Cell extract 39,400 4.67 1 . .68

Protamine sulfate 39,000 5.91 1.3 .84

Ammonium sulfate 29,500 18.7 4.0 1.22

Sephadex G-200 7,000 81.1 17.3 1.48

DEAE-cellulose . 2,220 141 30.2 1.60

 

*LqArabonate as substrate

130

most activity were pooled (75 m1 total). A typical
Sephadex G-200 elution profile for D-fuconate and D-galac-
tonate dehydratases is presented in Figure 28.

DEAE-Cellulose Chromatography— DEAE-cellulose
(Sigma, exchange capacity = 0.9 meq per g was pretreated
as recommended by Sober 33,2}. (50) and equilibrated with
0.01 molar sodium phOSphate buffer (pH 7.0). The pooled
Sephadex G-200 fractions were reduced in volume to 5 ml
using a Diaflow ultrafiltration cell equipped with a
10,000 molecular weight cut-off ultrafiltration membrane.
The Sephadex G-200 concentrate was applied to a DEAE-
cellulose column (2 x 6 cm) which was then washed with
60 m1 of the above buffer and then successively eluted
with a stepwise gradient composed of 60 ml each Of 0.05,
0.10, 0.15, 0.20, 0.25, and 0.30 molar sodium chloride
in 0.01 molar phOSphate buffer (pH 7.0). A typical
elution profile for D-fuconate and Degalactonate dehydra-
tase is presented in Figure 29. The DEAEecellulose step
must be performed as quickly as possible due to the
denaturing characteristics of DEAR-cellulose towards

D-fuconate dehydratase.
3. PrOperties

pH Optimum- D-Fuconate dehydratase activity as a
function of pH was maximal at pH 7.1-7.4 in Bicine and

Hepes buffers (Figure 30). The pH Optimum using L-arabonate

131

Figure 28. Fractionation of D-galactonate and
D-fuconate dehydratases on Sephadex G-200. The
details are presented in the text. L-Arabonate
was used as the substrate for D-fuconate dehydra-

tase. Fractions were analyzed by the semicar-

bazide assay.

ABSORBANCE (250 nm)

3000 ""

2.00

132

Figure 28

 

I I I I

o PROTEIN _
. D-GALACTONATE DEHYDRATASE

(3 D-FUCONATE DEHYDRATASE

 

 

 

 

 

480- 600 720 840
“ ML ELUATE

 

 

1.50

1.00

c0.50

PROTEEN (MG/ML)

133

.000: no: mcnom oneMud you mmmmm evanonnmo

:HaOm one .SnonOHQ .010 “ommnmapmmou mnmaonomHmmIQ .OIO ”omen

announce onmsooSmum .010 .ommnmapmsoc ond:oo¢m:n Hon onmnanSm

mfiu. mm Ummfi WGS QpGCODGHGIJ—U

.nwon osn an monsomoaa one mHHmnoQ

.omOHSHHOOIM<m0 so mcmmnmapmnmu no SOnnmsoHnodam .mm Tasman

134

Figure 29

(TN/0W) NIHLOHd

 

 

 

. Asa m0 mmmzoz zoaeosmm

 

 

 

 

 

 

 

 

 

 

.30: on
mm.0|l
92338
La H042
10m.0
\ as o om.o::
a...
W00

 

 

 

(mu 092) sostuossv

135

.UOnaOa aoHnomoa onssna on can no omhdoo can
wanaap cwsmco no: cap ma one .mOHSnNna sOnnocoa onSOHHaSU so downs
:Honou ones mn:oaOSSmmos ma one .amHoa om.0 was Seaman £00m .voms

mm: mngnnom OHMHooam newsman can snnz soHnOmam omOHSHHOOIm¢m0 was
.nsmnmcoo soHnmanaoosoo ommnmapanou can ands .conSOHUSH mm .coHHmb
one: SOHnnmoaaoo Seaman can use ma can nmsn namowo doaOHaao new

names vampsmnm one .ommnmhomSOU onmsoosnln no asaHnao ma .om oaswnm

136

Figure 30'

mm

 

 

on me. one
_ _ .
O
\ \
C
aozéamonnam . . . .
oz maps 0
$83
o
o
\ 0
Saga 0
a o o o. o H

 

 

mm

0m

mm

00H

XIIIAIIOV EAIIIIVIEH %

137
as substrate was superimposable with the pH Optimum using
D-fuconate as substrate. The pH Optimum in the absence
Of 2-thioethanol was also 7.1-7.4. but was broader than
the pH profile Obtained in the presence of 2-thioethanol.

Substrate Specificity— A relatively large number
of carbohydrates were examined as possible substrates
for the dehydratase. Of the following carbohydrates
only D-fuconate and L—arabonate were found to serve as
substrates at the 60 mM level: D-fuconate, L-arabonate,
D—arabonate, Iafuconate,. 6-iodO-6-deoxy-Degalactonate,
D-galactonate, D-ribonate, D-xylonate, D-lyxonate, D-
mannonate, L-rhamnonate, D-gluconate, Lpgalactonate, L-
gluconate, D-galactarate, D-glucarate, cellobionate, D-
glucuronate, D-galacturonate, 3-deoxy-Daxylg-hexonate,
N-acetyl-D-glucosamine, and glucpsamine.

From the LineweaverbBurk plots presented in
Figures 31 and 32 the Km values for D-fuconate and L-
arabonate were found to be 4.0 and 4.3 mM reSpectively.
The relative vmax values were 0.47 and 1.00 reSpectively.
2-Thioethanol had no effect on the Km value for D-fuconate
but did increase the Vmax Value, as previously noted.

Metalqun Activators- DeFuconate dehydratase
possesses an absolute requirement for a divalent cation

as Shown in Table XXV. Mg2+ was found to be the most

2+ 2+

effective activator with Mn2+, Fe2+, Cu , C02+, and Ni

being partially effective. The effect Of Mg2+ concentra-

138

.UONHHHnS

no: annbnnom oanooam nmoawns can gnaw SOHnomhn omOHSHHOOIH¢mQ
one .namnmaoo soHnmansOosoo ommnmachnou osn and: .0oanHOsH mm
.Uonamb mm: SoHnmanswoaoo onmao05wua osn nmsn namowo SOHOHaao
was memos cascadnm one .sOnnmanSOosoo onmaooSnaa on hngnnom

ommnmhpmsop onmaoosnum wanmHoH nOHa masmnaonmosoan .Hm onstm

139

Figure 31

0m.0

:05 maszoooano

 

H
3.0

 

 

H024m9m0HmBIN

HOdeBmOHmBIN
abomaHz

         

NEH:

/ .1...

«we

Il.m.m

Il.0.0H

 

a 0.3 w 0.0 u maszooamuo mom as

 

 

 

140

.vmnaaaps

was mpabapom camaoonm pmonwdn 0:» Spa: noapomhm mmOHzHHmolmde

mse .pCMpmcoo sodpmnpcmonoo ommpdnumsmu as» and: .vmpMoHuna mm
.uoanmb was :oapprCmoCOO mpmzondnmlq map pus» unmowm ummoanam

was mammm Uhmvcmpm msa .:o«pmnpnmo:oo mpmzopaHMIA op mpabapom

ommpdnohsmd mpmnoodmln wndpdamn poaa MysmunmbmmSmsdq .Nm mhfiwdm

141

2%: me «zomgauq

oaoo

H

mc.o

 

 

Figure 32

   

oo.N

 

a maoa u m.: n @4283an mom am

_

 

142

Table XXV. Effect of metal ions on Dbfuconate dehydratase.
The standard assay was used with the exceptions
that 2 umcles of the metal salt and 0.5 umole
EDTA (pH 7.4) were added. The DEAE-cellulose
fraction with the highest specific activity was
used. LqArabonate was used as the substrate.
The pH of all reagents was 7.4.

 

_, l

VJJ

 

Coupound _ . Relative Rate
MgClZ. MgSOu 100%
MgClz + Nanl 88
MnClz 85
Peso“ 37
CuSOu 29
00012 18
NiClz

Ca012

ZnCl2

MgClz + boiled enzyme

0000(1)

No ion added

143

tion upon D-fuconate dehydratase activity was studied
(Figure 33); the Optimum Mg2+ concentration was deter-
mined to be about 20 mM. The effect of concentration of
other ions on dehydratase activity was not tested.
Activity with 3.3 mM EDTA in the absence of added metal
ions was nil. Ammonium ion was 22% inhibitory at 3.3 mM.

Sulfhydryl Activation and Inhibit ion- Sulfhydryl
compounds activated the dehydratase from 30-60% depend-
ing upon the stage of purification (Table XXVI).
prChloromercuribenzoate (50 uM) produced 88% inhibition
although iodoacetate was without effect at 2 mM.

Beversibilityof the Dehydration- Incubation of
0.6 to 6 mM Z-keto-3—deoxy-D-fuConate (KDF) or 2-keto-3-
deoxy-L-arabonate (KDA) in the otherwise standard assay,
using Sephadex G-200 fractions containing the dehydratase
but devoid of KDF aldolase, resulted in no detectable
loss of TBA-positive material ((0.004 umole per hour per
mg protein). In addition, enzymatic preparations of KDF
or KDA were consistently obtained in yields amounting to
100% of the starting material. All attempts to demon-
strate the reversibility of dehydration were repeatedly
unsuccessful.

In addition to establishment of the equilibrium
position of the dehydratase reaction, it was also deter-
mined by the_same eXperiments that the dehydratase prepara-

tion did not cleave, or remove by other means, KDF or KDA.

14h

.opmnpmnsm on» me down was
opmcoosmua .uoms mos ommpmaomSoo cosmopomamwun no oaoboo :oapomhm

omoHSHHooum¢mQ « .oopoodosa mm doaamb mos soapsapsoonoo + ms on»

N
pas» noapmoowo on» spa: oomoaaao mos momma uaoosmpm one .mpabapoo

ommpmaomsoo oposoosMIQ so :oapmapzoosOo + wz Mo poommm .mm onwam

N

Figure 33

145

 

 

 

 

100

75—

25 4L—

1
O
m

IlIAIiOV HAILVTHH %

[M82+]

LOG

Table XXVI o

146

Sulfhydryl activation and inhibition of D-
fuconate dehydratase. The standard assay

was employed with the exception of the varia-
tion in the thiol or the thiol inhibitor
added. The DEAE-cellulose fraction with the
highest Specific activity was used. The pH
of all reagents was 7.#. LqArabonate was
used as the substrate. '

 

 

Compound

Concentration Relative Rate

 

No thiol or thiol
inhibitor added - 10Q%

Z-Thioethanol

2.85 mM 130

Glutathione (reduced) 2.68 mM 1&5

Dithiothreitol 2.68 mM 142

Iodoacetate

2.23 mM 104

p-Chloromercuribenzoate 0.05 mM 12

 

_ 11+?

Stability- D-Fuconate dehydratase is not heat-stable
relative to D-galactonate dehydratase as shown in Figure
34. The half-life at 53°C of the former, in a DEAE-
cellulose fraction, is 80 seconds compared to over 1 hour
for the latter. DaFuconate dehydratase, present in
Sephadex G-200 fractions, loses 50% activity upon freez-
ing overnight in 0.10 molar Bicine buffer (pH 7.0) and is
equally unstable towards 1y0philization in the same buffer
system. The enzyme, however, remains completely active
after 2h hours at room temperature in 0.10 molar Bicine
buffer (pH 7.“). D-Fuconate dehydratase, present in
Sephadex G-200 fractions, was found to be inactivated by
various buffers; the buffers (0.10 M) and the relative
activities of the dehydratase at pH 7.5 are: Bicine
(100%), sodium barbitol (60%), Hepes (03%), glycylglycine
(01%), phosphate (30%), Tris-H01 (15%), Tee (9%), and
Tricine (3%). _

Induction- The inducibility of D-fuconate dehydra-
tase was tested by growth on various substrates. The
data (Table XXVII) indicate that D-fuoose, D-galactose,
and L-arabinose induced the enzyme 13—fold over the non-
induced level as present in nutrient broth—grown cells.
D-Galaotose apparently acts as a gratuitous inducer of
D-fuconate dehydratase since the enzyme is inoperative

on D-galaotonate.

148

Figure 34. Line A: Thermal denaturation of D-
fuconate dehydratase at 53°C. A DEAE-cellulose
fraction was used. The protein solution (3.1 mg
per ml) was heated in 0.20 molar Bicine buffer
(pH 7.0). Aliquots were withdrawn at the indi-
cated times and were analyzed with the standard
assay using both L-arabonate and D-fuconate, with
no added thiol. Line B: Thermal denaturation of
D-galactonate dehydratase at 53°C. A Sephadex
G-200 fraction was used. The protein solution
(2.1 mg per ml) was heated in 0.20 molar Bicine
buffer (pH 7.0). Aliquots were withdrawn at the
indicated times and were analyzed with the standard

assay.

PERCENT ACTIVITY REMAIN ING

90

80

7O

60

50

30

20

 

149

 

 

 

 

 

MINUTES AT 53°C

Figure 30
f + + 4* 4
B /
O D-FUCONATE
_ . L-ABABONA‘I‘E
‘ D-GALACTONATE
u.
}_.
I
O
—— O
1.0 2.0 3.0 4.0

 

150

Table XXVII. Inducibility of D-fuconate dehydratase by
growth on various substrates. The standard
assay was employed with L-arabonate as the
substrate. Cell extracts were prepared by
sonic disruption in 0.01 molar Bicine buf-
fer and 2-thioethanol (pH 7.4). Protein was
estimated by the biuret assay.

__4__L

 

Growth Substrates Specific Activity
_. (umoles/hr/mg)

D-Fuco se 1+. 67

LnArabinose 0.27

D-Glucose 0.562

D-Galactose 4.25

Nutrient Broth 0.329

6-Iodop6-deoxy-D-ga1actose 0.570

151
0. Identification of the Dehydration Product

Among other things, the initial eXperiments indi-
cated that when cell extracts of D-galactose-, D-fucose-
or L-arabinose-grown cells were incubated with D-fuconate,
a TBA-positive product accumulated in amounts approaching
theoretical limits. The product was proposed to be a
2-keto-3-deoxy aldonic acid solely on the basis of the
physical pr0perties of the TBA chromogen. .This section
will concern the enzymatic synthesis, the structural
elucidation, and the chemical synthesis of the dehydra-
tion product...

Enzymatic Synthesis of the Dehydration Product- To
obtain sufficient material for characterization, the enzy-
matic preparation of the dehydration product was carried
out on a large scale (0.5 mmole). The reaction mixture
contained 50 umoles MgClz, 500 umoles potassium D-fuconate,
and 20 mg of a Sephadex G—200 fraction containing the
dehydratase,in a total volume of 10 ml. The pH was main-
tained at 7.“ with a Sargent recording pH stat using 0.01
molar sodium hydroxide as the titrant. Aliquots were
withdrawn at various intervals and were assayed with
TBA for 2-keto-3-deoxy aldonic acid formation. After
the reaction was complete (2 hours), the reaction mixture
was deproteinized by passage through a column of Dowex-
sow-3+ (0.5 x 35 cm). The column was washed free of non-

cationic material with deionized water (750 m1), and the

152

eluate was passed through an AG1-X8 formate column (200-
400 mesh, 0.5 x 35 cm) which was also washed with deion-
ized water (750 ml). The formate column was eluted with
a linear gradient of 0-0.08 molar formic acid using a
Technicon Autograd gradient device. Fractions (1“ ml)
were collected and analyzed for KDF by the TBA assay
(Figure 35). The pH of the eluate from the formate
column was also monitored using a flow-through cell
equipped with a micro-electrode. Fractions containing
KDF were pooled and reduced under vacuum to a syrup to
remove formic acid (bath temperature, 30°C). The free
acid was converted to the potassium salt with Dowex-SOW-K+
and lyOphilized to form a semicrystalline solid. The
yield was 490 umoles (98%). The solid was dissolved in
water, and the pH was adjusted to 7.0 with KOH. The
dehydration product was shown to be chromatographically
homogeneous on Whatman #1 paper in solvent systems 2 and
6 and possessed Rf values of .29 and .78 respectively.

Absorption Spectrum of the TBA Chromggggg 2-Keto-
3-deoxy aldonic acids and Z-deoxy sugars are cleaved by
periodate to form B-formyl pyruvate and malondialdehyde,
respectively. It was observed by Weissbach and Hurwitz
(17) that a compound resulting from periodate oxidation
of 2-keto-3-deoxy aldonic acids reacted with TBA to form
an intense pink color which absorbed maximally at 549-551

nm which they concluded to be a TBA-B-formyl pyruvate com-

 

153

spas oohdmmm one: Ada day mzoapooam one

mo.0Io a Spa? dopaflo mm: CESHOO 0:8

.Uaom ouhfipapaonoasplm
.pzoaomnw odes oaahou HdHoa

.posoohn soapmadhnod comma

loan madmoapmamuso 02» mo msndawopmaoaso opmahom Hawosoo .mm onsmdm

15h

mmmZDZ ZOHBUam

 

 

 

mm on mm
. —‘ — . CI..!J.¢: ‘ C
/ o
/ ,
o/I.
/
/
/
O //
O / /
ll 0 .
o: m .//
e \ //
O
/
O o
. o
o. o
o

5|8a
3
e
u
m
FL _ _ _

 

 

.H

(mu 199) HONVHHOSHV

155
plex. Malondialdehyde also reacts to form a TBA chromogen;
however, the Amax is 532 nm (6, 27, 57, 58). The malondi-
aldehyde-TBA chromogen is alkali stable, in distinct con-
trast to the B-formyl pyruvate-TBA chromogen which is
alkali unstable (14, 27). Both complexes exhibit shifts
to longer wavelengths in the presence of alkali, and no
other compounds tested except malondialdehyde and B-formyl
pyruvate yielded chromogens in the 530-550 nm range.

The absorption spectrum of the TBA chromogen
resulting from the periodate oxidation of the dehydration
product is presented in Figure 36. The chromogen possessed
a Amax at 551 nm and was alkali unstable. The results
indicate that the dehydration product possessed the
structure of a 2-keto-3-deoxy aldonic acid. No over-oxi-
dation of B-formyl pyruvate by periodate, which would
result in concomitant loss of color formation (17, 59),
was observed under the conditions ofthe assay (Figure 37).

Ceric Sulfate Decarboxylatigng To substantiate the
probable identity of the dehydration product as a 2-keto-
3-deoxy aldonic acid, a ceric sulfate decarboxylation was
performed. The dehydration product was first reduced with
NaBHn and then oxidized by ceric sulfate to form, theoret-
ically, a 2,5-dideoxy-pentose (Table XXVIII). The NaBHu
reduction products were TBA-negative since B-formyl lac-
tate resulting from periodate oxidation of the mixed

metasaccharinio acids does not react with TBA to form a

156

.madbhopza muzzaa or» no
seam» mos azhpomam obdmmooozm node no» no ma azhpooam pmoahonnz
one .mos spas m.aa on mg mg» no peoapmsnem mndxoaaoe mampnopea

cad» msoaamb pm pozooaa zoapmaU556o on» wo soapsdawo cumuoahoa

on» aohm wsapasmoa coonOHSO «me on» go mnpoonm soapmnomnd. .m
.pozcona soapmaomsoo on» no soapmodwo opmooahon on» scam

wzdpazmoa somoaohso «me on» no auhpoonm soapnaomnd .4 .mm ohswdm

Figure 36

157

 

B.

l

 

 

-—4.6

'-'102

10
1‘12

0.8
—-O.Ll'

 

 

A.

 

560

540

560

550

540

530

 

1.00 .—

0080 L—

.6
0.40

HONVHHOSEV

nm

 

158

.hmmmm on» yo msoapaozoo esp

Moons Noaaaoo emauopdbfiama Haahomnm on» go mpaflanmpm

.R 0.8me

159

oma

ONH

om

ooom ad mmBDZHz

0:

om

 

 

Figure 37

_

_

omN.o

.oom.o

 

 

(mu 19S) zonvauossv

160

oanmpm «Heads

as

an mmm a x
nowoaOHSO

dmaloUhnoonsonz

ooaaom
zomoaonno dma oz

 

 

dma «m9
monsooaoaosoama opmpoma Hmahomum
omo
.
omo Nmo oohsouampood
_ .
mmo New mmo Amovmw mmo
omo omo omo moon omo
QOHmm .monm
ewOpzon modem
nooHSpIQImNooodolm.N oazahmsoodmmpoa ooxaa
mmo
mmo moaeam
.
malolm N mlolom
. . @038 N . .IIIII
mlwlom N v\ mw mmmz
. mmw oo Amovmm
omo mooo

mam ho soapmahxonamooe opmMHsm canoe

manflmfi 39:...

was
a: amm 4

sowoaonso «ma

Iopmbbamm Haahounm

dma

mumbzhhn Haahomnu

.omw
«mo
_
one

.
mooo
e

QOHmm

 

mmo
mouwum
i-..
mmo
cum
meow

mam

.HHH>xx mange

161
visible chromogen (57). However, after ceric sulfate
decarboxylation of the mixed metasaccharinio acids, a
TBA-positive compound was formed which absorbed maximally
at 532 nm and was alkali stable and was thus character-
istic of a 2-deoxy aldose (Figure 38). The data substan-
tiated the presence in the dehydration product of keto
group at C-2 and the methylene group at C-3. Since
2,5-dideoxy pentose standards were not available, the
eerie sulfate oxidation product could not be rigorously
identified; however, the results strongly support the
contention that the dehydration product has the structure
of a 2-keto-3-deoxy aldonic acid.

Quantitative recovery of 002 by the procedure of
Meister (13) indicated that the dehydration product
yielded 0.996 mole 002 per mole KDF.

d-Keto Acid Derivatives- The dehydration product
reacts with 3-methyl-2-benzothiazolinone hydrazone to
form an azine which absorbs maximally at 325 nm, charac-
teristic of d-keto acids (20). The Spectrum is presented
in Figure 39. The dehydration product also forms a semi-
carbazone which absorbs maximally at 250 nm (Figure 40),
also characteristic of a-keto acids (19).

grPhenylenediamine will condense with 2-keto
hexonic acids to form 2-hydroxyquinoxalines (60). The
absorption Spectra of quinoxalines are highly Specific,
the 330/360 nm absorbancy ratio being 1.51 i 0.07 for

162

Figure 38. Absorption spectra of the TBA chromogen
resulting from the periodate oxidation of the sodium
borohydride-reduced and ceric sulfate-oxidized dehy-
dration product. The alkaline stability of the
malondialdehyde-TBA chromogen is indicated; the pH
11.2 spectrum was not changed after 2 hours. The

pH of the TBA assay is 2.0 which was adjusted to pH
11.2 with 0.5 molar KOH.

 

ABSORBANCE

163

 

 

 

 

 

Figure 38
l' - l I
1.500 —— PH 2'0
1.000 h-I—
0.500
I , l l I
525 535 545 555

160

Figure 39. Absorption spectrum of the 3-methyl-2-
benzothiazolinone hydrazone azine of the dehydration

product.

 

ABSOHBANCE

0.60

0.50

0.40.

0.30

0.20

0.1Q

165

Figure 39

 

| T T

AZINE DERIVATIVE OFeKDF WITH
3-METHYL-2-BENZOTHIAZOLINONE
___HYDRAZONE

 

l l 1

 

305 315 325

1 nm

335

345

 

166

Figure 40. .Absorption Spectrum of the semicarbazone
of the dehydration product.

‘ABSOBBANCE

1.00

1.30

1.20

1.10

1.00

0.90

0.80

167

Figure .40

 

 

KDF SEMICARBAZONE

 

 

168
2-keto hexonic acid derivatives. The quinoxaline Spectrum
for the dehydration product (Figure 41) possesses a 330/
360 nm ratio of 1.51: repeated analyses yielded the
average value, 1.51 t 0.02.

The data obtained from these three derivatives are
consistent with the previously obtained data which indi-
cated that the dehydration product possessed the structure
of a 2-keto-3-deoxy aldonic acid.

Egriodate Oxidation- It cannot be assumed that the
configuration of the hydroxyl groups remaining in the 2-
keto-3-deoxy aldonic acid has not been changed, deSpite
the absence of a precedent of such a reaction. The
assignment of the configuration of the hydroxyl groups at
C-4 and C-5 as thggg was possible, however, as a result
of studies of the different rates of release of B-formyl
pyruvate in the periodate oxidation of 3-deoxy-D-thgggg
and 3-deoxy-D-eythro-hexulosonic acids. The rate of
‘periodate oxidation of 3-deoxy hexulosonic acids is con-
tigent upon the configuration of the hydroxyl groups on
C-4 and C-5, and it has been found that the erythro con-
figuration is oxidized at a more rapid rate than the
thrgg configuration (14, 27, 61-65). The congruence of
the reaction rates for the periodate oxidations of the
dehydration product and 3-deoxy-D-thggg-hexulosonic acid
(2-keto-3-deoxy-D-galactonate) is indicated in Figure 42;
the results of these eXperiments clearly indicate that

169

Figure 41. .Absorption Spectrum of the quinoxaline
derivative of the dehydration product.

 

 

ABSORBANCE

170

Figure 41

 

1.0

0.9_—

0.8__.

0.7__

0.6__

0.5...

0.4-—-

 

QUINOXALINE DERIVATIVE OF
THE DEHYDBATION PRODUCT

 

 

325

340 355 370 385

l nm

400

 

171

 

.oaoo 0HsomoHSNoSIoHSpNaolnlmNoovum one .uaom canomoazwoslooHSp

Inshxooonm .posooaa zodpdaohnoo on» mo mmodpmoawo opmvodaon esp SH
. N: ohswam

ondbfihmm Hmahomnm mo ommoaoa no money on» no comandgaoo.d

172

comm 94 8992:
3 om

 

 

Figure 42

. . _

Amadzooemunuwxommnmnoamaumv 4
mas/6 monseamaommmeamuwxomfiono . m

 

Amazesofleououwxomaumnoemaumv o
medzomoqafimuommmadnaxomnum

 

Amaezoopqoupum Mmmmamnoemeumv 0
342883811 ommawmmuouaammam

 

 

 

om

ooa

NOIELE'IENOO %

173
the configuration of the hydroxyl groups at C-4 and C-5
has remained intact through dehydration at C-2 and C-3.

Quantitative determination of periodate indicated
that 0.89 mole periodate was consumed per mole a-keto
acid. .Acetaldehyde, the co-product of the periodate
oxidation, was also detected but in amounts of 0.2-0.3
mole per mole c-keto acid. Acetaldehyde, however, is
the only known compound which interferes with the TBA
assay (66) and probably condenses via an acid-catalyzed
aldol condensation with the enol formed by the rearrange-
ment of the active methylene carbanion of B-formyl
pyruvate.

Cleavage by 2-Ketoe3ydeoxy:D-fuconate- The dehydra-
tion product is cleaved by KDF aldolase in an endpoint
assay to yield 0.98 mole pyruvate per mole c-keto acid
and 0.96 mole lactaldehyde per mole a—keto acid. These
data provide additional structural proof of the dehydra-
tion product.

The previous data on the dehydration product are
summarized on Table XXIX. It has been concluded that the
structure of the dehydration product is 3,6-dideoxy-D—
thrgg-hexulosonic acid (KDF). AS a final proof of
structure, the chemical synthesis of KDF was undertaken

and is presented in the next subsection.

 

174

Kg

 

a: nmm an memuwam 4
ozauo ozonmndhn
ozofiddoNoaSpounonINIthpozlm
No.0 « an.H no.0 « am.a Amnaamwonaeeauonessumv
cause a: omm\omm .I
enaadaoosloSosmTo
Mme ,
a: 0mm 8: 0mm « onowmnnmo«aom
oanmpm oanmpm hpdaandpm aHmMHm
was
a: «mm a: «mm 4
soapddawo mummazm oaaoo
use noauosemn emmmz
Hopmm Sowoaoano dma
econ once Goapozuon :mmmz
nevus nowoaonne «ma
canwpmes. mapepmne.umpaaanmpm “Heads
a: Ham an ammumem “was ammoaonso any
oobhomno pdamom oopoogwm vogue: soapduaaopomhmno

 

 

mam mm pozooaa soapdhohsoc on» mo Soapmuanopodamno .NHNM wands

175

 

oHoa\oHoa wm.o

oHoa\oHoa mm.o

+

oaoa\oaoa mw.o

 

ooanp

oHoa\oHoa wmm.o

oHoa\oHoa oo.a

oHoa\oHoa oo.H

+

mHoa\oHoa oo.a

 

owns»

oHoa\oHoS oo.a

ommaouam max :9“: owmbmoao
amaze godpaampaa sesameasuoeq

ommaooam mam suds owo>moao
amped soapmaonaa opmhzhhm

noameaHo opdooaaom
nevus zoapoopoo oohsouaopood

soapQESmsoo cumuodhom

henna open
opmooaaoa zodpmnzwamsoo thondhm

soapoahwopnoooc oDSMHzm odaoo

176
5. Chemical Synthesis of 2-Keto-3-deoxy-D-fuconate

This compound was synthesized via modifications of
the method used for the synthesis of N-acetylneuraminic
acid (67). The reaction involves an aldol-type condensa-
tion of lactaldehyde with a pyruvyl carbanion generated
by alkaline B-decarboxylation of oxalacetate. DL-Lact-
aldehyde was prepared by the oxidative deamination of
DL—threonine using the procedure of Abeles gt a}. (68).
Chromatography of recrystallized DL-lactaldehyde on

Dowex-i-HSOB by the procedure of Huff (69) and analysis
of the fractions with 3-methyl-2-benzothiazolinone
hydrazone indicated the presence of at least 12 carbonyl
compounds with lactaldehyde in excess of 99%. Thin-layer
chromatography plates develOped in solvent system (i) and
visualized with 3-methyl-2-benzothiazolinone hydrazone
also indicated at least 6 carbonyl compounds were present
in minor amounts. Paper chromatograms developed in sol-
vent system 3 indicated 5 ninhydrin positive compounds
were present in the above recrystallized DL-lactaldehyde.
DeSpite the above contaminants, the DL-lactaldehyde dimer
melted at 103-05°C; reported 105°C (74). The 2,4-dinitro-
phenylhydrazone was also prepared, m.p. 156°C; reported
156.5-157°C (70). It Should be pointed out that previous
workers who have prepared lactaldehyde by the above proce-

dure have assumed purity on the physical evidence of melt-

ing points (68, 70-73, 262); the possible effects of the

177
13-15 contaminants upon the enzymatic systems which were
under investigation is a matter of conjecture. The
recrystallized lactaldehyde was used without further
purification for the synthesis of KDF.

DL-Lactaldehyde (0.34 g) was dissolved in 2.0 ml
water, and the pH was adjusted to 7.2. Oxalacetate
(0.226 g) was dissolved gradually in 2.0 ml water; the
pH was maintained at 8.5 with a Sargent recording pH
stat using 8.5 molar NaOH. The lactaldehyde solution
was then added to the oxalacetate solution, and the pH
was adjusted to and maintained at 11.0 at 25°C. The
progress of the reaction was followed by withdrawing
aliquots from the reaction mixture and analyzing for
2-keto-3-deoxy acids with TBA.

The reaction was judged to be complete within 20
minutes (Figure 43) after which the solution was neutra-
lized with AGSOW-X8 and filtered. The filtrate was
treated with Norit (1.5 8) until it was colorless, and
the resultant solution was applied to an AGi-XB formate
column (200-400 mesh, 0.5 x 15 cm) and eluted with one
liter of a 0-0.6 molar formic acid gradient followed by
one liter of a 0.6-5.5 molar formic acid gradient. The
column was washed free of non-ionic substances with
water (0.5 1) before initiating elution procedures.
Fractions (15 ml) were collected and analyzed with the

TBA and semicarbazide assays. One TBA-positive peak and

 

178

ad nobam ohm maampon

.pNop on»

.mommd dma on» an oohommm ones one oazpwaa

coupomoh can Bosh ssmhuspds ohms AH: mv mposvaad. .Uaom nanomOHSNos

Imwooodolm.m mo mamosvshm HmoaaoSo on» mo ouch noapooom .m: onzwam

 

Figure 43

179

 

£7

 

 

 

(um ISS ) somaosav

30

 

M INUTES

180
two semicarbazide-positive peaks were observed in the
elution profile (Figure 44). The fractions containing
the TBA-positive peak were pooled and freed of formic
acid by three evaporations under reduced pressure; yield
680 umoles, 40%. The second semicarbazone peak was-
eluted at approximately 5 molar formic acid and thus
resembled pyruvic acid (18); the fractions containing
the second semicarbazone peak were pooled and treated
as above and yielded a substance which reacted with
lactic acid dehydrogenase and NADH.

The pooled fractions (TBA-positive peak) were
rechromatographed on the AGi-X8 formate column with a
0-0.6 molar formic acid gradient; one slightly skewed
peak was obtained (Figure 45). The compound co-chromato-
graphed with enzymatically prepared KDF in solvent sys-
tems 2 and 6. Chemically prepared 3,6-dideoxy-DL-
hexulosonic acid, DL-KDF, prepared above, was cleaved to
0.485 mole pyruvate per mole by KDF aldolase (presented
in the next section). The rate of release of B-formyl
pyruvate in the periodate oxidation of chemically synthe-
sized DL-KDF was identical to that of enzymatically pre-
pared D-KDF using the techniques described in Figure 42,

. Chromatography of the pooled fractions containing
DLPKDF on an AG1~X8 carbonate column (200-400 mesh, 0.5 x
15 cm) by the procedure of Hershberger, Davis and Binkley
(75) or on an AGi-X8 borate column (200-400 mesh, 1 x 30

181

.hmmmm cosmonaooaaom

0:9 Spas oouhaoso one: mzoapomam .pwop on» ma zobaw cad cadence

.pSoaomaw used odsaoa adaoa m.mlo m wzdms opdfihom HIHoSoQ Scam odes

cabsahn one oaom odzomoHSNoSIAQIhHoeoHonw.m no soapzam .3: ohswam

182
(mu 19$) eonvasossv

Figure 44

mmmzbz ZOHBOdmm

 

 

one oma oma one we on ma
~\u _ A fi 01.. o_ _ -0
e . o
A x
,_ a i.
/
m.or| H 1 o Aw nl
~¢ N ..
i i 4r/// A
H a“ ,me«>pmmm A
s .
o.HIJ H x . _ II
A _ _
,1 \ . _
G ha . _
_ .
_ _
m an: mnHNammdonmm.|AY||AYI p In
_
ema laminar. “
CON] Ii
azmHoemo oHoa onmom
mm.mA.ule.o were?
iA VA

 

.mm.o

om.o

mm.o

oo.H

 

 

(mu oSz) souvanosev

183

Figure 45. Elution of 3,6-dideoxy-DL-hexulosonic
acid from Dowex-i formate using a 0-0.06 molar

formic acid gradient. Details are given in the
text.

 

ABSORBANCE (551 nm)

184
Figure 45

 

I I I I

3.0 *-

2.0

1.0

 

 

 

 

25 5b ‘ 75 100
FRACTION NUMBER

185
cm) by the procedure of Samuelson, Ljungquist and.Parck
(76) also yielded one peak. It was concluded that the
isomer of 3,6-dideoxy-DL—hexulosonic acid that was
formed was the 23339 isomer on the basis of co-chromatog-
raphy with enzymatically prepared 3,6-dideoxy-D-thggg-
hexulosonic acid (KDF), cleavage by KDF aldolase, and on
the basis of the Dowex-i-borate, -formate and ~carbonate
chromatography in which one peak was consistantly
obtained. It was also concluded that the condensation
was stereOSpecific under the conditions employed or that
the erythro isomer was less stable under the isolation
conditions.

D-KDF was also synthesized using the same condi-
tions employed for DLpKDF with D-lactaldehyde prepared
from Luthreonine. The product from the AGi-X8 formate
column was identical to enzymatically prepared KDF and
was cleaved to 0.98 mole pyruvate per mole KDF by KDF
aldolase. It is a possibility that the erythro isomer
was a contaminant to the extent of 1.5% and was not
separable from the three isomer utilizing the above
chromatographic procedures. It is not known whether the
erythro isomer is cleaved by KDF aldolase.

Synthetic KDF was found to be unstable under the
solvent extraction procedure and partially degraded to
a compound which reacts with lactic acid dehydrogenase

and NADH and co-chromatographs with pyruvate. High pH

186
()9) was found to be deleterious and resulted in orange
colored solutions and large losseSS of TBA-positive
material. KDF forms a lactone which is easily hydrolyzed
at pH 8.0. KDF may be stored indefinitely at -20°C and
pH 7.4 without deterioration. KDF could also be isolated
as the partially crystalline potassium salt by 1y0philiza-
tion.

The structure of the dehydration product of D-
fuconate has been rigorously determined by degradation,
derivatization. and, finally, by chemical synthesis.

The enzymatically prepared compound has been found to be
identical in all reSpectS to the chemically prepared
compound, and it has been concluded that the dehydration
product of Dufuconate is 3,6-dideoxy-Dethrgg-hexulosonic
acid.

F. CHARACTERIZATION OF 2-KETO-3-DEOXY-D-FUCONATE.ALDOLASE
1. Preliminary EXperimentS

By analogy to known reactions of deoxy hexulosonic
acids, the most likely degradative reaction for KDF would
be cleavage by an aldolase. The pathways of galactose
and glucose metabolism in some bacteria involve 2-keto-3-
deoxy-D-galactonate and 2-keto-3-deoxy-D-gluconate
reSpectively (78, 19), which undergo prerequisite phos-
phorylation by Specific kinases prior to cleavage by
Specific aldolases (19, 77-79); however, KDF cannot be
phosphorylated in a similar reaction due to the lack of
an hydroxyl group on C-6 and thus would not be eXpected
to be cleaved by these two aldolases. The pathway of 2-
keto-3-deoxy-L-arabonate degradation in some bacteria
involves dehydration to form d-keto glutarate semialde-
hyde (54-56): KDF cannot participate in a Similar reac-
tion, once again due to the lack of an hydroxyl group at
C-6.

An enzyme has been described in a pseudomonad,
however, which is instrumental in the degradation of the
rare carbohydrate D-arabinose and which cleaves 2-keto-3-
deoxy-D-arabonate to form pyruvic acid and glycolic acid
(80). Consequently, a Similar aldolase was sought which
could cleave KDF, presumably to pyruvate and D-lactalde-

hyde. Spectrophotometric analysis for such an aldolase

187

188
using KDF, cell extract, lactic acid dehydrogenase, NADH
and the appropriate controls indicated Slow cleavage
((0.3 mmole per hour per mg protein); further, incubation
of KDF with cell extracts resulted in a decrease in TBA-
positive material also amounting to 0.3 umole per hour
per mg protein. The enzyme was purified on Sephadex G-200,
and activity was located with the lactic acid dehydrogen-
ase and the TBA assays; the peaks were superimposable
(Figure 46).

Further investigation indicated that the enzyme
had an absolute requirement for a divalent cation, was
protected by thiols,and also cleaved 2-keto-3-deoxy-L—
arabonate (KDA), in addition to KDF. The products of the
cleavage of KDF or KDA were tentatively characterized as
pyruvate, by its reaction with lactic acid dehydrogenase
and NADH, and as an aldehyde, by the formation of a tetra-
azopentamethine cyanine derivative with 3-methyl-2—
benzothiazolinone hydrazone.

The observed cleavage of 2-keto-3-deoxy—L-arabonate
(L-KDA) was totally unexpected since the only previously
known pathway for LPKDA degradation, as elucidated by
Weimberg and Doudoroff (54) and Weimberg (55), involves
an entirely different sequence of reactions, as shown

be 10W 0

189

Figure 46. Sephadex G-200 profile of protamine sulfate-
treated cell extract from D-fucose-grown cells. The
fractions were analyzed for aldolase activity on KDF by
two means: (i) the TBA assay; and (ii) the lactic acid
dehydrogenase-coupled assay. The column (3 x 35 cm) was
equilibrated with 0.01 molar sodium phosphate buffer

(pH 7.0). The enzyme was eluted with the same buffer.

TBA endpoint assay: The assay consisted of 100 pl each
fraction, 0.0849 umole KDF, and 0.2 umole sodium
phosphate buffer (pH 7.0) in a total volume of
0.23 ml. The reaction was incubated at 27°C for
20 minutes before quenching with periodate.

Lactic acid dehydrogenase-coupled assay: The assay con-
sisted of 30 ul of each fraction, 0.07 umole NADH,
13 ug lactic acid dehydrogenase, 5 umoles KDF
(pH 7.0) and 5 umoles sodium phoSphate buffer in
a total volume of 0.15 ml. The absorbancy change
at 340 nm and 25°C was followed, and controls to
correct for endogenous pyruvate and NADH oxidase
were run.

 

uMOLES/HOUB/O . 1 0 ML ALIQUOT

190

 

Figure 46
a LACTIC AC ID
DEHYDROGENASE
O /
O
-5.0
(MG/ML)

. PROTEIN
/\

0.501

 

0.25;. '

 

I I I

 

 

240 280 320 360

ELUATE (ML)

 

191
COOH COOH COOH

w
(3..
a:
N
i
Q...
n:
N

HO-C-H H20 CH2 H20 CH2
l
CHZOH CHO COOH
L-KDA c-Keto glutarate d-Keto glutarate
semialdehyde

The previously known pathway for LnKDA degradation
involves the dehydration of L-KDA to form c-keto glutarate
semialdehyde which is subsequently oxidized to a-keto
glutarate. The L-KDA dehydratase of these workers was
routinely assayed by following the rate of NAD+ reduction
or the rate of decrease of TBA-positive material (Since
c-keto glutarate semialdehyde will not yield B-formyl
pyruvate upon periodate oxidation). Repeated attempts

to demonstrate the presence of the above pathway in the
pseudomonad under investigation were totally unsuccess-
ful: (i) there was no NAD+ or NADP+ reduction when Lp
arabonate or L-KDA were added to reaction mixtures con-
taining varying amounts of cell extracts, NAD(P)+, metal
ions, various buffers, and thiols; (11) no reduction of
c-keto glutarate with NADH or NADPH could be demonstrated;
(iii) loss of TBA-positive material when KDA was incu-
bated with Sephadex G-200 fractions of protamine sulfate-
treated cell extracts correSponded to the aldolase peak

as measured by KDF cleavage; (iv) the rate of decrease

192
of TBA-positive material when cell-free extracts were
incubated with KDA corresponded to the rate of KDF
cleavage measured in the same manner; and (v) nearly
quantitative conversions (98%) of L-arabonate to KDA, as
measured by the TBA assay, were carried out by cell-free
extracts, indicating no dehydration of KDA occurred.
Thus, it was concluded that KDA dehydratase was not
present in this microorganism and that a new pathway for
L-arabonate metabolism had been discovered.

The aldolase, hereafter referred to as KDF aldo—
lase, was totally inactive on 2-keto-3-deoxy-D-galactonate,
2-keto-3-deoxy-D-gluconate, and their reSpective 6-phos-
pho-derivatives. Keto deoxy galactonate formed by galactonate
dehydratase was found to be phoSphorylated by a kinase
prior to cleavage, and is demonstrated in Figure 47; in
addition, 2-keto-3-deoxy-6-phoSpho-D-gluconate was found
to be cleaved by cell extracts. Thus, it was concluded
that the pathways for glucose and galactose metabolism
in the pseudomonad under investigation were identical to
the previously described pathways (19, 77-79). Thus,
deSpite the structural Similarities between D-fucose and
D-galactose, the two carbohydrates are degraded by
entirely different routes; this will be discussed to a
greater extent in the next section.

The results of the initial experiments indicated
that an enzyme was present in D-fucose- and L-arabinose-

grown cells Which could cleave KDA or KDF to pyruvate

193

53.on OIIO

ommSaM oposopodacwnnlhwooUIMIopoMIN nvILu

omdaognd opdsopodddmlnlosamonenmlhwooolmlopoulm I
. omoHoUHd_mnM nTliu

.onomoq

. .0: onzwam :a confluence mo
oohoHQSo was onsoooona oahposoposmoauooam came one ”nomad omdaoofim mam

.ssa omHo one: oudbfihhn one omooawo mmdz

msosowouno you pooaaoo op mHoHpnoo .oomm use a: 03m pd hHHdOdhpoa

noposaonpoonm dosashopoc mes hpabdpod_ .Ha ma.o Mo ossaob Hdpop

o s« sodpoonm some no H1 ooH use .mmdz oaoan mo.o .omoSowoaonsoo

odes oapodfl m: ma .opdSopocamwlaaosamosalwuhwooolmlopoxlm macs:
o.H mo oopmamSoo homes 029 “homes omdaooam cacao»odamwualhwooUIMIopoMIm

.coxoaaom was Hmaaopos
obdpamoaldma SH ommoaooo on» .mSSp use .opdooanon an doudcawo no:
as .oposopooadmaanosamosauwlhwooolmuopoMIN .posooha one .Udod
odazpdnnmpoHSpnm sud: scammed son» use comm no mops:as on How
ooponzos« son» one: escapees“ one .zoapooam some no H1 ooa use
opozopomddwlmnhwoooIMIOpox:N macs: nu.o .NHowz macs: o.H .mad macs:
m.o no oopmHmsoo momma one “momma ommsas opdsopomaowlnnhwooolmnopoMtm
.Aoumfip came on» spa: confide one: moathO esp use
.An.n mav Hogans opmsnmosn azaoom HoHoa no.0 suds dopmhndaazdo no: “so
ow N av sasaoo one .modpabapoo omcaoodd mam end .omoaooad opczopodadw
IonmmosanmlhxoooIMIopoMIN use commas openopomaowlnnhwooolmuopoutN you
condoms one: msoapooam .mHHoo macaw omosandadiq Scam poonpwo HHoo mo

scapodam oudmaSm azdsosad moose: mo oaamoan oomlo Novosaom .5: ouswam

Figure

47

194

 

 

 

 

800‘

 

 

 

 

I I I l
0 UN 0 UN
0 {\- l-fi N
.4 a c3 5

TN/OW NIHIOHd

I I
O 0
° "2
u; e

'IM/HflOH/SE'IOMr1

700

600

500

400

ELUATE (ML)

195
and an aldehyde. The enzyme required a divalent cation,
was stabilized by thiols, was inactive on other deoxy
hexulosonic acids, and was distinct from previously
reported enzymes for D-galactose, D—glucose, and L-arab-
inose metabolism. In addition to the conclusion that
the aldolase was the terminal enzyme instrumental in
D-fucose metabolism, it was also concluded that a new

L—arabinose pathway had been discovered.
2. Purification

Cell extracts of L-arabinose-grown cells were pre-
pared as described in EXperimental Procedures with the
exception that the cells were disrupted in 0.10 molar
Bicine buffer and 0.143 mM 2-thioethanol (pH 7.4) to
'eliminate NADH oxidase. Except where indicated other-
wise, the fractionation procedures were carried out at
O-4°C. A summary of the purification is given in Table
XXX.

Protamine Sulfate Fractionation- The protein con-
centration of the cell extract was adjusted to 16 mg per
ml by dilution with 0.10 molar Bicine buffer and 0.143 mM
2—thioethanol (pH 7.4). .Ammonium sulfate (18.5 g) was
dissolved in 700 ml cell extract to a final concentration
of 0.20 molar, and then 140 ml of a 2% protamine sulfate
solution in the above buffer was added with stirring to
a final concentration of 0.33%. .After 30 minutes the

196

Table XXX. Purification of KDF aldolase*.

 

 

Fraction Units Specific Fold 280/260
Activity
Cell extract 66,600 4.1 1 .62
Protamine sulfate 66,100 4.9 1.2 .87
Ammonium sulfate 57,400 16.5 4.1 1.20
Sephadex G-200 14,600 69.8 17.0 1.38
Heat Step 10,600 208 50.7 1.59

*KDA was used as substrate.

197
suSpension was centrifuged at 40,000 x g for 10 minutes,
and the resulting precipitate was discarded.

Ammonium Sulfate Fractionation- The 40,000 x g
supernatant (840 ml) was brought to 40% of saturation by
the addition of 165 g of ammonium sulfate, and the
resulting precipitate was centrifuged down and discarded.
The supernatant was then brought to 60% of saturation
with 111 g of ammonium sulfate, and the resulting precipi-
tate was collected by centrifugation and dissolved in 82
ml 0.10 molar Bicine buffer and 0.143 mM 2-thioethanol
(pH 7.4). The protein concentration at this stage was
42 mg per m1.

Sephadex G-200 Chromatography- The above 40-60%
fraction was placed on a column (6 x 60 cm) of Sephadex
G-200 equilibrated with 0.05 molar Bicine buffer and
0.143 mM 2-thioethanol (pH 7.4). The enzyme was eluted
with the same buffer. Fractions (15 ml) were collected,
and those which contained the most activity were pooled
(105 ml total).

Heat Step- The protein concentration of the pooled
Sephadex G-200 fractions was adjusted to 2 mg per ml with
0.10 molar Bicine buffer and 0.143 mM Z-thloethanol (pH
7.4); Sufficient 0.10 molar magnesium chloride was added
to 10 ml of the pooled Sephadex G—200 fractions to bring
the concentration to 0.6 mM. The protein solution was

immersed in a 55°C bath for 8.5 minutes and then chilled

198
in ice. The flocculent precipitate was removed by centri-
fugation at 40,000 x g for 10 minutes. The supernatant
was used as the purified aldolase fraction. The protein

concentration was 0.5 mg per ml.
3. Properties

pH Optimum- KDF aldolase activity as a function of
pH was maximal at pH 8.1 in Hepes and Tris-H01 buffers
(Figure 48).

Substrate Specificity- KDF aldolase cleaved only
KDF and KDA among the nine deoxy acids tested. The deoxy
acids which were not cleaved by the aldolase were 2-keto-
3-deoxy-D-galactonate, 2-keto-3-deoxy-D-gluconate, 2-keto-
3-deoxy-6-phoSpho-Dbgalactonate, 2-keto-3-deoxy-6-phoSpho-
D-gluconate, 2-keto-4-hydroxy glutarate, 3-deoxy-D-x lo-
hexonate, and N-acetylneuraminic acid. The first four
deoxy hexulosonic acids which did not serve as substrates
also did not reduce the rate of cleavage of 20 mM KDF
when added at equimolar concentrations.

From the Lineweaver-Burk plots presented in Figures
49 and 50, the Km values for KDF and KDA were determined
to be 2.9 mM and 1.8 mM reSpectively. KDF was cleaved at
47% of the rate of KDA cleavage under saturating condi-
tions.

Metal Ion.Activation- KDF aldolase activity has an
absolute requirement for a divalent cation as indicated

in Table XXXI. Mn2+ was found to be the most effective

199

.oponpopzm on» no new Spas does no:

Homnmdna use .oponpmnzm on» mm «mm suns coo: mos nonnsn monom .annoa
zonpooon on» no oonsoo on» manned omsoSo no: one ma one .oonapuna soap
nooon opoonamso no dosnsnopoc onos mpsoaonSoooa ma one .noaoa o~.o mos
nonnsn zoom .ooms mos munbnpoo cannooam poosmas osp suns soapoonn oomlo
Novosnom oopoonpupoon osa .psopmSoo nonponpsoosoo oooHouHo on» spas
.oopooao:« mo .annob onos sonpnmonaoo nonnsp on» was we on» nos» pnoowo

oomoaaao mos homoo oncogene one .oooaooao mam no azadpno mm .m: onswnh

Figure 48

200

 

 

 

_

9.0

8.0

7.0

 

 

100

UN
l\

XIIAILOV HAIIVTHH %

50

pH

201

Figure 49. Lineweaver-Burk plot relating KDF con-
centration to KDF aldolase activity. The standard
assay was employed except that the concentration of KDF
was varied, as indicated, with the aldolase concen-
tration constant. The heat-treated Sephadex G-200
fraction with the highest Specific activity was

used.

Figure 49

202

 

<H‘

 

I

I

Km FOR KDF e 2.9 x 10-3 y;

3.0

2.0

1.0

 

 

 

 

0.500

KDF (mM)

1.00

 

203

woouaddus

mes naaeapem eacnemam persons one run: soaeoenn oomuo needsaem
consonplpoos one .oopoOHosa mo connon on: soaponpsoosoo «am

one non» anoowo uonoaaao mos momma onoosopo one .hpdnnpoo oooHooHo
mam on sonponpsooSoo dam weapoaon poaa anmnnobooxoSHA .om onswnm

204

oo.H

om.o

Azsv one

 

H

om.o:

 

Figure 50

 

 

 

rd>

 

 

Table XXXI.

205

Effect of metal ions on KDF aldolase. The
standard assay was used with the exceptions
that 0.2 umole of the metal salt and 0.05
umole EDTA (pH 7.0) were added. The heat-
treated Sephadex G—200 fraction with the
highest Specific activity was used. KDF was
used as the substrate.

W

 

Compound Relative Rate
MnClz 100%

MgC12 , MgSO)+ 5’-I

NiC12 45

Cu304 0

No ion added

2

 

206

activator at 3.3 mM with C02+, Mg2+, and N12+ being
partially effective. There was no variation in activity
when assays were run in the presence of 3 mM to 60 mM
MnClz. Aldolase activity with 0.33 mM EDTA in the
absence of added metal ions was 2% of the control activ-
ity.

Analysis of Sephadex G-200 fractions for KDF and
KDA cleavage in the presence and absence of added metal
ion yielded some interesting results (Figure 51). Under
conditions of saturating KDA and in absence of metal ion,
one small aldolase peak was obtained at about 740 ml of
the eluate (the 740 peak): activity in this peak was not
diminished in the presence of 0.66 mM EDTA. Under condi-
tions of saturating KDF another small aldolase peak was
obtained at about 660 ml of the eluate (the 660 peak):
the activity in this peak was completely abolished with
0.66 mM EDTA; When 3.0 mM Mn012 was incorporated into
the assay, a new peak 10-fold higher in activity was
observed with both KDA and KDF cleavage which was cen-
tered at the 660 peak. The 740 peak represents a lower
molecular weight protein than the 660 peak, and it is
tempting to suggest that the 740 peak represents a
"monomer" fraction and that the 660 peak represents a
"dimer" fraction. The "monomers," formed by dissocia-
tion of the "dimers," slowly cleave KDA but not KDF in
the absence of added metal ion; the "dimers" Slowly

 

207

.homoo enounopo ooaxnospo

on» an coon no: Hons» oz .NHoaz no oonooonn on» Ga duo cocoons on» ad

msoapoonn oomlo Novosaom an owobooflo saw one mam no odohdond

.am enemas

208

Figure 51

Ages messes

 

com cos .ooe can
i . . r
I e 4 O
i O I I .MO
< a) I 4 /
4 I
O
/
O I / .
c an... on .
I +N H mg
I f
I CE OS .. / II on
+ a
C
0.
AN .
+N 2 «an
_ _ con

 

 

 

LLI ALLOY 3A1 LV'IHH %

209

cleave KDF but not KDA, also in the absence of metal ion.
The "dimers," and not the "monomers“ cleave both KDF and
KDA at a 10-fold rate in the presence of added metal ion.
Attempts to reassociate the "monomers" in the 740 peak
with 2-thioethanol into "dimers" were consistently unsuc-
cessful. The 740 peak was not present in preparationsin
which 2-thioethanol had been incorporated. These results
are of a preliminary nature and deserve further investi-
gation.

Sulfhydryl Activation- The enZyme is insensitive
to thiol inhibitors and thiols during assay (Table XXXII),
but is stabilized by thiols during purification. IA time-
dependent renaturation process has been observed to occur
when Sephadex G-200 fractions containing the aldolase
were incubated with 2-thioethanol (Figure 52).

Eguilibrium Constant- Equilibrium positions of the
cleavage reaction with reSpect to both substrates were
determined (Table XXXIII). The equilibrium could be
approached from both directions, and the reactions were
exothermic in the direction of synthesis at 1 mM: the
equilibrium constants for KDF and KDA were determined to
be 0.12 and 0.36mM reSpectively.

Stability- KDF Aldolase possesses a half-life of
about 12 minutes at 55°C (Figure 53). The thermal dena-
turation profiles for KDF and.KDA are linear and super-
imposable, suggesting that the activity in the heat-

treated Sephadex G-200 fraction was due to a single

Table XXXII.

210

Effects of thiols and thiol group inhibitors
on KDF aldolase during assay. The standard
assay was employed with the exception of the
variation in the thiol or the thiol inhibitor
added. The Sephadex G-200 heat-treated frac-
tion with the highest Specific activity was
used. The pH of all reagents was 7.0. KDF
was used as the substrate.

 

 

._Compound Concentration Relative Rate
2-Thioethanol 3 mM 100
Reduced glutathione 3 mM 95
Dithiothreitol 3 mM 101
Iodoacetate 2 mM 97
prhloromercuribenzoate 2 mM 98

No thiol or thiol inhibitor added —- 100

 

211

Figure 52. Time-dependent thiol renaturation of KDF
aldolase present in Sephadex G-200 fractions not pre-
viously stabilized with thiols. Sephadex G-200 frac-
tions were incubated with 0.0143 molar 2-thioethanol
at (4°C for 11 hours in 0.10 molar Bicine buffer (pH
7.4). Each fraction was analyzed at 0, 2, 4, and 11
hour intervals. The aliquots were analyzed with the
standard assay. .A control with no thiol was also run

but was assayed only at to and til.

 

% RELATIVE ACTIVITY

200.

150

100

50

Figure 52

212

 

 

 

 

60

FRACTION NUMBER

70

 

213

as moa.o SE m.ma *** 2s mmqa 2s MHqH om.m .Qndbsnhm +

 

 

oehsooaopooq
SE N3H.o EH Noem tat 2E Nmm.o :8 mmw.o Chou ham
2s mem.o sss as Hm.m 2s mo.a as mm.a 03.5 «as
as mmm.o see 2s m:.m as mm.n 2s mw.n 93.5 opobsnmm +
monsoonenoonno
oohsoodopooq

mom son «as no someones me sad: soonopm

ounSooHoHoonHo sonpooom

msonponpsoosoo sanQnHHSdm
g

.oonsoedoaoondw oso poohsooaopooa,.opo>snmn .onoo nsoooumnopoxlm
non condoms onos use sonnsaom ssnnpnanswo so sonn ssonoSpnz onos
oposdnad .4QM ego was no owonooao on» non opsoposoo sannndHHSUm .HHHNNN oases

 

214

.4a& was mam anon wanes hound unocsdpm

on» sans Uohdomd use SkonUSpal onox oposddad. .A3.m may Hososno

noEpum as 915 one moses nodes 36 5 season mes 3s non we $6

.Hs av soapsaoo saoponn osa

.ooos mos sonpoonn comic Hoodsnom consonp

spoon one .oomm no omoaooao mam no soaponSposoe Hosnosa .mm onswam

215

 

 

 

 

 

Figure 53
I I I I I
é a “'
O O
I I I I I I I
a g a s a 3 5.:

DNINIVNEH KIIAIIOV %

14

12

10

MINUTES AT 55°C

216

enzyme. The enzyme, present in the heat-treated Sephadex
c-200 fractions, lost 70% activity when stored at -20°c
for three weeks in 0.10 molar Bicine buffer and 0.143 mM
2-thioethanol (pH 7.4), but was completely stable to
lyOphilization in the same buffer system. AS mentioned
previously, the aldolase appears to dissociate on Sephadex
G-200; this could be prevented by incorporating 2—thio-
ethanol into the eluant.

Induction- The inducibility of KDF aldolase was
tested by growth on various substrates (Table XXXIV).
D-Fucose and L-arabinose induced the aldolase to a level
3-4 fold over the non-induced level present in nutrient

broth-grown cells.
4. Identification of Cleavage Products

The cleavage products of KDF and KDA in the aldolase
reaction were initially characterized as a compound which
reacted with lactic acid dehydrogenase and as a compound
which reacted with 3-methyl-2-benzothiazolinone hydrazone
to form a tetraazopentamethine cyanine dye. The cleavage
products were suspected to be pyruvate and an aldehyde,
either lactaldehyde from KDF or glycolaldehyde from KDA.
More rigorous procedures were carried out to positively
identify the products and are presented in this section.

A reaction mixture composed of 5 umoles MgClz, 1o
umoles 2-keto-3-deoxy-D-fuconate, 7.35 umoles 2-thioe

ethanol, and 0.5 mg of a heat-treated Sephadex fraction

217

Table XXXIV. Induction of KDF aldolase by various growth
substrates. The standard assay was employed
with KDF as the substrate. Protein was esti-
mated by the biuret assay. Cell extracts
were prepared by sonic disruption in 0.10
molar Bicine buffer and 0.143 mM 2-thio-
ethanol (pH 7.4).

 

 

Growth Substrate Specific Activity
(umoleS/hour/mg)
D-Fucose 3.91
LqArabinose 4.23
D-Glucose 1.28
D-Galactose 1.34

Nutrient Broth 1.08

 

218
in a total volume of 1.0 ml was incubated at 27°C for 1
hour. The reaction was quenched by the addition of 1.5
ml of a saturated solution of 2,4-dinitrophenylhydrazine-
HCl in 2 N HCl. The hydrazones were allowed to form for
20 minutes after which the mixture was extracted with
three 10-ml portions of toluene. The toluene extracts
were combined and reduced in volume under vacuum (bath
temperature 20°C) to 0.5 ml. Thin-layer chromatography
plates were Spotted with 50 ul samples and were develOped
two-dimensionally in the following solvent systems (81):
lst dimension, n-butanol-ethanol-0.5 N NH40H (70:10:20);
2nd dimension, benzene-tetrahydrofuran-glacial acetic
acid (57:35:8). The plates were Sprayed with an aqueous
10% KOH solution after drying, and the Rf values and color-
ation of the 2,4—dinitrophenylhydrazones recorded. Two
well-defined Spots with distinct coloration were observed.
The 2,4udinitrophenylhydrazones were scraped from plates
not Sprayed with KOH and were eluted with 0.5 N NaOH and
the absorption Spectra of the alkaline solutions of the
2,4_dinitrophenylhydrazoneS recorded (82). Standard 2,4-
dinitrOphenylhydrazones of pyruvate and lactaldehyde were
prepared by the method of Haway and Thompson (83) and
Huff and Rudney (70), respectively, and were treated in
the same manner as the extracted 2,4-dinitrophenylhydra-
zones with the exception that the solid standards were

dissolved in toluene before Spotting on thin-layer

219

plates. The results, in Table XXXV, indicate that pyru-
vate and lactaldehyde were formed from the cleavage of
KDF. The absorption Spectra of the standard 2,4-dinitro-
phenylhydrazones of pyruvate and lactaldehyde were also
found to correSpond to the absorption Spectra of Spots
1 and 2, reSpectively. and are presented in Figures 54
and 55. It was concluded that pyruvate and lactaldehyde
were the cleavage products of KDF aldolase action upon
KDF.

Cleavagezgroducts of 2-Ket0:3pde9§1;L-arabonate-
The proceeding procedure was duplicated with KDA with
the exception that the 2,4-dinitrophenylhydrazone of
glycolaldehyde was prepared by the method of Powers,
Tabakoglu, and Sable (84). The results are presented
in Table XXXVI. It was likewise concluded that the
products of KDA cleavage were pyruvate and glycolaldehyde.

The identification of the cleavage products of KDF
and KDA were carried out by conversion of the cleavage
products to the reSpective 2,4-dinitrophenylhydrazoneS
and extraction with toluene, by two-dimensional thin-
layer chromatography, and by comparison of the visible
absorption Spectra of the isolated 2,4-dinitr0phenylhydra-
zones with standard 2,4-dinitrophenylhydrazones. It has
been concluded that KDF aldolase cleaves KDF and KDA in

an aldol-type cleavage reaction between C-3 and C-4.

220

 

me. He. szonm soapooon owobooflo mm ponm

 

 

mm. no. szonm onoosopm oomSovHopooAIQ
N3. on. szonnloamnsm sonnooon owobooao «k scan
as. mm. ssonnuoaansm onoosopm opobsnmm
wk sonmsosnm Hm Godmsosnm monmm mom mosoNonumsamsosaonpasnmnd.N

noses aonoo

 

mosao> nm
3

.psop one 2H sonaw one manopon
.onmnoonm man an ommemono was no mposoona one no sonomonnnosoon .>xxx oases

221

Figure 54. Absorption Spectra of the 2,4-dinitro-
phenylhydrazones of pyruvic acid and Spot #1
obtained from thin-layer chromatography of the KDF
cleavage products. The solvent was 0. 5 N NaOH.
Details are given in the text.

ABSORBANCE

222

Figure 54

 

c.6o.—-

 

PYRUVATE

0.40

 

 

 

400 440 480 560

223

Figure 55. Absorption spectra of the 2,4—dinitro-
phenylhydrazones of D-lactaldehyde and Spot #2
obtained from thin-layer chromatography of the KDF
cleavage mixture. The solvent was 0.5 N NaOH.

Details are given in the text.

 

 

ABSORBANCE

224

Figure 55

 

    

IACTALDERYDE

 

 

 

360 400 440 480 520

225

 

mm. ow. ssonnusoHHoH sodpooon owonooao «w poam

 

 

mm. . om. ssonpnsonnon oneoempm monsoonenoonno
as. mm. stonnnoasnsm sonpooon owobooao H¥ noun
0:. mm. szonnuoaansm onoesopm opensnmm
wk sodoSoan ax sonwsosnn monnm mom mosononUASHmsosaonndsnnld.N

nouns nonoo

 

 

 

oosHo>_nm

 

 

 

 

 

.pwop on» an sonnw ono manopom
.oooaooao mam an owonooao saw no oposoonn on» no sonpoonnapsooH .H>xxx oases

G. ANALYSIS OF MUTANTS LACKING D—FUCONATE DEHYDRATASE
AND 2-KETO-3-DEOXY-D—FUCONATE.ALDOLASE, AND THE
RELATIONSHIPS AMONG THE DEGRADATIVE PATHWAYS FOR
D-FUCOSE, LqARABINOSE, AND D-GALACTOSE

1. Mutant Strain 5-1-10-1 (D-Fuconate Dehydrataseless)

A mutant was isolated which was incapable of using
L-arabinose or Dbfucose as growth substrates but was
identical to the parental strain in other reSpects (see
Experimental.Procedures for the methods of mutagenesis
and isolation, and the Appendix for the prOperties of the
parental strain). The mutant, which was designated
strain 5-1-10-1, was found to be missing D-fuconate dehyd-
ratase, but possessed normal levels of the other enzymes
of the D-fucose pathway (Table XXXVII). The level of D-
galactonate dehydratase was also determined and was found
to be normal. The growth rates of mutant strain 5-1-10-1
on D-glucose and D-galactose were identical to those of
the wild type (Figures 56 and 57).

The above data thus support two conclusions:

(1) D-fuconate dehydratase is absolutely functional in

the metabolism of D-fuconate and L-arabonate: and (ii)
D-galactonate dehydratase is not functional in the metabo-
lism of D-fuconate. Previous evidence indicated that
D-galactonate dehydratase could dehydrate D-fuconate:
however, the Km value was about 120 mM. The Km value

was of such a magnitude to warrant the assumption that

the enzyme was not involved in the in vivo dehydration of

226

227

Table XXXVII. Enzyme levels in mutant strain 5-1-10-1, the
parental strain and a revertant derived from

the mutant.

Cell extracts of D-galactose

grown cells, unless otherwise indicated, were
prepared by sonic disruption in 0.01 molar
Bicine buffer and 0.143 mM 2-thioethanol

(PH 70“).
biuret assay.

Protein was estimated by the
Aldolase assays were identical

to the assay outlined in the Experimental
Procedures with the exception that 1.05 umole

KDF was employed.

Specific activities are
reported.as.umoles per hour per mg protein.

SpecificrActivity”

 

 

 

Enzyme Wild Type Mutant Revertant
DnAldohexose
dehydrogenase* 13.9 12.7 14.6
LnArabino-aldose
dehydrogenase 11.8 20.1 17.5
DbFuconate
dehydratase 4.25 < .0005 3.72
Aldolase 1.34 1.91 1.57
D-Galactonate
dehydratase 6.50 7.11 6.79
*Cell extracts were prepared from D-glucose grown cells.

228

Figure 56. Growth of wild-type cells on D-fucose,

L-arabinose, Dugalactose, and Dnglucose.

ABSORBANCE (600 nm)

229

 

 

 

 

 

 

Figure 56
.900 _. I I __
.800 F _
.700 _ fi’
.600 ._ . .V __r
.. ,v.v°‘
.500 __ o _4
'o
v v
.400 _.— v .0 ' __
0300 _ .—
e
V
o
.200 (_ . L-ARABINOSE L
' vD-GLUCOSE
V oD-GALACTOSE
. ' b v D-FUCOSE
0V '
O .
.V V
A o
o
0070 . ' .—
o °v .
.060 V _.
I I I I
5 10 15 20

nouns AT 32°C

230

Figure 57. Growth of mutant strain 5-1-10-1 on

D-fucose, L-arabinose, D-galactose, and D-glucose.

ABSORBANCE (600 nm)

231

 

 

 

 

 

Figure 57 .
.900 I I I .1
.80 ._
.70 __
. 500I._ g __
.4001. _,
030W .......
O L-ARABINOSE
.200— v D-GLUCOSE ——
o D-GALACTOSE
88 8 V D-FUCOSE
.100.— ' B —1
O
V V V _
’0705'7 Y Y Y Y 3’ Y o o o
0060— .—
.0507._ _
I I I I
5 10 15 20

HOURS AT 32°C

232
D-fuconate. The fact that the D-fuconate dehydrataseless
mutant possessed D-galactonate dehydratase but failed to
grow on D-fuconate corroborates this assumption.

Thus, not only can Dbgalactonate dehydratase
activity be distinguished from D-fuconate dehydratase
activity by means of Sephadex G-200 and DEAE-cellulose
elution profiles, thermal denaturation profiles, differ-
ential metal ion requirements, differential sulfhydryl
inhibitor responses, pH optima and mixed substrate studies,
but a mutant has been obtained which is lacking D-fuconate
dehydratase and yet contains normal levels of D-galactonate
dehydratase. A comparison of some of the properties of
the two dehydratases is presented in Table XXXVIII.

2. Mutant Strain 73-1-2 (Lacking 2-Keto-3-deoxy-D-

fuconateIAldolase)

Another mutant which exhibited defective growth on
D-fucose and L-arabinose was obtained by procedures iden-
tical to those used for the isolation of mutant strain
5-1-10-1. This mutant, designated 73-1-2, grew on L-
arabinose and D-fucose at about 40% the wild-type rate
but grew normally on Duglucose and D-galactose (Figures
56 and 58). This mutant when grown on L-arabinose or
D-fucose was found to possess about one-fourth the normal
level of 2-keto-3-deoxy-D-fuconate aldolase but possessed
normal levels of the other enzymes of the D-fucose pathway

(Table XXXIX). The slow-growing mutant strain is believed

Table XXXVIII.

233

A comparison of the properties of D-fuconate
and D-galactonate dehydratases

 

 

D-Fuconate D-Galactonate
PrOperty Dehydratase Dehydratase
Km for D—fuconate 4.00 mM 120 mM
Km for L-arabonate 4.25 mM not a substrate
KIn for D-galactonate not a substrate 12 mM
Half-life at 53°C 80 sec > 1 hour
pH Optimum 7.1-7.4 6.5-6.8
Metal ion requirement absolute 2—3 fold
(EDTA t excess cation) activation

Sulfhydryl inhibitor
reSponse

2 mM iodoacetate
50 uM -chloromercuri-

benzoa e

Mixed substrate studies

0-40% Ammonium sulfate
fraction

40-60% Ammonium sulfate
fraction

Calcium phoSphate gel step
(mixtures of equivalent
amounts of each enzyme)

no effect

88% inhibition
of D-fuconate
dehydration

non-additive
rates with
L-arabonate
and D-fuconate

10% total
activity

90% total
activity

100% inactiva-
tion

no effect

66% inhibition
of D-fuconate
dehydration

non-additive
rates with
D-galactonate
and D-fuconate

90% total
activity

10% total
activity

9% inactiva—
tion

234

Figure 58. Growth of mutant strain 73-1-2 on Dbfucose,

L-arabinose, D-glucose, and Dbgalactose.

 

ABSORBANCE (600 nm)

.900
.800

.700
.600
.500

.300

.200

.100

.090
.080

.070
.060

.050

235

 

 

 

 

Figure 58
_ I I I I A
I—- v' ' VFW
o o
h——- v —
OI o
-— v .—
1o
—— o —e
__ Y; 8, __
. Y
g V
_ .V A
‘v
1. LnARABINOSE
3 V D-GLUCOSE
_ o D-GALACTOSE '—
n— 8 -—
_ V D-FUCOSE __
a B ._
3 3
— . v —D
8 o

V§VVV —

I I I I

5 10 15 20

HOURS AT 32°C

236

Table XXXIX. Enzyme levels in mutant strain 73-1-2 and the
parental strain. Cell extracts were prepared
from L-arabinose grown cells unless otherwise
noted and were prepared as in Table XXXVII..
Protein was estimated by the biuret assay.
Specific activities are reported as umoles

per hour per mg protein.

 

 

Specific Activity

 

 

Enzyme Wild Type Mutant
Dqudohexose dehydrogenase* 16.3 14.1
LnArabino-aldose dehydrogenase 15.7 11.8
D-Fuconate dehydratase 4.27 3,32
KDF Aldolase 4.23 0.91

 

*Cell extracts prepared from D-glucose grown cells.

237
to be a leaky mutant, in which a change in an amino acid
residue in the aldolase has resulted in a less catalyti-
cally active protein.

The isolation of mutant strain 73-1-2, which
simultaneously exhibited a reduced growth rate on L-
arabinose and Dbfucose and a correSpondingly diminished
level of KDF aldolase, supports the prior conclusions
that KDF and KDA are degraded by the same enzyme and that
L-KDA dehydratase, the only otherenzyme reported to act
upon L-KDA (54-56), is not present in this microorganism.

The fact that the two mutant strains, 5-1-10-1
and 73-1-2, possess correSponding defects in both the
L-arabonate and D-fuconate pathways suggests the common
identity of the two pathways and substantiates the pre-
vious conclusion that both D-fuconate and L—arabonate
are dehydrated by the same enzyme and that both KDF and
KDA are cleaved by the same enzyme. In summary, the
enzymes of D-fuconate and L-arabonate metabolism are
identical, but, deSpite the structural Similarity between
D—fuconate and D-galactonate, the enzymes reSponsible
for D-galactonate degradation do not participate to a
detectable extent in the degradation of either D-fuconate

or L-arabonate.

DISCUSSION

The data presented in this thesis show that, in
the pseudomonad under investigation, D-fucose is degraded
by the pathway shown in Figure 59. D-Fucose is degraded
through a modified Emtner-Douderoff pathway by oxidation
to the lactones, hydrolysis to the free acid, dehydration
to the keto deoxy acid, and cleavage to form pyruvate
and Delactaldehyde. The data also indicate that the
enzymes of the D-fucose pathway function in a new path-
way for the degradation of L-arabinose (Figure 60A);
instead of being dehydrated to form c-keto glutarate
semialdehyde, 2-keto-3-deoxy-L-arabonate is cleaved to
form pyruvate and glycolaldehyde. All attempts to demon-
strate the previously known pathway (Figure 60B) were
unsuccessful.

In this section, the major conclusions, and the
data on which they are based, will be discussed, and an
attempt will be made to relate the results obtained from
this work to those from related studies in the literature.

The initial eXperiments indicated the presence of
at least two soluble, pyridine nucleotide-linked dehydro-
genases which were capable of oxidizing D-fucose and
which could be differentiated on the basis of nucleotide

and substrate Specificity and Sephadex G~200 elution pro-

238

239

Figure 59. Pathway of D-fucose degradation

D-Fucose \\\\\\\§
[Pyranose] v\\\\\\[FuranoseJ
NAD NAD(P )
D-Fucono-b-lactone D-Fucono-Y-lactone
D-Fuconate
H20

2-Keto-3-deoxbe-fuoonate

Pyruvate D-Lactaldehyde

240

ZFigure 60. L-Arabinose degradation in pseudomonads.

.A. New pathway

Lparabonate

H20

2~KetO-3-deoxy-L-arabonate

PYruvate / \‘ Glycolaldehyde

B. Previously known pathway (54—56)

Z-Keto-3-deoxy-L-arabonate
H20
c-keto glutarate semialdehyde

NAD(P)

/

I-Nmem
0

c-keto glutarate

 

241
files. NO activity which modified D-fucose by means of
isomerization, phOSphorylation, epimerization, or reduc-
tion was detected in crude cell extracts. Thus, it was
concluded that pathways for D-fucose metabolism other
than the oxidative pathway described above did not exist
in this pseudomonad; it was also concluded that D-fucose
was not degraded by a pathway analogous to those which
occur in other organisms for the degradation of L-fucose
or L-rhamnose, which are the two other common 6-deoxy
aldohexoses (263-265).

Based on substrate specificities of the two fucose
dehydrogenases, the NAD-dependent enzyme may be desig-
nated a D-aldohexose dehydrogenase and the NAD(P)-depen-
dent enzyme may be designated an L-arabinO-aldose dehydro-
genase. Substrate Specificity cannot, however, be used
as the sole criterion in ascertaining the participation
Of a particular enzyme in a metabolic process. The induc-
tion by a Specific carbohydrate of an enzyme which, in
turn, is Operative on the carbohydrate suggests that the
enzyme may be functional in the metabolism of the carbo-
hydrate. Both dehydrogenases were induced by growth on
D-fucose and,in the absence of any mutational evidence
to indicate the contrary, may be considered to be func-
tional in the degradation of D-fucose. The D-aldohexose
dehydrogenase is also induced by growth on D-glucose

whereas the L-arabino-aldose dehydrogenase is also induced

242

by growth on D-galactose, L-arabinose, and 6-iodo-6-deoxy-
D-galactose. The last carbohydrate is readily metabolized
by the pseudomonad and is probably oxidized by the latter
enzyme in_zizg. For the sake Of clarity, the two dehydro-
genases will be distinguished in the remainder of this
section on the basis of substrate Specificity rather than
nucleotide Specificity.

Evidence was Obtained which establishes the products
of carbohydrate oxidation to be a O-lactone with the D-
aldohexose dehydrogenase and a y-lactone with the L-arabino-
aldose dehydrogenase. The ring sizes of the lactones
were determined by chromatography and by the ability of
the respective dehydrogenases to convert 6- and y-lactones
to the corresponding aldose (Tables X and XVII).

The Km values at pH 8.1 for the physiologically
important carbohydrates ranged from 0.86 mM to 5.8 mM for
the D-aldohexose dehydrogenase and from 0.14 mM to 0.50
mM for the L-arabino-aldose dehydrogenase. The Km values
for the latter were also runat the pH Optimum, pH 9.4,
and were found to be 5 to 14-fold higher than the Km
values at pH 8.1. The Observed increase in Km and Vmax
values at pH 9.4 provokes thought about the implications
inherent in the determination of a Km value at only one
hydrogen ion concentration, as is the case for most
enzymes reported in the literature, except those enzymes
which, due to an abnormally high Km value, were investi-

gated to determine the lowest and yet most meaningful Km

243
value (85, 86)-

The dehydrogenases were easily distinguished on
the basis of thermal denaturation. The half-lives at
55°C were found to be 42 seconds and 13 minutes for the
D-aldohexose and L-arabino-aldose dehydrogenases, reSpec-
tively (Figure 23). These properties were used in the
heat step of the purification of the latter enzyme.
Both enzymes were insensitive to thiols, thiol group
inhibitors, metal ion activators, and EDTA.

Due to the fact that pseudomonads are noted for
possession of multiple carbohydrate dehydrogenases, it
was of interest to determine whether the pseudomonad
under study possessed only two dehydrogenases, or if the
activities on these various sugars could be resolved
further. This was investigated by measuring the ratios
of activities on various substrates when the enzymes
were subjected to the following: (i) chromatography on
Sephadex G-200 and DEAF-cellulose: (ii) adsorption on
and elution from calcium phOSphate gel; (111) thermal
inactivation; (iv) assaying at different pH values; (v)
mixing substrates for possible additive rates. In all
cases, the ratios remained constant, indicating that
the L-arabino-aldose dehydrogenase and the D-aldohexose
dehydrogenase are Single enzymes and that there is, in
the pseudomonad under study, no other dehydrogenase
active on D-glucose, D-galactose, L-arabinose, or D-

fucose when the pseudomonad is grown on any of these

244

carbohydrates. An additional dehydrogenase has been
observed in D-xylose grown cells. It appears to be a
D-xylose Specific dehydrogenase which does not oxidize
any of the above four carbohydrates and D—arabinose,
D-glucose, L-fucose, and D-lyxose; this enzyme is not
detectable in cell extracts of the pseudomonad when
grown on carbohydrates other than D-xylose.

The presence of bacterial enzymes which oxidize
carbohydrates to the correSponding lactone is well estab-
lished. Many of these enzymes are soluble and Specific
for non-phOSphorylated aldoses. Table XL lists the
Specificities of various soluble aldose dehydrogenase
that have been found in other bacteria by other workers.

The only enzymes closely related to the unique
D-aldohexose dehydrogenase, herein described, have been
characterized by Cline and Hu (87) and Avigad'gt‘gl. (42).
The NAD-dependent aldose dehydrogenase of Cline and En
exhibited entirely different substrate requirements, was
totally inactive on D-mannose, and was active on D-xylose.
The NADP-dependent aldohexose dehydrogenase Of Avigad
23‘s}. was active only on D-glucose, D-mannose, 2-deoxy-
D-glucose, and D-mannosamine whereas the NAD-dependent
D-aldohexose dehydrogenase described in this thesis was
active on all the D-aldohexoses tested and possesses the
broadest substrate Specificity of any soluble dehydrogen-
ase previously described. Microbial dehydrogenases

operative on D-mannose are a rarity, there being only

245

6H Non comm omdsowOHUhnod

 

 

 

A o: v > Donald . Howls . +32 mososoo30u m ooosa fiendish

. Mflmmmonomooom omosowononsoo

Ac: .smv > nononq .Howus +942 goososooSoon oooSnnonth

. Hmmnm oooSomononswo

Ammo +mq<z .+aaz meeosooSomm ooosapenaws

osnlq ooosowononSoo

name none .nemuq .nenmum +ma<z .+asz omsoaoosoem omoannenarn

oanmmonosoooo omoSowononsoo

Aomv > +de mososoosomm ooosanondta
assets

.SHonQ ossmnoo ooosowononsoo

Amen e .smaun .enmaa +moez noooenosoosno onososoone
nonouq

.osnln .Hnwna omoSowononsoo

flame Inna .snmam .nemna +maz oesoaoosomm omoone

omosomonOhsoo

cum 0 ooowosooaoua +Q<z oososOOSoom omowosooadrn

Donald oooSowononsoo

mum > .Howun .osnln +mm<z .+de oososOUSomm ooooHOIOSanondmA

 

osoooes no noaonnaooam neaoannooam
.nom nososo. snow msnm oponuonsm nopmoood oonsom oshusm
sonPOOHm

lIIII II.

 

mooom I AosOnoo: I omom Ammonnoanooomosos eopoahnosm
Imonnsos non canaoomm momosowonohsoo Hoanopoop oansaoo osoanon no sooanomsoo d. .Ax oases

2M6

.mmoanoaamo .oaamo “omopoma .poaH ”mmopawa

.pama ”mmopomamwlq .Hmwnq ”mmoonmlq .odmtq .mmozanmnmln .nmhwlm
ummannualmlhxomUINnocaadlm .ndadlm ”omonapmhmlq .nmnmnq ”mmozdmaln
.naanm «omoofiusa .osmln ”omoahxla .Hmwsm ”mmopowammnhnooclm .Hmwcln

”mucosamlmnhxoouum .5Hmcum ”mmopodadwlm .Hdwlm ”mmoousIQ .aawlm

.ummu uncapdabonnpd.

 

A0:

Ammv
Ammv
Am3v
A::v
Ammv
Aamv
AoHV
Aumv
as
.mnv

Ammv

IQIm

noun

OHHmo

. p03 53
.Hhuna .samna

samoum .sawnm

Swain .sawnm

sawuan .samnm

sawua

fiawlm

panauq .Hawcum

.osgna .wanm

nahmlq .walm

thwtq .walm

:zonxqs
+nmmdz

+mm¢z

+942
+942
+de
+m©<z
+Q¢z
+Q<z

+mdz

 

mumommopnmm

mfiooOOOHdmm

.mmMWHmmmmm
Ampomnopmad

mcouwmopdm

Hmuodponmod

 

 

 

 

 

muonoo
mafiaaowm

 

canoaovSomm
uwnoaOUSmmm
cQGQEOUSmmm

adammonmnoomm

manoaoUSQmm

 

cascaOdsmmm

, mmdamwOHuhSmu

omoosawum

mmdnmwOHUhsmc
mmoo§H0IQ

mmanowOHUhnmd
omooaaolm

mmmnmwonchnuu
mmoosausn

ommcowonchnod
omooaaulm

mmmnmwOHUhSmu
omoosaolm

mmwnmwonvhsmd
mmooSHolm

mmmnmwOthsmc
mmopomawonm

mmdnmwOHUhnmu
amopomamona

amazemenuhsmc
mmopomdmotn

omdnmwOQUESmc
mmopodadolm

247
two recorded instances in the literature (42, 43); the
D-aldohexose dehydrogenase herein described constitutes
the third instance.

The ring form of the lactone resulting from the
oxidation of D-glucose by the NADP-dependent aldohexose
of Avigad gtugl. was determined to be a 6-1actone on the
basis of reversibility of the reaction only with D—
glucono-é-lactone. The ring form of the product was not
determined by Cline and flu. All soluble dehydrogenase
Operative on D-glucose, for which the lactone ring size
has been determined, yield D-glucono-b-lactone (39, hZ-hh,
93). This is in agreement with the results obtained in
the current study. However, all previously reported
studies of soluble dehydrogenases Operative on D-galactose,
in which the lactone ring size has been determined. have
indicated the product to be a y-lactone. Thus, the D-
aldohexose dehydrogenase isolated from the pseudomonad
under investigation is the first reported instance of
D-fucose or D-galactose oxidation in which the pyranose
form of the carbohydrate is oxidized resulting in forma-
tion of the 5-lactone. D-Fucono-b-lactone and D-galac-
tono-b-lactone have never been isolated due to their
instability (46-48).

.A substantial number of carbohydrate dehydrogen-
ases have been characterized as to whether the configura-
tion of the first carbon atom of the substrate is impor-

tant in the oxidation of the sugar (38, 85, 87, 93. 9#).

248
In all cases examined which are valid under a set of con-
ditions discussed below, the B-D or the correSponding
d-L form Of the carbohydrate was determined to be the
more reactive anomer. It is felt that this type Of
analysis must be restricted to those carbohydrates which
exhibit simple mutarotation (95), that is a single slow
mutarotative step solely involving pyranose interconver-
sions; included in this series would be D-glucose, D-
mannose, D-xylose, D-glucose and the disaccharides con-
taining these carbohydrates in the reducing moiety. The
carbohydrates which exhibit complex mutarotation, that
is those which form substantial amounts Of furanose forms
in solution, cannot be used in an anomeric prefence
analysis with certainty since other unknown rate parameters
are introduced into the analysis; these carbohydrates are
D-galactose, D-fucose, L-arabinose, and D-talose (95).
A further obvious restriction must necessarily involve
the utilization of only those dehydrogenases which are
Operative on the pyranose form Of the carbohydrate.

In compliance with the previous restrictions, the
investigation Of the steric requirements Of the D—aldo-
hexose dehydrogenase with a-D-glucose and d,B-D-g1ucose,
showed that the B-D-pyranose form was preferred over the
d-D-pyranose form. This is in agreement with the results
Obtained with other soluble dehydrogenases. The L-arabino-

aldose dehydrogenase was not similarly investigated due

2#9
to the fact that the enZyme oxidizes only those carbohyd-
rates which exhibit complex mutarotation and also due to
the fact that enzyme appears to require the furanose form
for catalysis.

The L-arabinO-aldose dehydrogenase is also a
unique enzyme and is not easily insertable into any pre-
viously established class Of soluble carbohydrate dehydro-
genases due to its substrate and nucleotide requirements.
It is one of the rarer carbohydrate dehydrogenases Opera-
tive with both NAD and NADP; it also possesses stringent
substrate requirements and is Operative on only those
carbohydrates possessing the L-arabino—oonfiguration.

The product Of this dehydrogenase has been identified as
the y-lactone, which is in agreement with results Obtained
with other soluble dehydrogenases Operative on D-galactose.
Since the y-lactone is the product, it can be assumed that
the enzyme is Operative on the furanose form of the carbo-
hydrate.

The relatively high content Of furanoses in solu-
tion Of some carbohydrates has been documented. The rate
of mutarotation in aqueous solutions Of galactose, arabin-
ose and, presumably, fucose is substantially faster than
the rate Of mutarotation Of mannose or glucose as was
pointed out earlier; this has been interpreted as an
indication that substantial amounts of the furanose forms

Of the former series are present at equilibrium (96).

250
The prOportion of furanoses in an equilibrium solution
depends upon the relative stabilities Of the pyranose
and the furanose forms. The stability Of the furanose
forms, likewise, depends on whether the substituents
are gig or trans to each other; when the three substitu-
ents Of the 0-2, 0-3, and C-5 are in a Eraggfitrgng
arrangement, the furanose form is quite favorable, and
its presence is indicated by fast mutarotation. From a
detailed mathematical treatment Of the mutarotation of
D-galactose, Nelson and Beegle (53) postulated that an
equilibrium Of galactose contains 12% furanose at 20°C.
A more recent study Of the equilibrium composition Of
L-arabinose and D-galactose solutions at 25°C by gas
chromatography Of the trimethylsilyl ethers indicated.
that the B-furanose form amounts to 5.3 and 4.9% reSpec-
tively; the B-pyranose to B-furanose conversion Of galac-
tose was noted to be more rapid than the B-pyranose to
the a-pyranose, substantiating in part, the previous
interpretation Of fast versus slow mutarotation rates (97).
Wells, Sweeley, Bentley. and coworkers have demonstrated
also by gas chromatography Of the trimethylsilyl ethers
that equilibrium solutions Of D-galactose contained 5.4%
furanose (98). Since structures can be drawn for D-
galactofuranose, D-fucofuranose, and L-arabinofuranose
which exhibit the allgggggg configuration (Figure 61), it

is probable that the L-arabinO-aldose dehydrogenase is

251

Figure 61. ‘All-trans configuration Of B-furanose forms

Of D-fucose, D-galactose, and L-arabinose.

 

 

 

 

 

NAD(P) NAD(P)H O
x /
, \OH H 0
LnArabinO-aldose __,
dehydrogenase H on
HCOH HCOH
l l
B B
R=H : d-L-Arabinofuranose LnArabonO-Y-lactone
B=CH20H : B-D-Galactofuranose DbGalactonO-Y-lactone

B=CH3 : B-D-Fucofuranose D-FuconO-Y-lactone

252

also Operative on the B-anomer. It is not evident, how-
ever, why enzymes would be evolved which are Operative on
B-D—galactofuranose rather than B-D-galactopyranose,
present in 15-fold excess or d-D-galactOpyranose, present
in a 6-fOld excess. There exists the possibility that
the LparabinO-aldose dehydrogenase possesses the capabil-
ity Of converting B-pyranose, present in 15-fold excess,
to a B-furanose in a manner similar to a mutarotase.

Since D-fuconO-Y-lactone is very stable and hydro-
lyzes slowly even at pH 9 (#8) and since 20% of the y-
lactone formed from oxidation Of D-fucose by the L-arabino-
aldose dehydrogenase (Table V) was converted to the free
acid by hydrolysis, it was evident in the preliminary
eXperiments that an aldonolactonase was present. Aldono-
lactonases are a poorly characterized group Of esterases
which are Operative on internal esters Of aldonic acids.
They are a necessary group Of enzymes for the hydrolysis
Of those lactones which are stable towards non-enzymatic
hydrolysis in the 6-8 pH range. Weimberg (88) has
Observed that Pseudomonas £325; is incapable Of efficiently
utilizing L-arabinose, D-xylose, and D—ribose because of
the absence Of a lactonase; the Y-lactones of the reSpec-
tive aldoses were found to accumulate in the media due to
their stability at a neutral pH. Due to the stability Of
D-fuconO-y-lactone, the growth rate Of the pseudomonad
under study would probably be diminished in the absence

Of a lactonase.

253

A literature survey was undertaken to correlate
the properties Of various delactonizing enzymes with the
prOperties Of the lactonase isolated from the pseudomonad
under study. The results Of the survey are presented in
Table XLI.

The general properties Of aldonolactonases can be
summarized as the following: (1) they are usually Speci-
fic for 6- or y-lactones; (ii) they require a metal ion
and a thiol for maximum activity; (iii) the pH Optima are
in the range 6-7.5: and (iv) they all possess high
Miohaelis constants, as noted also by Yamada (103).

The enzyme partially characterized in this study
was not activated by thiols or metal ions, nor was it
inhibited by high concentrations of thiol group inhibi-
tors. The Km value and the pH Optimum were found to be
21 mM and pH 7.6 reSpectively. There appear to be no
striking differences between the prOperties of this lac-
tonase and the prOperties of the previously described
lactonases.

The degradation Of D—fucose to the keto deoxy acid
provides another example Of a wideSpread metabolic mechan-
ism in bacteria whereby utilization Of carbohydrates and
polysaccharides involve formation of deoxy sugar acid
intermediates. 2-KetO-3-deoxy sugar acid intermediates
are involved in the degradation Of glucose (19, 125),
galactose (78), D-altronic and mannonic acids (106),

D-arabinose (117), L-arabinose (59-56), D-xylose and

 

 

 

 

5.0 Aolpomqv ¢H0|>IQ .l economa
AHOHV nm.w 0:02 2s om Hmoa>uo .pdmu>na mooaoobsmw .m nonpoaq
pomquo
+moo Asaousuov .saououm1ouo ommnopoaq
Amav we» +mcz .+Nmz :5 m.m .saououo paces m.noaam sHannO
. .+Noo Ommzopodq
Aooflv o o +mzz +~mz saououo umpda mea>om saouono
madannmsooflw Ommzopowq
Anny .m nmn4r>aq
U.
M0 Ommzopomq
Adsv nm>aa psm Haou>um
.l OmGCOpOmA
Amwv Hmong .m pandr>lm
mwanaoamsoQ£M Ommsopomq
Aomv .m pmn4r>uo
waasmonm£oommv ommcopodq
awn. v . m **H.m0l>lm
05M|>IQ

Onoz as om namu>nm .Hmun>lm cocoaoozomm ammonopooql>
psmaOaHsuOm “mommaoaommv

.cmm .pgo mm Hodge eOH Hmpmz am mpdoacaomom monsom manage

i

 

.mOmOQOpOmHOQOOHO msOHHOP no nomanmaaoo 4. .qu canoe

255

.Osopodaocoonm .Ozm ”OnOpomaocoasw
.Hso ”economaononzmz .nmz “Onopowaononam .pam nOnopomaozoofiflmlosnmosm1w .5Holm10
”Odom camoanopomq .pomq .Onopomaonoosdo .us "economaonondnd_.nwnd Monopomaozopomamo .Hmo**

ommHOHuhm *

 

 

 

”muoapodboannd
OmOHOHOhs
meaaonmpsn
aonasspma
Amoav m.n Osoz 21 mm osmomOHOOHm (M amHOnndol>
afloa>nq umpda
Amoav .5HUI>:Q .Houu>nm cmaamaaoz HH omonopomq
nm2|>lm _
s ”sHOuHuO ”saouwnq secede swaps
0. H60! ID Ham] IA USN Hobaa
0H Im. new Q2 2 SE o la HSOI JD H50! IA :ddHMESdz H mmmnopodq
A m V w +N . +Nm N . > . >
sHou>uo .nmn¢r>uo secede
.A30H .+NON ”Howarun HDHMI>IQ use nobda Omanopoma
moav mm» +mez +moo as oouoa asun>uo zaza>no cadamssmz nonoead

256
D-ribose (88, 119), D-glucosaminic acid (107), alginic
acid (61), polygalacturonic acid (63, 116), hyaluronic
acid (108), chondroitan sulfate (115), pectin (109, 110),
glucaric and galactaric acids (18, 26, 111, 112, 119),
mygyinositol (113), and y-hydroxy glutamate (114).
Z-Keto-3-deoxy-D—arabinO-heptonic acid-7-phOSphate (17,
59, 121), 2-ketO-3-deoxy-D-mggng-Octonic acid-8-phOSphate
(1a, 75, 118, 120), N-acetyl-neuraminic acid-9-phosphate
(122), and N-acetyl-neuraminic acid (122, 123), also
belong to the group Of keto deoxy sugar acids but are
involved in biosynthetic reactions and are formed by aldol
condensation of two smaller moieties. The Z-ketO-B-deoxy
sugar acids which contain 5 and 6 carbon atoms are formed
from their respective aldonic acid by the action Of a
Specific dehydratase.

The quantitative conversion Of D-fuconate by cell-
free extracts to a substance which after periodate oxida-
tion yielded a TBA chromogen with xmax 551 nm suggested
that the substance was a 2-ketO-3-deoxy aldonic acid and,
thus, directly implicated the intervention of a dehydra-
tase. The dehydration product was shown to possess the
structure 3,6-dideoxy-Dgthggg-hexulosonic acid (KDF) by
analysis of the condensation products Of the dehydration
product with semicarbazide-H01, Orphenylenediamine, and
3-methyl-2-benzothiazOlinone hydrazone,by ceric sulfate

decarboxylation, by periodate rate and total consumption

257

studies, by analysis Of the KDF aldolase cleavage products,
by identification of the periodate cleavage products, and
by chemical synthesis. Dehydration was found to occur at
C-2 and C-3 leaving the C-4 and C-5 hydroxyl groups intact.

LnArabonate was also dehydrated by the same enzyme
to a substance which, after periodate oxidation, yielded
an alkalei-unstable TBA chromogen with Amax 551 nm. The
common identity Of the enzymes which dehydrated L-arabon-
ate and D—fuconate was indicated by a variety Of means,
among which were the following: (i) Sephadex G-200 and
DEAE-cellulose elution profiles and the pH Optima Of the
enzyme were superimposable; (ii) rates from studies
utilizing mixed substrates were not additive; (iii) ther-
mal inactivation profiles of the enZyme with both sub-
strates were superimposable and exhibited first order
kinetics; and (iv) a mutant strain was isolated which
lacked dehydratase activity on L-arabonate and D—fuconate,
and a revertant was Obtained from the mutant strain which
regained the ability to utilize D-fucose and L-arabinose
and possessed normal dehydratase activity on L-arabonate
and D-fuconate. It was concluded that the Observed
activities were due to a single enzyme.

An enzyme reported to dehydrate L-arabonate has
been found to occur in Pseudomonas saccharophila by
Weimberg and Doudoroff (5#) and Weimberg (55), and in

Pseudomonas fragi by Weimberg (88). Weimberg and

258
Doudoroff were unable to purify and characterize the
g. saccharOphila enzyme due to its extreme lability: in
contrast, the dehydratase isolated from the pseudomonad
under study is very stable and retains 100% activity after
24 hours at 30°C, whereas the g, saccharophila enzyme is
totally denatured after 5 minutes under the same condi-
tions (54). Thus, this is the first reported purifica-
tion and characterization Of a dehydratase active on
L-arabonate and D-fuconate. In 3, 3325;, Weimberg only
reported the appearance Of TBA-positive material when
crude extracts were incubated with L-arabonate.

A literature survey was undertaken to correlate
the properties Of other aldonic acid dehydratases and
the results Of this survey are presented in Table XLII.
The properties exhibited by most Of the dehydratases can
be listed as the following: (i) dehydration occurs at a
position a,B tO the terminal carboxyl group: (ii) the
configuration Of the hydroxyl groups at the dehydration
site is usually thrgg (the exceptions are Damannonic
dehydratase and transeliminase which dehydrate erythro
hydroxyl groups): (iii) a metal ion is usually required
for maximal activity (the exception is D-galactarate
dehydratase): (iv) most dehydratases requiring metal
ions also require thiols (the exception is D-mannonic
dehydratase): (v) the pH Optimum ranges from 7-8; (vi)
the Km's are 0.1-8 mM: and (vii) the dehydratase reac-

tion appears to be irreversible.

 

280 .MV

 

 

 

 

 

 

 

 

 

omdnonvhsoo
Aboav :8 c on m.m +quA+moo OHSDNHO Odom canonzdacn opoconzozlo
3.80 .9
omOpmHOhsOo
Aboav boaHH as m mom m.m +Nom cons» Odom canonpamln opmnonpadtm
Amnoomonosay {My
+Nwzx+mom omdwOHOhsoo
Aoma «+mnza+mwz opmnoondw opmzooaaw
.mmi >33 as 0 mm» o.m..n.s 4+mnza+mmm 8.2» noscmosmam noncmosmsm
Ammamdbonoo {my
ommpmnpmsoo
nyammav :.nlm.u +Nwz : Odom hwohOthQ
5 II.
2 39m 380 .m:
+~cz OaHOHmb Hampoa oompmhomzoo
Amway mom o.num.u «+mwza+~om uawowemsdeua.s .= chow awonemsam
mumpomaopoom Anomnanmv
.anm OanHmbomH OmOpmHOmsoc
Ammav o: «.muo.m consanwz omen» swonesnaeum.s edom awoneasdo
+Nom opdzonmndra omOpOAOhnoo
Ammav both 28 d mom :.m:a.m 4+N224+Nwz OOHSp oudfioofihnm opmnoonmlm
.dmm whopmbdpod opam :OHpmHOh20Q
.mom mom am Hodna .pmo ma zOH Hmpoz 63% pm MQSOHO mpmhpmnsm mofihusm
H Mono m go
noapmndwahzoo

2
a

mommpmncmzmo msodhmb mo somahmnaoo .HHAN OHQOB

 

 

 

 

Amway

“mafia
wines
2

Aaov
“anal
Amaze

AHHHV

Aomv

as ma.

28 m

28 :.

bonhH 28 m.

21 on

no» N.wuo.o

m.mum.m

m.s

mom m.m
on o.m

mm» m.mnm.m

mom m.n

+mn2 4+~

+~ao +~

+N

0C

00

:2

mm

03

00
ma

 

cons»

 

owns»

 

owns»

Ohspuyo

 

coax»

 

omen»

 

cons»

Odom canopmndtq

Odom canonmhdln

O “coca

Ia m open
I . opamona
mo 8 opoMIN

Udod

IIIIIIhMHamm

loudnoomm .mv
ommpdnomsoo
opwmonmwdrq

.Illlllhmaamn

Ohmnoomm .mv

Omdnmnumnoc
mpmnopondta

 

Amonmwonom
Honomnohodv
ommpdhomnoo

Hopdmocd opom

 

AochaOudommv

mamas
OHOpomHmmeOm ommsaaaflo
OH:

Uaom OHCAde

Odom oanmpnda

Odom
oahmpomamwln

mumnmoaum
.Opdnmosawln

anosommv
Add: mmmzawdd

Amedpsm .mv
omdpmHUhSOO
opmnmphda

 

“Hoe {we
Omwpmwomgoo
opmampomaooum

 

:8 .mv
omMdeOhnoo
opdhdosaoua

“mammm

Ionmsooom .mv

«om «manageasoo «OM

261

In comparison to the previously characterized
dehydratases, D-fuconate dehydratase does not exhibit
any unusual prOperties. Dehydration occurs at thrgg
hydroxyl groups at C-2 and C-3, and a divalent cation
and thiol are required for maximal activity. The pH
Optimum is 7.1-7.h, and the dehydration is an essen-
tially irreversible reaction.

D-Fuconate dehydratase activity is influenced
by the buffers used in the assay. The enzyme is par-
ticularly susceptible tO Tris and the Tris derivatives
prepared by Good.gt_g;, (129). It is interesting to
note that the crude labile enzyme from.§, saccharophila
also exhibited denaturation by Tris (54): buffer effects
on dehydratases have not been described except for the
previous report. Inhibition of D-fuconate dehydratase
by the buffers presented in the text is not believed to
occur by chelation Of ionic cofactors since no correla-
tion exists between the pKM2+ (metal ion binding con-
stants) and the Observed activities in the various buf-
fers.

D-Fuconate dehydratase is induced by D-galactose
(Table XXVII), but is totally inactive on D-galactonate.
D-Galactose may thus be considered to be a gratuitous
inducer of D-fuconate dehydratase. This presents an
interesting situation in that D-fucose is a gratuitous
inducer Of the galactose operon in other organisms (132,

133), whereas in this instance, galactose is a gratuitous

262
inducer Of an enzyme instrumental in D—fucose degradation.

.A mutant, strain 5-1-10-1, was isolated which was
unable to utilize Dbfucose or L-arabinose and was shown
to lack Dbfuconate dehydratase but possessed normal levels
Of D-galactonate dehydratase. This evidence Justifies
the conclusions that D-fuconate dehydratase is absolutely
functional in the metabolism Of D-fuconate and L-arabon-
ate, that D-galactonate dehydratase is not functional in
the metabolism Of D-fuconate, and that there is little
likelihood that this organism possess an alternative
route for D-fucose degradation, a conclusion reached
previously after the inability to detect other enzymes
Operative on D-fucose.

The further degradation of deoxy hexulosonic acids
is characterized by one Of three routes: (1) by dehydra-
tion (56); (ii) by direct cleavage (18, 117, 119, 122,
123, 134, 135): or (11) by cleavage Of a phOSphorylated
intermediate (19. 77-79). An aldolase was discovered
which was induced by growth on L-arabinose or D-fucose
and which catalyzed the cleavage of KDF and KDA. KDF
aldolase was purified over 50-fold and some Of its
properties determined.

The cleavage products of KDF and KDA were deter-
mined tO be pyruvate and lactaldehyde and pyruvate and
glycolaldehyde.,reSpectively. Equivalent amounts Of each

product were formed from the respective substrate.

263
Pyruvate, and not hydroxypyruvate, was shown to be the
product Of the KDF aldolase-catalyzed cleavage of KDF and
KDA by chromatography and absorption Spectra Of the 2,4-
dinitrophenylhydrazones.

A literature survey was undertaken to correlate the
sulfhydryl requirements, the pH Optima, and the metal ion
requirements of aldolases which cleave deoxy hexulosonic acids
and is presented in Table.XLIII.

In contrast to the other aldolases in Table XLIII
KDF aldolase has an absolute requirement for a divalent
cation. The only other aldolases Operative on a deoxy
hexulosonic acid which possess.' a metal ion requirement
are 2-ketO-3-deoxy-D-glucarate aldolase (18) and 2-ketO-3-
deoxy-D-arabino-heptonate-7-phoSphate synthetase from
sweet potato (121).

The pH Optimum for KDF aldolase was found to be
8-8.2 in Tris and Hepes buffers with each substrate. All
similar aldolases possess pH Optima in this range where
the substrates are completely ionized. The requirement
for metal ions in the presence of ionized substrates
suggest the formation of an enzyme-M2+-substrate chelate
is necessary for catalysis to occur.

KDF aldolase in the assay reaction has no sulf-
hydryl requirement as is evident from the absence Of
stimulation in the presence of thiols or from the absence

of inhibition in the presence Of thiol group inhibitors.

264

 

 

 

 

AHNHV «.5 on mom opapon poosm omdpospzhm opennmonn
Inlopdnopnoslonapmnm
Inlhwoovlm:0nomnm
omgonpflmm opowgm ose
. . .l Imlopanpnonlosanmha
Ammv :.ml:.o o: o: 0: «H00 .m .Injhwooutm3opOMIN
.I OmOHOOHm opdsopodamwosm
Anny m.o mHasmOnwsOomm .m Imonnnouhwooonmaopomum
.l ommaooam opOSOOusonm
Anny m.o OHaSMmhdsoomm .m ImosnnmuhwooOIMI090M|N
Aomfl .tl .I OmOHOOHs opmsoosawomn
.mNHV o.w|o.n on on o: msoomonosam .m Imonmlwlhwooosm3090MIN
modaouao encampaaw
Ammav m.m mos pswdam nmpda pas nopoausnswouesmu>
odeOOHO onshapsaw
Aimav o.mnm.m mom on o: nobda pom IopoxldthOHchmlr
ommaouam oposoosu
«.mno.w on we» mom Odsoaoosomm :QIhNOOOImlopoMIN

.hom .990 mm donasuom madam: mpnmwd oohfiom Oahunm

onsoaaoo pnoamban msapmaoso
mcdozeom an .pdpoq an .pdan

%

.mOHos hwooo opox Obwoao Seas: mommaooam msodhmb mo somanmnaoo

.HHHAN OHDQB

265

 

“may

Ammav

ANNHV

Anmflv

nomad

©.ml:.u

N.m

«.5

OS

mom

we» won

me»

pswdam

 

hmbdd pam

ameoaa mom

msomzanmhom .lm

mmosammaom 2M

Omdaoudm ondnmodaw
Inuhwooolm10pomam

ommpospnhm
momsnmosmue show
canasdnnonahpoodlz

omoaooam Odom
canasdwsOsampoodtz

OmmHOOHO Odom
oazasoasosampoodlz

composushm opmsgmonm
Imuuaoo cacopOOtOssma
unaswooe1maopoaum

 

266

However, when extracts were prepared in the absence Of
thiols and fractionated on Sephadex G-200, a time— and
thiol-dependent renaturation process was Observed to
occur. It is suggested that thiols prevent the disso-
ciation or denaturation Of KDF aldolase by reducing
sulfhydryl groups necessary for the maintenance Of con-
formational integrity.

The equilibrium constants for the KDF aldolase-
catalyzed cleavage of KDF and KDA were determined to
be 0.12 and 0.38 mM reSpectively. The position of
equilibrium favors the synthesis Of KDF or KDA at 10"3 M,
and thus is not in agreement with the findings Of other
workers on aldolases of non-phosphorylated deoxy hexul-
sonic acids. Aldolases Operative on non-phOSphorylated
deoxy hexulosonic acids possess equilibrium constants
down to 10-2 M (111, 112, 118, 122, 123, 134, 135);
aldolases Operative on phOSphorylated deoxy hexulosonic
acids possess equilibrium constants in the range 10'3-
10'5 M (77, 129, 130) which are similar to the aldolases
which cleave vicinal hydroxyl groups (136, 137) and to
KDF aldolase. The results Obtained in this study pro-
vide an exception to the conclusion Of Fish (18) that
all aldolases which cleave non-phOSphorylated deoxy
hexulosonic possess equilibrium constants that favor
cleavage.

The aldolases listed in Table.XLIII also possess

267
strict substrate requirements. This high degree Of
Specificity has also been found for KDF aldolase which
cleaves only KDF and KDA. Neither 2-keto-3-deoxy-D—
galactonate nor 2-keto-3-deoxy-D—gluconate or the
phOSphorylated derivatives served as an inhibitor of the
cleavage reaction.

KDF aldolase has been shown to be distinct from
previously reported enzymes for the degradation Of D-
galactose, D-glucose, and Lparabinose. In addition to
the conclusion that the aldolase was the terminal enzyme
in the first known pathway for D-fucose metabolism, it
was concluded, on the basis Of mutant strain 73-1-2,
which simultaneously possessed a reduced growth rate on
L-arabinose and D-fucose and a correSpondingly diminished
level Of KDF aldolase, that KDF aldolase was instrumen-
tal in the degradation Of L-KDA. The only previously
known pathway for L-KDA degradation involved its dehydra-
tion by L-KDA dehydratase to form c-keto glutarate semi-
aldehyde (56). L-KDA dehydratase could not be demon-
strated in the microorganism under investigation, and it
has been concluded that a new L-arabinose pathway has
been discovered.

The fact that the two mutant strains, 5-1-10-1
and 73-1-2, possess corresponding defects in both the
L-arabonate and D-fuconate pathways suggests the common

identity of the two pathways and substantiates the pre-

268
vious conclusion that both Dbfuconate and L-arabonate
are dehydrated by the same enzyme and that both KDF and
KDA are cleaved by the same enzyme. In summary, the
enzymes Of D-fuconate and L-arabonate metabolism are
identical, but, deSpite the structural Similarity
between D-fuconate and Dbgalactonate, the enzymes
responsible for D-galactonate degradation do not par-
ticipate to a detectable extent in the degradation Of

either D-fuconate or L-arabonate.

SUMMARY

The first known biodegradative pathway for D-fucose
has been elucidated. The pathway was determined to be:
D—fucose ____9.D-fuconO-Y-lactone + D-fuconO-b-lactone
——-) D-fuconate -——-) 2-keto-3-deoxy-D—fuconate ——9
pyruvate + lactaldehyde. The enzymes were purified and
some Of their prOperties determined. The metabolic
intermediates Of D-fucose degradation were isolated and
identified by derivatization and chemical synthesis. The
enzymes Of the D-fucose pathway were found to act upon
L-arabinose and were found also to function in the bio-
degradation Of L-arabinose as demonstrated by mutant
strains which possessed simultaneous defects in both L-
arabinose and D-fucose metabolism. A.new pathway for the
degradation Of L-KDA was demonstrated through which L-KDA
is cleaved by an aldolase to form pyruvate and glycolalde-
hyde.

269

(1)

(1A)

(2)
(3)

(4)

(5)

(6)
(7)

(8)

(9)

(10)

(11)

(12)

(13)
(14)

(15)

BIBLIOGRAPHY

Dahms, A. S., and Anderson, R. L., Fed. Proc.. §§,
540 (1969). ""

Sumner, J. B., and Howell, S. F., g, Biol. Chem.,
108. 51 (1935).

Hestrin, 3.. _._J_. Biol. Chem.. 180, 249 (1949).

 

 

 

Bean, H. C., and Porter, G. G., Anal. Chem.,ll,
1929 (1959).

Koch, R. B., Geddes, w. F., and Smith, F., Cereal
2112.12” 2.8. 424 (1951). ——

Travelyan, w. E., Proctor, D. P., and Harrison, J. A.,
Nature, 166, 444 (1950).

Warren, L., Nature, 186, 237 (1960).

Warburg, 0., and Christian, w., Biochem. §,, 310,
384 (1941).

Gornall, A. G., Bardawill, C. J., and David, M. M.,
:1. £322;- 2.1222” .112. 751 (1949).

Dische, Z., and Borenfreund, E., g. Biol. Chem..

$25; 583 (1951).

Friedemann, T. B., and Haugen, G. B., Q. Biol. Chem..
142. 415 (1943).

Sayne, P. W., and Greenberg, D. M., g, Biol. Chem.,
232, 787 (1956).

Itagaki, B., and Suzuki, 8., g. Biochem. (Tokyo), 56,
77 (1964).

Meister, A., g, Biol. Chem., 122, 309 (1952).

Ghalambor, M. A., Levine, E; M., and Heath, E. G.,
g. Biol. Chem., 241, 3207 (1966),

Paz, M. A., Blumenfeld, O. 0., HOJkind, M., Henson,
E., Furfine, C., and GallOp, P. M., Arch. Biochem.

mic '__Soa L029 548 (1965).

270

(16)

(17)

(18)

(19)

(20)
(21)

(22)

(23)

(24)

(25)

(26)

(27)

(28)

(29)
(30)
(31)
(32)

(33)
(34)

271

Sugimoto, K., and Okazaki, H., g. Biochem. (Tokyo),
_6_§,. 373 (1967).

Weissbach, A., and Hurwitz, J., g, Biol. Chem., 224,
705 (1959).

Fish, D. C., Ph.D. Thesis, University of Michigan,
Ann Arbor, Michigan, 1964.

Mac Gee, J., and Doudoroff, M., g, Biol. Chem., 210.
617 (1954).

Soda, K., 553. Biol. Chem. (Tokyo), 21, 1054 (1967).

Freundenberg, K., and Hixon, H. M., Chem. Ber., 56,
2119 (1923).

Schmid, H., and Karrer, P., Helv. Chim..Acta, 22,
1371 (1949).

Raymond, A. L., and Schroeder, E. F., g, Amer. Chem.
8°C.. 0, 2785 (1948).

Moore, 8., and Link, K. P., g. Biol. Chem., 133, 293
(1940).

Evans, W. L., Edgar, H. H., and Hoff, G. P., g.
Amer. Chem. Soc., 48, 2668 (1926).

Fish, D. C., and Blumenthal, H. J., Methods Enzymol..
2n 53 (1966).

Waravdekar, U. 8., and Saslaw, L. D., g, Biol. Chem.,

323. 1945 (1959).

Wolfrom, L., and Thompson, A., Methods Carbohydrate
Chem.,‘;, 121 (1962).

Mayer, W., and Tollens, B., Chem. Ber., 8, 3021 (1905).
Votocek, E., g. Chem.,.g, 803 (1900).
Tollens, B., and.Muther, B., Chem. Ber., 21, 306 (1904).

Levvy, G. A., and Mc Allan, A., Biochem.‘g..‘§Q. 433
(1961).

Clark, E., g, Biol. Chem., 4, 65 (1926).

Dimfier, B., and Link, K. P., g. Biol. Chem.. 150. 345
19 3 .

(35)

(36)

(37)

(38)

(39)

(4O)

(41)

(42)

(43)
(44)

(45)
(46)

(47)

(48)

(49)
(50)

(51)

(52)
(53)

272

Sapico, V., Hanson, T. E., Walter, H. M., and Andersondc
H..L., g. Bacteriol., 26, 51 (1968).

Mayo, J. W.. Ph.D. Thesis, Michigan State University,
East Lansing, Michigan, 1968.

Simpson, F.,J., Wolin, M. J., and Wood, w. A., J,
Biol. Chem., 230, 457 (1958).

Strecker, H., and Korkes, S., 6, Biol. Chem., 196,
769 (1952).

Lee, C. K., and Dobrogosz, w. J., g, Bacteriol.. 22,
653 (1965). "

Doudoroff, M., Contopoulou, C. B., and Burns, 5..
Proceedings 23 the International S§mposium‘66 Enzyme
Chemistré, Tokyo and Kyoto, I957, cademic Press,

N0 0' 9 8. pr, 3130

Cuatrecasas, P., and Segal, 8., g. Biol. Chem., 241,
5904 (1966).

Avigad, G. A., Alroy, Y., and Englard, 3.. g, Biol.
Chem., 24}, 1936 (1958).

Okamoto, K., g. Biochem. (Tokyo), 55, 348 (1963).

King, '1'. E., and Cheldelin, V. H., Biochem. ._I_.. 68,
31 p (1958). ““""

Weimberg, H., Biochim. Biophys..Acta, 61, 359 (1963).
Jermyn, M. A., Biochin. BiOphys. Acta, 51, 78 (1962).

Levene, P..A., and Simms, H. S.,.g. Biol. Chem., 65,
31 (1925). '" " "'

Levoy, c. A., McAllan, A., and Hay, A. J., Biochem.
go. 323. 225 (1962).

Wood, w._A.. Methods Enzmolu g. 212 (1955).

Sober, H. A., Gutter, F. J., Wyckoff, M..'g..Amer.
Chem. Soc., 26, 756 (1965).

Cori,)o., and Lipmann, F., g. Biol. Chem., 195, 417
1951 . -

Lien, 0. G., Anal. Chem., 5;, 1363 (1959).

Nelson, J. M., and Beegle, F. M., 6,.Amer. Chem. Soc.,
3;, 559 (1919).

(54)

(55)
(56)

(57)

(58)

(59)

(60)

(61)

(62)

(63)

(64)

(65)

(66)
(67)

(68)

(69)
(70)

(71)
(72)

273

Weimberg, H., and Doudoroff, M.,‘g. Biol. Chem., 212,
607 (1955). w

Weimberg, B., g. Biol. Chem.. £24. 727 (1959).

Stoolmiller, A. C., and Abeles, H. H., 6. Biol. Chem.,
241, 5764 (1966).

Patton, S., and Kurtz, G. w., 6. Dairy Sci., 4, 669
(1951 .

Jennings, w. G., Dunkley, w. L., and Heiber, H. G.,
Food Hes., £6, 13 (1955).

Srinivasan, P. H., and.Sprinson, D. B., Q. Biol.
Chem., 224, 716 (1959).

Tanning, M. C. and Cohen, S., 6. Biol. Chem., 189,
189 (1951).

Preése, J. and Ashwell, G., g. Biol. Chem.. 252, 309
19 2 ,

Preéss, J. and Ashwell, G., J. Biol. Chem., 252, 317
(19 2 .

Preése, J. and Ashwell, G., g. Biol. Chem., 228, 1571
(19 3 . '

Preéss, J. and Ashwell, G., g. Biol. Chem., 228, 1577
19 3 .

Clagss D., Biochem. Biophz . Res. Commun.. 32. 745
19 5 .

Warren, L., g. Biol. Chem., 323, 1971 (1959).

 

 

 

Cornforth, J. V., Firth, M. E., and Gottschald, A.,
Biochem..g.,‘6§. 57 (1958).

Zagalak, B., Frey, P. A., Karabotsos, G. L., and
Abeles, H. H., g. Biol. Chem., 241. 3028 (1966).

Huff, E., Anal. Chem., 5;, 1626 (1959).

Huff, E., Rudney, H., g, Biol. Chem., 224, 1060
(1959).

Huff, E., Biochim. Biophys..Acta, 66, 506 (1961).
Van Eys, Jan,‘g. Biol. Chem., 226, 1531 (1961).

(73)

(74)
(75)

(76)

(77)

(78)

(79)

(80)

(81)
(82)

(83)

(84)

(85)

(86)

(87)

(88)
(89)

(90)

(91)

271+

Gupta, N. K., and Robinson, W. G., g, Biol. Chem.,
225, 1609 (1960).

WOhlg Ac. Chemo we. LL16; 3599 (1908).

 

 

Herschberger, C., Davis, M., and Binkley, S. B.,
in 2121-.Esss-o.£&3o 1585 (1968)-

Samuelson, 0., Ljungquist, K. T., and Parck, C.,
Svensk.Papperstidn., 6;. 1043 (1958).

Shuster, C. W., and Doudoroff, M.,.Arch. Mikrobiol..

De Ley, J. and Doudoroff, M., 6, Biol. Chem., 221,
745 (1957).

Kovachevich, R. and Wood, W. A., g. Biol. Chem., 212,
757 (1955)-

Palleroni, N. J., and Doudoroff, M., 6, Biol. Chem..
218. 535 (1956).

Berlet, H. H., Anal. Biochem., fig, 525 (1968).

 

Dancis, J., Hutzler, J., and Levitz, M., Biochim.
BiOphys. Acta, 8, 85 (1963).

Hawary, M. F. S., and Thompson, H. H. 3., Biochem. 6,.

Powers, H. H., Tabakoglu, G., and Sable, H. Z.,
Biochem. Prep., 3, 56 (1955).

Metder, H. P., Wilcox, S. S., Wick, A., 6. Biol.
9.11211.” 22- 1769(1964). """"

Metder, H. P., Wilcox, S. S., Wick, A. N., 6. Biol.
Chem., 240, 2767 (1965).

Cline, A. L., and Hu, A. s. L., 6. Biol. Chem.,'§&Q,
4493 (1965).

Weimberg, H., g. Biol. Chem., ggé, 629 (1961).

Walgenfels, K., and Kurz, G., Biochem. §,. 225. 559
19 2 .

Cline, A. L., and Hu A. S. L., Biochim. Biophys.
Acta, 25, 237 (i964):

Bach, J., and Sadoff, H. L., g. Bacteriol., 62, 699
(1962).

(92)

(93)
(94)

(95)

(96)

(97)

(98)
(99)
(100)
(101)
(102)
(103)
(104)
(105)
(106)
(107)

(108)

275

D01, H., Halvorson, H., and Church, B., J. Bacteriol..
22. 43 (1959).

Hauge, J. G.,‘g. Biol. Chem., 352, 3630 (1964).

Bean, H. C., and Hassid, W. Z., 6, Biol. Chem., 215,
425 (1955).

Bentley, H., and Bhate, D. S.,‘£. Biol. Chem., 255,
1225 (1960).

 

 

Eliel, E. L., Allinger, N. L., Angyal, S. J., and
Morrison, G. A., Conformational.Analysis, John Wiley
and Sons, New York,1967, p. 404.

Conner, A. H., and Anderson, L., 157th National
Meeting Of the American Chemical Society, Minneapolis,
Minn., April 1969, NO. D 12.

Sweeley, C. C., Bentley, H., Makita, M., and Wells,
W. W., i, Amer. Chem. Soc., 65, 2497 (1963).

Brodie, A., and Lipmann, F., g. Biol. Chem., 212, 677
(1954) .

Suzuki, K., Kawada, M., and Shimazono, N., J. Biochem.
(Tokyo), _2, 448 (1961).

Nishizuka, r., and Hayaishi, 0.. g. gigl.‘ghgm,. 221.
2721 (1962).

Winkelman, J., and Lehninger, A. L., g.

Yamada, K., Ishikawa, S., and Shimazono, N., Biochim.
Biophys. Acta, 2, 255 (1959).

Bublitz, C. and Lehninger A. L. Biochim. BiOphys.
Acta, 61, 288 (1961). . .

Biol. Chem..

 

 

Sistrom, W. H., and Stanier, R. E., J. Biol. CLem.,
210, 821 (1954). '

Smiley. J. D., and.Ashwell, G., J. Biol. CLem.. 255,
1571 (1960).

Merrick, J. M. and Roseman, S. J. Biol. Chem. 225
1274 (1960). ' ' ' '

Linker, A., Hoffmann, P., Meyer, K., Sampson, P.,
and Korn, E. D., g. Biol. Chem., 225, 3061 (1960).

276

(109) Albersheim, P., and Killias, U., Arch. Biochem. Bithys..
22. 107 (1962). """"

(110) Starr, M. P., and Moran, F., Science, 135, 920 (1962).
(111) Blumenthal, H. J., Methods Enzymol., 2, 660 (1966).

(112) Blumenthal, H. J., and Jepson, T., Methods Enzymo .,
2. 665 (1966).

(113) Berman, T., and Magasanik, B., Q. Biol. Chem., 241,
800 (1966).

 

(114) Maitra, U., and Dekker, E. E., J. Biol. Chem., 228,
3660 (1963)-

(115) Hoffman, P., Linker, A., Lippman, V., and Meyer, K.,
g, Biol. Chem., 225, 3066 (1960).

(116) Preiss, J., and Ashwell, G., Biochem. Bioghys. Res.
Commun..8 _, .357 (1962)-

(117) Palleroni, N., and Doudoroff, M., J. BLol. CLem., 222,
499 (1956)

(118) Ghalambor, M., and Heath E. C. Biochem. Bioghzs.
Res. Commun., 3;, 288 (1963). '

(119) Dagley, S., and Trudgill, P. W., Biochem. g., 25, 48
(1965 .

(120) Levine, D. H., and Backer, E., J. Biol. Chem., 224.
2532 (1959).

(121) Minamikawa, T., and Uritani, I., g. Biochem. (Tokyo),
61.367 (1967)

(122) Boseman, 8., Jourdian, G. W., Watson, D., and Rood, B.,
Proc. Natl. Acad. Sci. Q.‘§.,‘41, 958 (1961).

(123) Comb, D. G., and Boseman S., J. BLol. CLem., 225,
2529 (1960).

(124) Good, N. E.,Winget, G. D., Winter, W., Connolly,7 T. N.,

Izawa S. and Singh, M. M. Biochemistry
(1966).. 0 O D 29%

(125) Entner, N., and Doudoroff, M., J. BLol. CLem., 196,
853 (1952).

(126) Kanamori, M., and Wixom, R. L., J. BLol. CLem., 238,
1004 (1963).

(127)
(128)

(129)

(130)

(131)

(132)
(133)
(134)

(135)

(136)

(137)
(138)

(139)
(140)

(141)

(1h2)

(143)
(1M4)

(1“5)

(146)

277
Myers, J. W., g. B101. Chem.. 329. 1416 (1961).

Wixom, B. L., Shatton, J. B., Strassan, M., J. B101.
CLem.,22§.128 (1960).

Kovachevich, K., and Wood, w. A., J. B101. CLem.,
_2. 745 (1954).

Meloche, H. P., and Wood, W. A., Methods Enzzgol..
2. 653 (1966). """"""'

Hurlbert, R. E., and Jawby, w. B., Methods Enzzgol..
2. 680 (1966). ""“'

Buttin, G., Cold Spring Harbor. fig, 213 (1962).
Buttin, G., g. 1112;. B__1__o1., 1, 16L» (1963).

Maltra. U., and Dekker, E. U., B101. Chem.. 322.
1#85 (1964).

Kuratom1,K.. and Fukunga, K., Biochim. Biophzs.
Acta, Z_. 617 (1963).

Ghalambor, M. A., and Heath, E. C., J. B101. CLem.,
222, 2427 (1962).

Charalampous, F. C., Methods Enzxgol.. 5, 283 (1962).

Warren, L., and Felsenfeld. H., J. B101. CLem., £22.
1421 (1962).

Okano, A., CLem. Pharm. Bull. (Tokyo), 5, 176 (1957).

Crumpton, M., and Davies, D..A.. Biochem. g,. 22,
729 (1958).

MaoPhillamy,‘A.. and E1derfeld,J.. J. Org CLem.,
‘_0 150 (1939)-

Khare, M. P., Helv. Chim. Acta, 35, 1515 (1962).
Khare, M. P., Helv. Ch1m.LActa,‘&5, 153# (1962).
Khare, M. P., Helv. CLim. Acta, 21, 155 (1954).

Tshesche, B., and Grimmer. G., CLem. Ber., 88. 1569
(1955)-

Satoh, D., and Ishii, H., CLem. Pharm. Bull. (Tokyo),
'fi, 284 (1956).

278

(147) Licht1, H., Tamm, C., and Reichstein, T., Helv. Chlm.
Acta,‘22, 1919 (1956).

(148) Zelnlck, B., and Schindler, 0., Helv. Chim. Acta, 59,
‘ 2110 (1957).

(149) Power, F. B., and Bogerson, H., g. Chem. Soc., 101,
(1912).

 

(150) Pigman, W., Chemistrz 2£_Carboh drates, 2nd ed.,
Academic Press, New York, N. Y., 1935, p. 107.

(151) Votocek, E., Collect. Czech. Chem. Commun., 1, #6

(152) Votocek, E., Collect. Czech. Chem. Commun., 1, 606
(1929 . '

(153) Votocek, E., g, Amer. Chem. §gg., jg, 1749 (1928).
(154) Kilianl, H., Chem. £333.. 552. 2116 (1892).
(155) Mannlch, C., Arch. Pharm. (Weinheim), 216, 211 (1938).

(156) Beber, T., and Reichstein, T., Helv. Chim. Acta, 22,
343 (19116). “""’ "'""' ““—

(157) Schmidt, 0.. gag” 518. 70 (1947).

(158) Jacob. Wm 23 1.3.12;- Shsmu 2.6.. 355 (1932).

(159) Lamb, 1.. and Smith, 3.. ,1. 9m. §9_9_.. 4&2 (1936).
(160) Young, W. T., g. 935. 211311., 1, 2111 (1942),

(161) Shah, N. M., Meyer, W., and Reichstein, T., Pharm.
Acta Helv., 33, 113 (1949).

(162) ?uw, g., and Gurtler, J., Helv. Chim. Acta, 39, 2079
1959 .

(163) Votocek, E., Chem. Zentra.,ll, 1042 (1901).
(16“) Votocek, E., Chem. Zentra., 2;, 1361 (1902).
(165) Votocek, E., Chem. Zentra., 2;, 1527 (1905).

(166) Goldberg, M. L., and Sternbach, S. H.,'g..Amer. Chem.
Soc., §_g, 1639 (1958).

(167) Berger, J., Sternbach, L., and Calhoun, K. M., g,
Amer. Chem. Soc.. g9. 1636 (1958).

Ill 1‘11
In|.lll .

(168)
(169)
(170)

(171)
(172)
(173)

(1711)

(175)

(176)

(177)

(178)

(179)
(180)

(181)
(182)
(183)

(181»)

(185)
(186)
(187)
(188)

279
Leach, B. E., g. Amer. Chem. Soc., 12, (+01 (1953).
Votocek, E., Chem. Zentra., ;, 1818 (1906).

Schmidt, 0., and Mayer, W. , Naturwissenschaften, 11;,
247 (1943).

Schmidt, 0.. and Mayer, W., Ann., 255, 26 (1944).
Schmidt, 0., Naturwissenschaften, £12, 130 (1965).

Bosselet, J. P., and Hunger, A., Helv. Chim. Acta,

23, 2103 (1951).

Stoll, A., and Junker, E., Modern Methods 91 Plant
ggilynsliséVggfi, Paech, K., Ed., Springer Verlag,
Leach, B. E., ,1. A_I_n_§_r_. £1129}. £93., 15, (+011 (1953).

Calhoun, K. M., and Johnson, J. E., Antibiotics and
Chemotherapy, _6_, 294 (1956).

Springer, G., and Williamson, W., Fed. Proc., _1_.§_,

2348 (1959) .

Springer, G., and Williamson, W., Biochem. 51., §j,
282 (1962).

Kubat, E., Biochem. _._I_., Q, 291 (1961).

McMahon, T., and Nolan, M. 0., _J. Cell. Comp. Physiol”
0. 219 (1957).

Sharon, N., and Shif, 1., Biochem. 1., 21, 210 (1961+).
Letherwood, E., Fed. Proc.. _2_2_. 239 (1963).

Levvy, G., and McAllen, A., Biochem. J., 82, 206
(1963).

Levvy, G., and McAllen, A., Biochem. J., £11, 361
(1963).

Levvy, G., Nature, 182, 1027 (1960).
Marney, A., and Got, 0., EXperienta, 22, (+41 (1964).
Hemmer, H., Ann. Rev. Pharm., 2, (+11 (1965).

Braunvald, E., and Kloche, F., Ann. Rev. Med., 16,
371 (1965).

(189)

(190)
(191)

(192)
(193)

(191+)
(195)
(196)

(197)

(198)
(199)
(200)
(201)
(202)

(203)
(204)

(205)

(206)
(207)
(208)
(209)
(210)
(211)

280

MacLennon, A. P., and Randall, R. M., Biochem. _J_.,
.222. 309 (1961). """‘""""

Baker, J. B., Biochem. J., g3. 352 (1962).

Spencer, J., and Giori, P., Can. _J_. Microbiol., 1.
185 (1961).

Frommhagen, L. H., Virology, 2, LL30 (1956).

Itaki), 0., Bull. Agric. Chem. Soc. Japan, 32, (+36
(1958 .

Hijama, W., Tohoku g. EXEtl. Med., 25, 333 (1952),
Hirst, E. L., _J. Chem. Soc., 323 (1951).

Gefretin, T., Compt. Rend. Soc. Biol., 145, 115
(1951 .

Dillon, T., Proo. Royal Irish Acad. Sci., 558, 189
(1953).

Glegg, R. E., Science, _11_8_, 611+ (1953).

Masamune, H., mg. M. 91139.. fig. 37 (1952)-
Bertalanffy, F., _C_g_r_1. 212;. M., .7_o_, 196 (1954).
Evans, A., J. Em. Technol., 23, 147 (1954).

Hosimoto, H., Bull. Japan Soc. Sci. Fishery, £9,
#06 (1954).

Emlllanl, F0. BEE. £29. £1.50 Qulm.. 21-22, 11 (1952-53).
Kent, P. w., Brit. ,1. Eggtl. Pathol., 6, 1&9 (1955).

I()oh1ur)1an, c. H., Arch. Biochem. Biophzsu fl. 445
1955 .

Kirby, K. 8., Biochem. J., fig. 582 (1955).

Minganti, A., m. 921;. Egg. M” 2, 248 (1955).
Eagon, J., .CLOEBE- £2513” _2_f_P_1._. 527 (1955).

Glegg, R. E., E353. 99;; 21351... g. “53 (1955).

Odin, L., 531:3 91322. M” 2, 1235 (1955).

Butt, H. R., g. Endocrin.., $2, 167 (1956).

(212)

(213)
(214)
(215)
(216)
(217)

(218)
(219)
(220)
(221)
(222)

(223)
(224)

(225)

(226)
(227)

(228)

(229)
(230)

(231)
(232)
(233)

281

Black, W. A. P., International.Peat Symposium,
Section B3 (1964).

Quirk, J. P., 2222;. 22. 357 (1956).

Ejirma, T., lgggg‘gg Serbutsugaka, 22, 85 (1854).
Vasseur, E., ALkil Kgmi, 1, 39 (1949).

Paigen, K., J. Bacteriol., 22, 1397 (1966).

Beg gs, W. H., and Rogers, Palmer, J. Bacteriol.,
21, 81869 (1966).

Paigen, K., g. Bacteriol., 21, 1201 (1966).

Ganeson, A. K., g. M. E231... l._6_, 42 (1965).

McBrain, D. C. H.,'g. Bacteriol., 2;, 1391 (1966).
Williams, B., and.Paigen, K., Fed. Proo., 24, 417 (165).

Wilgéams, B., Biochem. Biophys. Res. Commun., 24,143
19 .

Jayararnan, K., J. Mol. Biol., 18, 339 (1966).

Echols, E., and Adkger, S., J. Bacteriol., 2, 601
(1966).

Bertland, A. and Bugge B. Arch. Biochem. Biophys.
116,280 (1966). ' """" """‘"""

Alvardo, F., Biochin. Biophys. Acta, 41, 233 (1960).

Hu, A. S. L., and Cline, A. L., J. Biol. Chem., Zio,
4493 (1965).

Cuatrecasas, P., and Sega1,S., J. Biol. Chem., 241,
5910 (1966).

Kiliani, H., Chem. Ber., 53, 2866 (1930).

Rheiner, A., and.Reichstein, T., Helv. Chim._Acta,
2:. 687 (1952)

Ribas, I., Helv. Chim. Acta,‘24, 1477 (1951).
Reichstein, T., Helv. Chim. Acta, 22, 422 (i952),
Reichstein, T.. Helv. Chim. Acta, fig, 370 (1953).

(234)

(235)
(236)

(237)

(238)
(239)
(240)
(241)
(242)
(21(3)
(214)
(245)
(246)
(247)

(248)
(249)
(250
(251)
(252)

(253)
(254)

282

Richter, R., and Reichstein, T., Helv. Chim. Acta,
21. 76 (1954). " " " "'"

Hegediis, H., Helv. Chim. Acta, 23;. 1133 (1955).

Katz, A., and Reichstein, T., Pharm. Acta Helv., $2,
231 (1944).

Rittel, N., Hunger, A., and Reichstein, T., Helv..
Chim. Acta, 6, 434 (1953).

Jager, R. H., §_e_l_y_. 3133.2. £21189 2%. 977 (1959)-
Kiliani, H., Aggh, £2§£§., £20, 250 (1892).
Kiliani, H., 91139;. _B_g_x_~_., 2;. 2460 (1898).
Kiliani, H., 9239;. 1325., 5_1_. 1633 (1918).
Okada, 3., ,1. M. £93. m, 21. 118 (1953).
Sakasawa, P., P;gg£g_fl§g,,'4, 20 (1956).

Okano, A., nggg.Ԥg;l. (Tokyo), 6, 173 (1957).
Tschesche, R., Agg,, 696, 160 (1957).

Hoji, K., Egggg. 22;}. (Tokyo), 2, 289 (1961).

Kundig, W., Kundig, F. D., Anderson, B., and Roseman,
S.,.g. Biol. Chem., 241, 3243 (1966).

Luderitz, 0., Staub, A. U., and Westphal, 0.,
Bacteriol. Rev., 29, 192 (1966).

Blumberg, W., Levine, W., and Persack, J., g. Biol.
2222-. 23.2.. 2850 (1967).

 

) Ashwell, G., and Volk, P.,‘g. Biol. Chem., 240, 4549

(1965).

Adye, J., and Powelson, D., J, Bacteriol., 81, 780
(1961).

Bailey, J. M., and Pentchev, P. G., Amer. g, thsiol.,
208, 385 (1965).

Rindmarsch, G., g, thsiol. (London), 186,166 (1966,

%evgg5 G., and Conchie, J., Methods Enzxmo ., 2, 571
19 .

(255)
(256)

(257)

(258)

(259)

(260)

(261)
(262)

(263)

(264)

(265)

(266)
(267)

283
Hayaishi, D., Methods Enzzgol., 2, 731 (1966).

Esterly, J., Standen, A. C., and Pearson, E.,
.g. Histochem. Cytochem., $j, 470 (1967).

MacLennon, A. P., and Smith, D. W., Biochem. J.,
“9 3 P (1960)-

Wheat, B., Rollins, E. L., and Leatherwood, J.,
Nature, 202, #92 (1958).

Wheat, B., Rollins, E. L., Leatherwood, J. M., and
Barnes, B. L., g, Biol. Chem.. 228, 26 (1963).

Bailey, J. M., Fishman, P. H., and Pentchev, P. G.,
g. Biol. Chem.. 244, 781 (1969).

Suzuki, N., Biochim. BithzS. Acta, £12, 371 (1969).

Chiu, T., and Felngold, D. 3., Biochemistry, g, 98
(1969).

Wilson, D. M., and A31, 3., g. Bacteriol. , 12, #10
(1957).

Engle?berg, E., Arch. Biochem. Biophzs.. 1;, 179
(1957 .

Heath. E. C., Ghalambor, M. A., 1. Biol. Chem.. 222.
2u23 (1962).

 

 

Shah, N. M., Pharm. Acta Helv.. fig. 120 (1953).

 

Busch, H., Introduction Engiochemistr g£_the Cancer
Cell, Academic Press, N. Y., p. 253, 2&0.

APPENDIX A
OCCURRENCE AND METABOLISM OF D-FUCOSE AND ITS DERIVATIVES

D-Fucose (6-deoxy-D-galactose) is a constituent of
cell walls, cardiac glycosides, mucoproteins, mucOpoly-
saccharides and numerous other glycosides. This sugar
and its 2-0-methyl—, 2,3-di-O-methyl-, 3-0-methy1-, and
2-amino-2-deoxy-, 3-amino-3-deoxy-, and D-acetyl-B-amino-
B-deoxy-derivatives have been isolated from the hydroly-
sis products of heteroglycosides obtained from plants,
fresh-water gastropods, and bacteria (139-176, 180-182,
229-246).

D-Fucose and D-digitalose (3-methoxy D-fucose) are
liberated upon acid hydrolysis of chartreusin, an anti-
biotic active against Gram-positive organisms and myco-
bacteria; it is produced by Streptomzces chartreusis from
African soil and by Streptomzces Sp. from North American
soil (166-168, 175-176).

DAFucose has been isolated from cardiac glycosides
obtained from seeds of Strophanthus sarmentosis and §.
eminii (159, 162), and from leaves of Digitalis purpurea
and‘Q. lanata (145, 1&6). D-Digitalose has been isolated

from the seeds of‘§. gerardii.‘§. commontii, §, armentosis,

(173), and Digitalis purpurea (172). StrOphanthus and

Digitalis belong to the family Apocynaceae, which provides
284

285
numerous D-fucose-containing drugs widely used as heart
stimulants. D-Fucose is a carbohydrate component of a
cuproprotein. Stellacyanin (249) and has been isolated
from the root glycosides of the morning glory, Ipomoe
orizanesis (17k), and from the family Cruciferae, which
comprises the cabbages, cauliflowers, brussel Sprouts,
and mustard seed (174). 3-0-Methyl ethers of D-fucose
have been obtained from the naturally occurring pure
cardiac glycosides, strospesid and panstrosid (173).
Other naturally occurring methoxy derivatives and methoxy
glycosides of D-fucose are also documented (146-1&8,
177-179. 229-2ué).

The roots of various plants of the Convolvulaceae
contain glycosides of D-fucose and.have been used
medicinally as purgatives for many centuries. The plants
are cultivated in Mexico, Jamaica, and South America
(161). In addition. four common Species of aquatic
gastropods have been found to possess Specific poly-
saccharides containing D-fucose (180).

D-Fucosamine has been found to be present in bacteria.
This amino sugar was first located in a Specific lipo-
polysaccharide of the Gram-negative bacterium, Chromo-
bacterium violaceum (luO). Subsequently. it has been
isolated from Bacillus lighenformis (181).I§. subtilis
(181), and g, cereus (182, 258, 259), Salmonella (248),
Xanthomonas campestris (250). and Pseudomonas aeruginosa

(261). Thus, Dbfucose and its derivatives are quite

286

widespread: the cases cited represent but a few of the
many sources that are documented (139-171, 17u-176, 180).

The literature contains many instances in which
fucose was detected and identified by co-chromatography
with L-fucose, but in which no specific rotation studies
were performed. Thus, D-fucose could conceivably be
present in any of the following instances: O-methyl
fucose in flyobacterium'bgzig (189) and fl, tuberculosis
(257), the organic material if egg shell matrix (190).
the fucose polysaccharides produced in Corynebacterium
insidiosium and Q, sepedonicuml(191). the polysaccharide
found in the P38 and the Lee strains of influenza virus
(192), hemicellulose from bean seeds (193), fucose in the
gastric mucosa (19“), exudate from the bark of HlEEE
fglzg (195), mucoprotein from sedentary polychete worms
(196), saponins hydrolyzate from Gypsophila (266), seed
exudate of Ascophxllum nodosum (197), carbohydrate com-
ponents of reticular fibers from lymph, lung, testis, and
adipose tissue (198), placental carbohydrates (199),
carbohydrate content of tissues of cancerous individuals
(26?), lung reticulin of cattle (200), polysaccharide
produced by a strain of‘g. polzmzxa (201), carbohydrates
of mollusk style (202), fucose polysaccharide from cor-
tical fibers (203), fucose content of human renal
reticulin (201+), Descement membrane of bovine cornea (205),
quebracho tannin extract (206), amphibian egg jelly (207),

capsular polysaccharide of Pseudomonas fluorescens (208),

287
guinea pig spermatozoa (209), human cervical mucus (210).
urinary gonadotrOpins (211), Sphagnum moss and peat (212),
aSpen wood (213), cerebroSpinal fluids from patients with
mental disorders (21#), sea urchin eggs (215), microcysts
of Exxococcus xanthus (251). The presence of L—fucose was
inferred in each of the cases mentioned. However, because
of the current realization of the prevalence of the D-
isomer. it would be anticipated that in at least some of
these cases D-fucose would be present instead of, or in
addition to. L-fucose. This observation has also been
made by Levvy (185), who stated, “Although L-fucose has
been positively identified in mammalian tissue in at least
two instances, in many cases, its presence has been
inferred from evidence that does not distinguish it from
its-D-enantiomorph."

Thus, the literature substantiates the ubiquity
of D-fucose. Although D-fucose was once thought to be
a rarity compared to the supposed relative abundance of
the L-isomer, it is felt that the wideSpread presence of
D-fucose must now be recognized and that numerous new
sources will continue to be unveiled.

Specific B-D-fucosidases have been reported in
organisms on varying positions of the evolutionary scale.
The visceral hump of the limpet, Patella vulgata, has
been described as providing a good source of glycosides
(183). It has been reported also that the limpet possesses
an enzyme that hydrolyzes B-thucosides and that is dis-

288
tinguishable from B-D-galactosidase (183). In addition,
Levvy gg‘gl. have reported a B-D-fucosidase from ox
liver (184). Since then, Specific B-D-fucosidases have
been discovered in liver, kidney, and epididymis tissues
of rat, ox, and pig (185). There are also documentations
for Specific B-D-fucosidases in cow ovary and uterus
(148), in the digestive joice of the snail, §§;;§,pomatia
(186), and in the jejunum of rat, rabbit, mouse, and
guinea pig (256). D-Fucose-containing cardiac glycosides
have been shown to be cleaved in rat liver by a reportedly
unknown glycosidase (187); presumably this enzyme is iden-
tical to the B-D-fucosidase isolated from rat liver by
Levvy (185). It has also been shown that 92% of 6-deoxy-
hexose-glycosides from Digitalis Sp. are metabolized by
human subjects (188), and it is conceivable that man also
possess a Specific B-D-fucosidase. D-Fucono-Y-lactone
has been found to be a competitive inhibitor of mammalian
B-fucosidases (25k).

No pathway for the metabolism of D-fucose has been
elucidated in Spite of its ubiquity. In fact, very
little data concerning any aSpect of D-fucose metabolism
is available in the literature. Kundig gthgl. (247)
have reported that in.§,'gg;i D-fucose induces a phos-
photransferase which catalyzes a phoSphoryl transfer
from a phosphoprotein to galactose, thereby forming
galactose-6-phoSphate. Buttin (132-133) has reported

D-fucose induction of galactokinase, galactose-i-P-

289
uridylyltransferase, and UDP-galactose-4-epimerase in‘g.
22;}, He also Specifically reports that g, 32;; does not
use D-fucose as a carbon source. Various other researchers
have used D-fucose as a gratuitous inducer of the galactose
operon on the assumption that the carbohydrate was not
metabolized (216-229).

D-Fucose has also been reported to be a most effi-
cient repressor of the lac operon (217, 218, 221) and to
markedly enhance the fluorescence of UDP-galactose-4-
epimerase (225). Alvardo (226) has shown that galacto-
kinase isolated from Saccharomyces fragilis did not phos—
phorylate D-fucose using ATP as the phosphoryl donor.

Hu and Cline (227) and Wallenfels and Kurz (89)
have isolated from D-galactose-grown Psuedomonas Sp. a
D-galactose dehydrogenase which oxidized D-fucose at the
same rate as D-galactose. Cuatrecasas and Segal (228)
have recently reported on a NAD-Specific mammalian galac-
tose dehydrogenase from rat liver which exhibited low
activity on D-fucose. Hayaishi has also reported on a
particulate lactose dehydrogenase of Eseudomonas graveolens
which was highly active on D-fucose (255). There is no
report in the literature that any of the above organisms
which possess D-fucose dehydrogenases are capable of grow-
ing on D-fucose.

Echols gt 3;. (224) have shown through inducer
uptake eXperimentS that D-galactose and D-fucose enter

the cell via different tranSport systems. This is quite

290
interesting in light of the fact that Specific B-glyco-
Sidases for D-fucosides and D-galactosides have been dis?
covered (183-186); these findings suggest, perhaps, that
two distinct operons are present in some bacteria for D-
galactose and D-fucose metabolism. D-fucose has also
been reported to be actively tranSported by rat and ham-
ster small intestine and rat kidney (252, 253). A
mutarotase from tovine kidney cortex has been shown to
be very active on Defucose (260) whereas mutarotase from
rat kidney has been demonstrated to be inhibited by D-
fucose (253). DeSpite the plethora of information con-
cerning D-fucose, its biodegradative pathway has not been

previously described in any organism.

APPENDIX B

ISOLATION AND CHARACTERIZATION OF THE PSEUDOMONAD USED
IN THE PRESENT INVESTIGATION

The organism used in this study was isolated from
commercial D-fucose. Solid, non-sterile D-fucose (50 mg)
was added to a culture tube containing the mineral medium
(10 m1) described in Experimental Procedures. Turbidity
was noticeable after 5 days on a reciprocal shaker at
32°C. The culture was streaked on D-glucose-mineral
agar; a colony was selected from the plate and was grown
on D-fucose. This culture was streaked again on D-glu-
cose-mineral agar; the above procedure was repeated twice
again until purity was assured.

The organism thus selected was tested for classi-
fication by the Michigan State Veterinary Bacteriological
Laboratory and the Michigan State Agricultural.Pathological
Laboratory. The characteristics of this organism, as
determined by these two laboratories are as follows:

Gram stain: coccobacilli, Gram-negative
Bloodplate: sharp odor, similar to Pasteurellg

Tryptone agar: slightly yellow, no pigment, sticky
colonies

MacConkey agar: small, fine colonies

Kligler's agar: negative, growth light

291

292

Seller's agar: no gas butt, no fluorescence
Oxidase: positive
Catalase: positive, but Slow
Nitrate reductase: positive
Citrate: positive
Urea: positive
Fermentation: non-fermentative, aerobic
Voges-Proskauer: positive
Indole: negative
Lys ine : no growth
Motility: negative
Liquid Culture:

D-Glucose: positive

Lactose: negative

Starch: negative

L-Sorbose: negative

Adonitol: negative

Dulcitol: negative

On the basis of the above data, the organism may be clas-

sified as a pseudomonad.

The organism was further tested for growth sub-

strate Specificity. D-Glucose log-phase cells were inocu-

lated into mineral medium containing 0.5% carbohydrate.

The carbohydrate and the time required to reach the mid-

point of log-phase are listed below.

293

Carbohydrate Time
D-Gluco se i 4 hours

 

D-Galactose "
D-Galactarate "
DuGalactonate "
D-Glucarate "
D-Fucose "
L—Arabinose "
Sucrose, Lactic Acid 24 hours

6-Iodo-6-deoxy-D-galactose "

D-Gluconate #8 hours
Glycerol "
D-Fructose 4 days
D-Glucosamine "
D-Xylose "

There was no growth after two weeks on D-arabinose, L-
rhamnose, cellobiose, mannitol, sorbitol. lactose, pro—
pylene glycol, Dbmannose, L-Sorbose, L-mannose, L-galac-
tose, L-glucose, or L-gluconate. .After 2.5 weeks, slight
growth was noticed on D-mannose, and full growth was
noticed on propylene glycol.

Electron microscopy of this organism was performed
by Richard F. Hamman on a RCA.EMU-2E electron microscope.
Specimens to be assayed were appropriately diluted and a
1 ul sample was placed on a 200 mesh copper grid (No. 2200.
Ernest F. Fullam Assoc.) covered with 0.5% Formvar, air

dried, and shadowed with tungsten oxide in a Kinney High

 

'- if

294

Vacuum Evaporator (New York Air Brake Co.) at an angle of
180-200. The tungsten oxide was generated by heating
in air the 0.025 in. tungsten wire (Ladd Research Indus-
tries, Inc.). Polystyrene latex Spheres of 0.26 p,
standard deviation 0.006 u, (Dow Chemical Physical
Research Lab, Midland, Michigan) were placed on the grids
as a Sizing reference prior to sample application.

Carbon stabilized grids were prepared by grinding
a 1/16" dia. tip, 3/16" long on a carbon rod 1/8" in dia.
and butting it against a flat-ended 1/8" rod. The evapor-
ator "filament glow" rheostat was advanced to 100% (45-50
amps) and the "filament glow" switch was flashed on and
off as rapidly as possible. This deposited an even
layer of carbon over the entire grid surface. The use
of carbon stabilized grids was necessary due to the fail-
ure of the Fbrmvar film under conditions of negative
staining with PTA. Samples to be negatively stained were
mixed 1:1 with 2% PTA, pH 6.8 and placed on the grid to
dry. If the staining was too heavy, appropriate dilutions
were made. Samples were viewed at 50 KV accelerating
voltage without an objective aperature in the EMU-2E.
Electron micrographs were eXposed on 2" x 10" projector
Slide plates, developed in D-19 for 1-2 minutes, fixed,
and dried. Two electron micrographs of the pseudomonad
are presented in Figure B-i. The microorganism is about

1.25 u in length and is not flagellated.

295

Figure B-l