IllilllllllllllllIllllllllllllllllllllll

3 1293 01691 8884

This is to certify that the

dissertation entitled

DEGRADATION OF 3-PHENYLPROPIONIC
ACID BY HALOFERAX SP.D1227

presented by
Weijie Fu

has been accepted towards fulfillment
of the requirements for

PhD. Microbiology

degree in

 

 

fig 22 iv;

 

Major professor

Date W7

 

 

LIBRARY

Michigan State
University

 

 

 

PLACE IN RETURN BOX
to remove this checkout from your record.
TO AVOID FINES return on or before date due.

 

DATE DUE MTE DUE DATE DUE

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

1/98 chtHODataanfiS—p.“

 

DEGRADATION OF 3-PHENYLPROPIONIC ACID BY
HALOFERAX SP. D1227

By

Weijie F u

A DISSERTATION

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

DOCTOR OF PHILOSOPHY

Department of Microbiology

1997

ABSTRACT

DEGRADATION OF 3-PHENYLPROPIONIC ACID BY
HALOFERAX SP. D1227

By

Weijie Fu

Haloferax sp. D1227, an extreme halophile isolated by Emerson et al.
from soil contaminated with highly saline oil brine, is the only reported
aerobic archaeon capable of metabolizing aromatic compounds as the sole
carbon source for growth. In contrast to the extensive research on aromatic
degradation by bacteria and fungi, little is known about the aromatic
metabolism in Archaea, which represents a third domain of life
phylogenetically distinct from Bacteria and Eucarya. The degradation of the
important plant-derived aromatic compound 3-pheny1propionic acid by Hf
D1227 was studied to understand the archaeal strategies for
phenylpropanoid catabolism. I propose that Hf D1227 metabolizes 3-
phenylpropionic acid by initial 2-carbon shortening of the side chain to
benzoleoA, via a mechanism similar to fatty acid B-oxidation, followed by
aromatic degradation using a gentisate pathway based on the following

findings: (1) the extracellular accumulation of cinnamic acid, benzoic acid,

3-hydroxybenzoic acid, and gentisic acid in the cultures of Hf D1227 grown
on 3-phenylpropionic acid; (2) the presence of a 3-phenylpropionyl CoA
dehydrogenase; (3) the ATP-, CoA— and NAD-dependent conversion of
cinnamic acid to benzoleoA; and (4) the presence of gentisate 1,2-
dioxygenase. I also demonstrated that the upper aliphatic pathway from 3-
phenylpropionic acid to benzoic acid is separately regulated from the lower
gentisate pathway.

The gentisate 1,2-dioxygenase involved in the 3-phenylpropionic acid
degradation by Hf D1227 was purified using phenyl-Sepharose CL-4B
chromatography, hydroxyapatite chromatography, and gel filtration with
Superose 12. The enzyme was a 174,000i6,000 Da homotetramer
composed of 42,000i1 ,000 Da subunits. The maximal enzyme activity was
obtained in 2 M KCl or NaCl, 45°C and pH 7.2. The enzyme had a 9.2%
excess acidic over basic amino acids. The gene encoding Hf. D1227
gentisate 1,2-dioxygenase was cloned, sequenced and expressed in H.
volcanii. Four histidine clusters (H-X-H) and a segment containing key
conserved residues in eubacterial extradiol dioxygenases fingerprint region
were identified in the deduced amino acid sequence of Hf D1227 gentisate

1 ,2-dioxygenasc.

To my parents and my husband for their love and support.

iv

ACKNOWLEDGMENTS

I would like to express my sincere thanks to: Dr. Patrick Oriel, my
mentor, for his guidance, kindness and patience. Dr. Michael Bagdasarian,
Dr. John Breznak, Dr. Shelagh F erguson-Miller and Dr. Robert Hausinger,
my committee members, for their invaluable advice and time. Dr. Olga
Maltseva and Dr. Tae-Kyou Cheong for all the helpful discussions and for
teaching me many techniques. Dr. Sadhana Chauhan and Dr. Trevor

D'Souza for their help at the beginning of my Ph.D. career.

PREFACE

This dissertation includes four chapters and an appendix. The first
chapter is a general introduction and the fourth chapter is a summary and
recommendations for future research. The second and third chapters
present the results of the dissertation research in manuscript form. Both
have been submitted to Extremophiles. I also participated in a project to
isolate a pinene-degrading thermophile, Bacillus pallidus BR425. The
paper presenting the results of the pinene work is included in this thesis as

an appendix.

vi

TABLE OF CONTENTS

LIST OF TABLES ............................................................................... ix

LIST OF FIGURES .............................................................................. x

CHAPTER 1

INTRODUCTION ............................................................ 1
Objective of this study .................................................... 1
Background ................................................................. 2
References ................................................................. 19

CHAPTER 2

DEGRADATION OF 3-PHENYLPROPIONIC ACID

BY HALOFERAX SP. D1227 ............................................. 26
Abstract .................................................................... 26
Introduction ............................................................... 27
Materials and Methods .................................................. 30
Results ..................................................................... 39
Discussion ................................................................ 54
Acknowledgments ...................................................... 58
References ............................................................... 58

CHAPTER 3

GENTISATE 1,2-DIOXYGENASE FROM

HALOFERAX SP. D1227 ................................................. 63
Abstract .................................................................. 63
Introduction .............................................................. 64
Materials and Methods ................................................. 66
Results .................................................................... 77

vii

Discussion ................................................................ 87

Acknowledgments ....................................................... 90
References ................................................................ 90

CHAPTER 4

SUMMARY AND RECOMMENDATIONS

FOR F UTURE RESEARCH .............................................. 94
Summary of the dissertation research findings ...................... 94
Future research ........................................................... 95
References ................................................................ 97

APPENDIX

DEGRADATION OF PINENE BY

BA CILL US PALLID US BR425 .......................................... 98
Abstract .................................................................. 98
Introduction ............................................................. 98
Materials and Methods ................................................ 100
Results .................................................................. 104
Discussion .............................................................. 1 12
Acknowledgments .................................................... 1 17
References .............................................................. 1 17

viii

LIST OF TABLES

Table 2.1 Aromatic metabolites observed in 48 hour
Hf D1227 cultures grown on various substrates ........................... 43

Table 2.2 Stimulation effect of MgClz on ligase activity .................. 47
Table 2.3 Accumulation of intermediates when 3-phenylpropionic

acid and 2,2'-dipyridyl were added to cells pregrown on various
substrates ........................................................................ 53

Table 3.1 Purification of gentisate 1,2-dioxygenase from Hf D1227 .. 78
Table A1 Growth of B. pallidus on monoterpenes ....................... 107

Table A2 Metabolites formed during grth of Bacillus pallidus
on a-pinene .................................................................... 108

Table A3 Metabolites formed during two-phase incubation of
BR425 suspended cells ....................................................... 111

Table A4 Stimulation of B. pallidus BR425 pinene hydroxylase
activity by electron acceptors ................................................ 113

LIST OF FIGURES

Figure 1.1 A universal phylogenetic tree based on small subunit ribosomal
RNA sequences. The tree has been rooted by analysis of duplications
in protein sequences. (Adapted from Olsen and Woese, 1997) ............ 4

Figure 1.2 Fatty acid B-oxidation cycle. Individual enzyme reactions:

1. Fatty acyl-CoA dehydrogenase; 2. 2,3-enoyl-CoA hydratase; 3. 3-
hydroxyacyl-CoA dehydrogenase; 4. 3-ketoacy1-CoA thiolase.

(Adapted from Ratledge, 1994) ................................................ 15

Figure 1.3 Aerobic benzoate degradation pathways in various
Microorganisms ................................................................... 17

Figure 2.1 Phylogenetic position of strain D1227 within the genus
Haloferax based on a maximum-parsimony analysis. The bar
represents a 0.5% difference in evolutionary distance .................. 41

Figure 2.2 Accumulation of intermediates when 3-phenylpropionic
acid and 2,2’-dipyridyl were added to 3-phenylpropionic
acid-grown cells ................................................................ 45

Figure 2.3 Effect of NAD on the conversion of cinnamleoA to
benzoleoA ..................................................................... 49

Figure 2.4 Inhibition of gentisate 1,2-dioxygenase activity by
2,2’-dipyridy1 ................................................................... 51

Figure 2.5 Proposed pathway for the degradation of
3-phenylpropionic acid by Hf D1227 ...................................... 56

X

Figure 3.1 Strategy for cloning the coding and flanking regions of Hf
D1227gentisate 1,2-dioxygenase gene. The coding region is enclosed

in the box. Solid lines represent Hf D1227 sequences and dashed lines
represent pBluescript SK(+) sequences .......................................... 73

Figure 3.2 CTAB polyacrylamide electrophoresis of Hf D1227 gentisate
1,2-dioxygenase at different stages of purification. l. marker proteins;

2. crude extract; 3. phenyl Sepharose C1-4B; 4. hydroxyapatite; 5.
superose 12 ......................................................................... 79

Figure 3.3 Effect of salt concentration on Hf D1227 gentisate 1,2-
dioxygenase activity ............................................................... 82

Figure 3.4 Nucleotide sequence of Hf D1227 gentisate 1,2-dioxygenase
gene and the deduced amino acid sequence. Amino acid sequences
determined by protein sequencing are underlined. A putative promoter
sequence is boxed. The histidine clusters are in bold letters. The

putative conserved residues are double-underlined ........................... 85

Figure A1 Growth of Bacillus pallidus BR425 in varied
concentrations of (R)-a-pinene ................................................ 106

Figure A2 Fragmentation pattern of acid metabolite from
BR425 growth on (at-pinene or carvone ....................................... 109

Figure A3 Hypothetical pinene degradation pathway for
Bacillus pallidus BR425 ........................................................ 115

xi

Chapter 1

INTRODUCTION

Objective of this study

An extremely halophilic, aerobic archaeon Haloferax sp. D1227 was
isolated by Emerson et al. from soil contaminated with highly saline oil
brine near Grand Rapids, Michigan (20). This strain requires 2 M NaCl for
optimal growth and is the only reported aerobic archaeon capable of
utilizing aromatic compounds as sole carbon and energy sources for growth.
The degradation of aromatic compounds has been studied extensively in
bacteria and fungi (8, 24, 45), but little is known about the aromatic
metabolism by Archaea, which represents a third domain of life
phylogenetically distinct from Bacteria and Eucarya (53). The Hf D1227
metabolism of 3-phenylpropionic acid, which is a central metabolite in plant
physiology (25), was studied to understand the archaeal strategies for
phenylpropanoid catabolism.

Since cinnamic acid, benzoic acid, 3-hydroxybenzoic acid and gentisic

acid were detected in Hf D1227 cultures grown on 3-phenylpropionic acid,
it was hypothesied that 3-phenylpropionic acid was metabolized by B-
oxidation of the side chain to benzoic acid, which was subsequently
degraded via a gentisate pathway. The Hf D1227 gentisate 1,2-
dioxygenase, one of the key enzymes in the 3-phenylpropionic acid
degradation pathway, was purified and characterized as a study of the first
ring-fission dioxygenase from an extreme halophile. The gene encoding
Hf D1227 gentisate 1,2-dioxygenase was cloned and sequenced to obtain
the first gentisate 1,2-dioxygenase gene sequence and information on the
relationship between this archaeal gentisate 1,2-dioxygenase and other ring-

cleavage dioxygenases.

Background
1. Archaea as a third domain of life

Haloferax sp. D1227 used in this study is a halophilic archaeon. It was
twenty years ago when Carl Woese first announced his discovery of the
Archaea - a group of organisms so different from all other living organisms,

that he placed them in a separate domain of life (54). By comparison of

ribosomal RNA sequences, Woese articulated the now-recognized three
primary lines of evolutionary descent, termed domains: Eucarya, Bacteria,
and Archaea (53). A universal phylogenetic tree is shown in Figure 1.1.
Within the archaeal domain, there are two taxonomic divisions
(Euryarchaeota and Crenarchaeota) and three phenotypic groups, the
methanogens, extreme thermophiles, and extreme halophiles. Among
characteristics typical of Archaea are: (1) membrane lipids composed of
ether-linked isoprenyl phosphoglycerides; (2) lack of muramic acid-
containing peptidoglycan in the cell envelope; (3) the "common arm" of the
RNAs contains pseudouridine or l-methylpseudouridine instead of
ribothymidine; and (4) the sequences of SS, 168, and 23S rRNA differ
significantly from those of eubacteria and eucaryotes (29).

It is becoming clear that the three fundamental cellular information
processing systems of Archaea - DNA replication, transcription and
translation - are more similar to their eucaryotic than their bacterial
counterparts (42). Archaea] DNA polymerases and replication fork proteins

are eucaryote-like; bacteria and archaea-eucaryotes replication fork proteins

Bacteroldes

Escherichia
Baclllus
Synechococcus

i Chloroflexus
Thermotoga

 

  

 

 

 

 

 

Thermococcus
Methanococcus

Methanobacterlum

    
  

Memanomlcreblum

Halobacterlum

 

 

 

\ Homo

Zea

Saccharomyces
Paramecium

 

Trypanosoma

 

Valrlmorpha

Bacteria

Crenarchaeota

Archaea

Euryarchaeota

Eucarya

Figure 1.1 A universal phylogenetic tree based on small subunit ribosomal
RNA sequences. The tree has been rooted by analysis of duplications in
protein sequences. (Adapted fi'om Olsen and Woese, 1997)

have little or no sequence similarity (21). In addition to the universally-
distributed three largest subunits of the RNA polymerase, the archaeal RNA
polymerase contains additional small subunits not found in the simpler
bacterial enzyme, but all having counterparts in the eucaryotic version (33,
57). The transcription initiation system in Archaea is a simpler version of
that seen in eucaryotes and basically different from the bacterial system (34,
57). The archaeal ribosomal proteins and tRNA charging enzymes are more
similar to their eucaryotic than to their bacterial homologs (2).

Archaea also resemble Eucarya in many other aspects. Introns have been
found in archaeal rRNAs and tRNAs, but not in mRNAs; and RNA splicing
in Archaea resembles that of eucaryotes (23). One distinction between
Eucarya and Bacteria is the presence of histones in Eucarya. Archaea]
histones have been found which compact DNA into histone-DNA
complexes analogous to the eucaryotic nucleosome, and sequence
comparison shows that the eucaryotic and the archaeal histones evolved
from the same ancestor (2).

The similarity of the fundamental information processing systems in

Archaea (DNA replication, transcription, and translation) to their eucaryotic

counterparts and other resemblances between Archaea and Eucarya (RNA
splicing, histones, etc.) strongly suggest that Archaea are more closely
related to Eucarya than to Bacteria. However, the bacterial features in
Archaea should not be overlooked. Archaea and Bacteria are quite similar
in morphology, lack of nuclei, genomic organization, and metabolic
pathways. Like Bacteria, Archaea have single circular chromosomes and
genes are organized in operons; some archaeal operons are identical in gene
order to bacterial operons (21). Archaea] metabolic genes in many key
pathways such as carbon fixation, biosynthesis of amino acid, nucleotide,
and coenzyme, are strikingly similar to bacterial counterparts (43).

The recently obtained first archaeal genome sequence from
Methanococcusjannaschii provides valuable information on Archaea (7).
A striking feature of the M. jannaschii genome is that about 50% of the
potential coding sequences match no sequence in the databases, indicating
the uniqueness of Archaea. Open reading frames encoding eucaryote-like
DNA polymerase, replication proteins and DNA repair enzymes were
detected in the genome of M. jannaschii. Ten of the eleven translation

initiation factors, two of the three elongation factors, and the only release

factor recognized in M. jannaschii genome are of the eucaryotic type. In the
M. jannaschii genome, genes for subunits of a complex such as RNA

polymerase and ribosome are linked in an operon.

11. Overview of extremely halophilic, aerobic Archaea

Haloferax sp. D1227 used in this study is an extremely halophilic,
aerobic archaeon. There are six recognized genera of halophilic Archaea,
four genera of non-alkalophilic halophiles (i.e. Halobacterium, Halococcus,
Haloferax, and Haloarcula) and two genera of alkalophilic halophiles (i.e.
Natronobacterium and Natronococcus). Non-alkalophilic halophiles have
been isolated from a variety of highly saline environments, including salted
food products (fish, hides, bacon, and sausage), salt lakes such as the Great
Salt Lake in Utah, the Dead sea, and hypersaline soils. In contrast,
alkalophilic halophiles have mainly been found in highly alkaline soda
lakes such as the Wadi Natrun in Egypt and Lake Magadi in Kenya. These
halophiles are usually of a bright red-orange color due to the synthesis of
bacterioruberin pigment, which protects the organisms against bright

sunlight present at most habitats of extreme halophiles. The cell shapes of

halophilic Archaea, which are rods, cocci, square-shaped, or pleiomorphic,
can vary depending on the growth conditions.

The optimum NaCl concentrations for the growth of extreme halophiles
is typically between 2.6 M and 4.3 M. Some can even grow in saturated
NaCl concentration (6 M). Na+ and Cl' ions are required for growth, but Na+
can be partially replaced by K”. Isolates from high Mg“ environments
such as the Dead Sea usually have high requirements of Mg“.
Haloalkalophilic strains require a pH around 8.5 to 9.5, while the optimal
range for nonalkalophilic halophiles is between pH 7 to 7.5. All halophilic
Archaea grow well at 37 to 40 °C. Oxygen supply is an important
parameter to consider in cultivating extreme halophiles, since the high salt
concentrations decrease the oxygen solubility in the media. While H.
halobium and related strains grow primarily on amino acids, Haloferax and
Haloarcula species show the ability to use carbohydrates as carbon and
energy sources. Some Halobacterium strains produce bacteriorhodopsin
and are therefore capable of using sunlight as an energy source. In general,
these organisms do not grow rapidly, exhibiting generation times 2 to 12

hours. Halophilic Archaea are highly resistant to most antibiotics.

To balance the high osmotic pressure and ionic strength in the
hypersaline environments, extremely halophilic Archaea accumulate high
concentrations of KC] inside the cells (9), unlike halophilic Bacteria and
Eucarya where compatible solutes such as glycine and betaine are
accumulated as osmotica. The high concentrations of intracellular KCl,
approximately the concentration of NaCl in the medium, are deleterious to
many common proteins and other biological macromolecules. High KCl
causes aggregation of proteins because of enhancement of hydrophobic
interactions, it interferes with electrostatic interaction because of charge
shielding, and it reduces the availability of free water because of salt ion
hydration (although K+ hydrates less water than Na+) (35). By introducing
additional acidic residues (glutamic acid and aspartic acid) into the proteins,
halophilic Archaea proteins are highly adapted to function in a milieu
containing 2 to 5 M KCl. The bulk protein of the extreme halophiles
contains an excess acidic over basic amino acids of close to 10%, whereas
the bulk protein of the non-halophilic organisms is about chemically neutral
(48). The X-ray crystal structures of halophilic malate dehydrogenase and

2Fe-28 ferredoxin (19, 22) show that most of the acidic residues are located

10

on the surfaces of the proteins. These acidic residues are more highly
hydrated than other amino acids and can coordinate a hydrated salt ion
network at the protein-solvent interface as demonstrated in crystals of
halophilic ferredoxin. Acidic amino acids can also be used to form internal
salt bridges with strategically-positioned basic residues inside the protein to
provide structural rigidity. Excess acidic residues also reduce protein
hydrophobicity. Comparison of halophilic archaeal protein-encoding genes
with their nonhalophilic homologs exhibit an inordinately high proportion
of nonsynonymous nucleotide substitutions, and many of the resulting
amino acid replacements involve the addition, removal, relocation, or
rearrangment of acidic residues (17).

Proteins from extreme halophiles are not only stable in high
concentrations of salt, but actually require them for their function. Thus,
purification of halophilic proteins requires procedures that can be carried
out in the presence of high concentrations of salt such as gel filtration,
hydroxyapatite, hydrophobic, and sulfate-mediated chromatography.
Procedures that take place at low salt concentrations such as ion exchange

chromatography can be effective only if inactivation can be reversed.

ll

Enzymes purified to date from extreme halophiles include glutamine
synthetase, superoxide dismutase, or-amylasc, peroxidase, and
ketohexokinase (5, 32, 37, 39, 46 ).

Unlike Bacteria characterized by a single cell wall polymer, the murein,
Archaea exhibit a variety of non-murein polymers with remarkable
structural and chemical diversity (51). In contrast to the ester linkage-based
bacterial and eucaryotic lipids, archaeal core lipids are isopra(e)nyl glycerol
ethers, and all lipids in Archaea contain a 2,3-sn-glycerol and Czo-CZO, C20-
C25 or CZS-C25 diethers (31). The polar lipids of halophilic Archaea are
phospholipids and glycolipids.

Comparison of the central metabolism (i.e. conversion of hexose sugars
to pyruvate, and then to CO2 and H20 via the citric acid cycle) in Archaea,
Bacteria, and Eucarya suggests that the basic central pathways were
established before the divergence of the three domains, with variations
occurring thereafter (16). Halophilic archaeal membranes possess typical
components of bacterial respiratory system including flavoproteins,
quinones and cytochromes of types a, b, c, and o; the only difference being

that a high salt concentration is required for maintaining structure and

12

firnction of the respiratory chain in extreme halophiles (26).

The genomes of halophilic Archaea are similar in size to those of
Bacteria, both chromosomes and widely distributed plasmids are circular.
The G+C content of the bulk of the DNA is > 60 mol% and total DNA can
be separated into two fractions according to G+C content in most cases.
Chromosomal sequences of 62 to 71 mol% G+C constitute the major
fraction, whereas the more A + T rich minor fraction (55 to 58 mol% G+C)
consists mainly of covalently closed circular DNAs (cchNA). The
genomic instability of Halobacterium halobium due to insertion elements is
not a common feature of all halophilic Archaea (44).

Genetics in extreme halophiles is still in the early stage. Bi-directional
genetic exchange has been demonstrated in H. volcanii which requires
cell-cell contact, but no cell fusion occurring (40). Polyethylene glycol
(PEG) mediated spheroplast transformation of halophilic Archaea has been
successful in transforming plasmids and linear genomic DNA, as well as
phage DNA transfection (10). Shuttle vectors with selective markers
(mevinolin and novobiocin) which can replicate in both H. volcanii and E.

coli have been developed (28, 36). Transposons with selectable marker

13

genes have also been constructed, although no natural transposons have
been detected in Archaea (18). With these useful tools, a modest number of
halophilic archaeal genes has been characterized including genes encoding
rRNAs, ribosomal proteins, bacteriorhodopsin, halorhodopsin, gas-vesicle
proteins, and superoxide dismutase (4, 30, 38, 41, 49). A consensus
TTA(A/T) sequence for the archaeal promoters is usually located 27i4 bp
upstream from the translation start codon (5 7). It has been calculated that
Methanococcus jannaschii uses initiation codons ATG, GTG, and TTG
about 70%, 25%, and 5% of the time respectively (7). All halophilic rRNA
operons seem to share a common l6S-23S-SS gene organization

reminiscent of that found in eubacteria (41).

III. Aerobic degradation of phenylpropanoids by microorganisms
Degradation of the aromatic compound 3-phenylpropionic acid by
Haloferax sp. D1227 was the focus of this study. As aromatic compounds
with aliphatic side chains, phenylpropanoids present microorganisms with a
choice as to their mode of attack. One choice is to undergo hydroxylation

of the aromatic ring prior to degradation of the side chain as demonstrated

 

 

 

14

in a species of Achromobacter and two strains of Pseudomonas (3, 11, 15).
Another is shortening of side chains first via a mechanism similar to fatty
acid B-oxidation cycle as suggested by Webley et al. (50) based on whole
cell experiments, but not enzyme studies. The fatty acid B-oxidation
pathway (Figure 1.2) is usually induced by growth on oils, fatty acids or
alkanes. After entering the cells via various mechanisms, fatty acids are
first activated to acyl-CoA thioesters by chain length-specific acyl-CoA
synthetases prior to their oxidative degradation. In the following acyl-CoA
dehydrogenase reaction, one molecule of FAD, which is often tightly bound
to the bacterial and mitochondrial enzymes, is reduced and reoxidized
through the electron transfer chain yielding ATP; whereas the peroxisomal
enzyme is a flavin enzyme in which the flavin is reoxidized by oxygen
directly producing H202. After the 2,3-enoyl-CoA hydratase step, another
dehydrogenase reaction is carried out by NAD+-dependent 3-hydroxyacyl-
CoA dehydrogenase. The generated NADH is linked to oxidative
phosphorylation in bacteria and mitochondria, and to glycerol-3-phosphate
dehydrogenase in the peroxisomes. The final reaction of the cycle involves

the CoA-dependent cleavage of the acyl chain by 3-ketoacyl-CoA thiolase

 

 

15

R-CHZ-CHZ-CO-S-CoA FAD/flavin

  
 

FADH2/reduced flavin

R-CO-S-CoA
CH3-CO-S-CoA
R-CH=CH-CO-S-C0A
4 H10
2
H-S-CoA
R-CO-CH,-CO-S-CoA 3 R-CHOH-CH z-CO-S-CoA
NADH NAD+

Figure 1.2 Fatty acid B-oxidation cycle. Individual enzyme reactions: 1.
Fatty acyl-CoA dehydrogenase; 2. 2,3-enoyl-CoA hydratase; 3. 3-
hydroxyacyl-CoA dehydrogenase; 4. 3-ketoacyl-CoA thiolase. (Adapted
from Ratledge, 1994)

16

resulting in a two carbon shorter fatty acyl-CoA molecule, which reenters
the cycle without further activation (47). Although in mitochondria, all
enzymes of the B-oxidation appear as separable, mono-function proteins,
there is a remarkable degree of structural organization in the bacterial and
peroxisomal enzymes (56).

After scission of the aliphatic side chains, the resulting aromatic
compounds are usually transformed by monooxygenases and dioxygenases
into central intermediates such as catechol, protocatechuate and gentisate
before ring cleavage (24). It is generally accepted that dihydroxylation is a
prerequisite for enzymatic fission of the aromatic ring, which renders the
aromatic ring more reactive to electrophilic attack (27). A number of
different pathways for transforming benzoic acid into the dihydroxylated
intermediates have been demonstrated (Figure 1.3). As shown in pathwayl ,
some species such as Alcaligenes eutrophus and Pseudomonas arvilla
metabolize benzoate to catechol by benzoate 1,2-dioxygenase, a two-
component ring-hydroxylating dioxygenase. One component is the
NAD(P)H reductase containing a FAD and a [2F e-ZS] redox center and the

second component is the terminal dioxygenase in an (orB)3 configuration

l7

 

COOH
benzoate
1 Pathway 1 J Pathway 2 1 Pathway 3 J Pathway 4
00°“ COOH coscm
0H
OH
H OH
catechol 4-hydroxybenzoatc 3-hydroxybenzoatc benzoleoA
/ \CHO COOH >< i
COOH
coo
C000: Ox COOH COSCoA
OH OH O Q
muconic acid 2-hydroxymuconic H @014 OH
serrualdehyde protocatechuate 3-hydroxybenzoleoA
gentisate l
/ \ I
COOH O
H o= OH
\
COOH COOH OH gentisate
COOH
COOH maleylpyruvate '

mucomc 301d muconic seruialdehyde

ICOOH
O:

i 1 mil
\

OH

maleylpyruvate

l

3-carboxy 2-hydroxy-4-carboxy 1

Figure 1.3 Aerobic benzoate degradation pathways in various
microorganisms.

18

(55). Pathway 2 and 3 were demonstrated in some members of
Pseudomonas, Streptomyces and Bacillus where benzoate was metabolized
to gentisate or protocatechuate via either 3- or 4-hydroxybenzoate (12, 13).
4-Hydroxybenzoate-3-hydroxylase and 3-hydroxybenzoate-6—hydroxylase
involved in these two pathways have been studied in detail, both are single-
component flavoproteins with NAD(P)H as external electron donors (27,
52). Recently, Altenschmidt et al. (1) proposed a fourth benzoate
degradation pathway involving CoA derivatives. Their research shows that
a facultative denitrifying Pseudomonas strain KB740 metabolizes benzoate
aerobically via benzoleoA and 3-hydroxybenzoleoA rather than via free
acids, and 3-hydroxybenzoleoA is further degraded via gentisate. The
aerobic benzoate-CoA ligase in this pathway is not closely related to its
anaerobic counterpart based on immunological comparison. The activity of
the purified benzoleoA-3-monooxygenase is dependent on a flavin
nucleotide (FAD or FMN), NADPH, and benzoleoA. Although the
involvement of CoA thioesters in the anaerobic degradation of aromatic
compounds is well known, only a few reports have described aerobic

pathways with CoA derivatives as intermediates, i.e. the metabolism of 2-

l9

aminobenzoate and the dehalogenation of 4-halobenzoate (6, 14). The
authors postulated that one reason for utilizing CoA thioesters in aerobic
metabolism is that the degradation of naturally abundant C‘s-C3 compounds
by B-oxidation results in benzoleoA and analogues, and it is advantageous
to metabolize these compounds directly since free aromatic acids are
membrane-permeable. The dihydroxylated aromatic ring of the central
intermediates are opened by ring-cleavage dioxygenases leading to

intermediates of the tricarboxylic acid cycle (8).

References

l. Altenschmidt, U., B. Oswald, E. Steiner, H. Herrmann, and G. Fuchs.
1993. New aerobic benzoate oxidation pathway via benzoyl-coenzymeA

and 3-hydroxybenzoyl-coenzymeA in a denitrifying Pseudomonas. J.
Bacteriol. 175: 4851-4858.

2. Amils, R., P. Cammarano, and P. Londei. 1993. Translation in archaea.
393-438. In: M. Kates et al. (ed.), The biochemistry of Archaea
(Archaebacteria). Elsevier Science Publishers, New York.

3. Blakley, E. R. and F. J. Simpson. 1964. The microbial metabolism of
cinnamic acid. Can. J. Microbiol. 10: 175-185.

4. Blanck, A., and D. Oesterhelt. 1987. The halo-opsin gene. 11.
Sequence, primary structure of halorhodopsin and comparison with
bacteriohodopsin. EMBO J. 6: 265-273.

 

20

5. Brown-Peterson, N. J ., and M. L. Salin. 1993. Purification of a catalase-
peroxidase from Halobacterium halobium: characterization of some unique
properties of the halophilic enzyme. J. Bacteriol. 175: 4197-4202.

6. Buder, R., and G. Fuchs. 1989. 2-Aminobenzoyl-CoA
monooxygenase/reductase, a novel type of flavoenzyme. Purification and
some properties of the enzyme. Eur. J. Biochem. 185: 637-643.

7. Venter, J. C., C. J. Bult, 0. White, G. J. Olsen, L. Zhou, R. D.
Fleischmann, G. G. Sutton, J. A. Blake, L. M. FitzGerald, R. Clayton, J. D.
Gocayne, C. I. Reich, R. Overbeek, E. F. Kirkrress, K. G. Weinstock, J. M.
Merrick, A. Glodek, J. L. Scott, N. S. M. Geoghagen, J. F. Weidman, J. L.
Fuhrrnann, M. C. Hanna, M. D. Cotton, K. M. Roberts, M. A. Hurst, B. P.
Kaine, M. Borodovsky, H. Klenk, C. M. Fraser, H. O. Smith, C. R. Woese,
A. R. Kerlavage, B. A. Dougherty, J. Tomb, M. D. Adams, D. Nguyen, T.
R. Utterback, J. M. Kelley, J. R. Peterson, P. W. Sadow. 1996. Complete
genome sequence of the methanogenic archaeon, Methanococcus
jannaschii. Science. 273: 1058-1073.

8. Cain, R. B. 1980. The uptake and catabolism of lignin-related aromatic
compounds and their regulation in microorganisms. 21-56. In: T. K. Kirk
et al. (ed.), Lignin biodegradation: microbiology, chemistry, and potential
applications. CRC press, Inc., Florida.

9. Christian, J. H. B., and J. A. Waltho. 1962. Solute concentrations within
cells of halophilic and non-halophilic bacteria. Biochem. Biophys. Acta.
65: 506-508.

10. Cline, S. W., W. L. Lam, R. L. Charlebois, L. C. Schalkwyk, and W. F.
Doolittle. 1989. Transformation methods for halophilic archaebacteria.
Can. J. Microbiol. 35: 148-152.

11. Coulson C. B., and W. C. Evans. 1959. Microbiological degradation of
trans-cinnamic acid by soil Pseudomonads. Chem. & Ind. 543-544.

12. Crawford, R. L. 1975. Degradation of 3-hydroxybenzoate by bacteria
of the genus Bacillus. Appl. Microbiol. 30: 439-444.

21

13. Crawford, R. L. 1976. Pathways of 4-hydroxybenzoate degradation
among species of Bacillus. J. Bacteriol. 127: 204-210.

14. Crooks, G. P., and S. D. Copley. 1993. A surprising effect of leaving
group on the nucleophilic aromatic substitution reaction catalyzed by 4-
chlorobenzoyl-CoA dehalogenase. J. Am. Chem. Soc. 115: 6422-6423.

15. Dagley, S., P. J. Chapman, and D. T. Gibson. 1965. The metabolism of
B-phenylpropionic acid by an Achromobacter. Biochem. J. 97: 643-650.

16. Danson, M. J. 1993. Central metabolism of the archaea. 1-24. In: M.
Kates et al. (ed.), The biochemistry of Archaea (Archaebacteria). Elsevier
Science Publishers, New York.

17. Dennis, P. P., and L. C. Shimmin. 1997. Evolutionary divergence and
salinity-mediated selection in halophilic archaea. Microbiol. Mol. Biol.
Rev. 61: 90-104.

18. Dyall-Smith, M. L., and W. F. Doolittle. 1994. Construction of
composite transposons for halophilic Archaea. Can. J. Microbiol. 40: 922-
929.

19. Dym, O., M. Mevarech, and J. L. Sussman. 1995. Structural features

that stabilize halophilic malate dehydrogenase from an archaebacterium.
Science. 267: 1344-1346.

20. Emerson, D., S. Chauhan, P. Oriel, and J. A. Breznak. 1994. Haloferax
sp. D1227, a halophilic Archaeon capable of grth on aromatic
compounds. Arch. Microbiol. 161: 445-452.

21. Forterre, P., and C. Elie. 1993. Chromosome structure, DNA
topoisomerases, and DNA polymerases in archaebacteria (archaea). 325-
365. In: M. Kates et al. (ed.), The biochemistry of Archaea
(Archaebacteria). Elsevier Science Publishers, New York.

22. Frolow, F., M. Hare], J. L. Sussman, M. Mevarech, and M. Shoham.
1996. Insights into protein adaptation to a saturated salt environment from

22

the crystal structure of a halophilic 2Fe-ZS ferredoxin. Nat. Struct. Biol. 3:
452-458.

23. Garrett, R. A., C. Aagaad, M. Anderson, J. Z. Dalgaard, J. Lykke-
Anderson, H. T. N. Phan, S. Trevisanato, L. Ostergaard, N. Larsen, and H.
Leffers. 1994. Archaeal rRNA operons, intron splicing and homing

endonucleases, RNA polymerase operons and phylogeny. System. Appl.
Microbiol. 16: 680-691.

24. Gibson, D. T. and V. Subramanian. 1984. Microbial degradation of
aromatic hydrocarbons. 181-252. In: D. T. Gibson (ed.), Microbial
degradation of organic compounds. Marcel Dekker Inc., New York.

25. Hahlbrock, K., and D. Scheel. 1989. Physiology and molecular
biology of phenylpropanoid metabolism. Annu. Rev. Plant Physiol. Plant
Mol. Biol. 40: 347-369.

26. Hallberg, G. C., and A. Colmsjo. 1989. Four different b-type
cytochromes in the halophilic archaebacteriurn Halobacterium halobium.
Arch. Biochem. Biophys. 272: 130-136.

27. Harayama, S., and M. Kok. 1992. Functional and evolutionary
relationships among diverse oxygenases. Annu. Rev. Microbiol. 46: 565-
601.

28. Holmes, M. L., and M. L. Dyall-Smith. 1990. A plasmid vector with a
selectable marker for halophilic archaebacteria. J. Bacteriol. 172: 756-761.

29. Holt, J. G., N. R. Kieg, P. H. A. Sneath, J. T. Sraley, and S. T. Williams.
1993. Bergey’s manual of determinative bacteriology. Ninth edition. p.
12. Williams & Wilkins. Baltimore.

30. Home, M., C. Englest, and F. Pfeifer. 1988. Two gas vacuole proteins
in Halobacterium halobium. Mol. Gen. Genet. 213: 59-464.

23

31. Kamekura, M., and M. Kates. 1988. Lipids of ha10philic
archaebacteria. Vol. 2: 25-65. In: F. Rodriguez-Valera (ed.), Halophilic
bacteria. CRC Press, Inc., Florida.

32. Kobayashi, T., H. Kanai, T. Hayashi, T. Akiba, R. Akaboshi, and K.
Horikoshi. 1992. Haloalkaliphilic maltotriose-forming a-amylase from the

archaebacterium Natronococcus sp. Strain Ah-36. J. Bacteriol. 174: 3439-
3444.

33. Langer, D., J. Hain, P. Thuriaux, and W. Zillig. 1995. Transcription in
archaea: similarity to that in Eucarya. Proc. Natl. Acad. Sci. USA. 92:
5768-5772.

34. Nikolov, D. B., and S. K. Burley. 1997. RNA polymerase II
transcription initiation: a structural view. Proc. Natl. Acad. Sci. USA. 94:
15-22.

35. Lanyi, J. K. 1974. Salt-dependent properties of proteins from
extremely halophilic bacteria. Bacteriol. Rev. 38: 272-290.

36. Lam, W. L., and W. F. Doolittle. 1989. Shuttle vectors for the
archaebacterium H. Volcanii. Proc. Natl. Acad. Sci. USA. 86: 5478-5482.

37. Manitz, B., and A. W. Holldorf. 1993. Purification and properties of
glutamine synthetase from the archaebacterium Halobacterium salinarium.
Arch. Microbiol. 159: 90-97.

38. May, B. P., P. Tam, and P. P. Dennis. 1990. The expression of the
superoxide dismutase gene in Halobacterium cutirubrum and H. volcanii.
Can. J. Microbiol. 35: 171-175.

39. May, B. P., and P. P. Dennis. 1987. Superoxide dismutase from the
extremely halophilic archaebacterium Halobacterium cutirubrum. J.

Bacteriol. 169: 1417-1422.

40. Mevarech, M., and R. Werczberger. 1985. Genetic transfer in
Halobacterium volcanii. J. Bacteriol. 162: 461-462.

24

41. Mevarech, M., S. Hirsch-Twizer, S. Goldman, E. Yakobson, H.
Eisenberg, and P. P. Dennis. 1989. Isolation and characterization of the

rRNA gene clusters of Halobacterium marismortui. J. Bacteriol. 171:
3479-3485.

42. Olsen, G. J ., and C. R. Woese. 1997. Archaeal genomics: an overview.
Cell. 89:991-994.

43. Olsen, G. J ., and C. R. Woese. 1996. Lessons from an Archaeal
genome: what are we learning from Methanococcusjannaschii? Trends.
Genet. 12: 377-379.

44. Pfeifer, F. 1988. Genetics of halobacteria. Vol. 2: 105-133. In: F.
Rodriguez-Valera (ed.), Halophilic bacteria. CRC Press, Inc., Florida.

45. Rochkind, M. L., J. W. Blackburn, and G. S. Sayler. 1987. Metabolism
of non-chlorinated compounds. 45-68. In: M. L. Rochkind et al. (ed.),
Microbial decomposition of chlorinated aromatic compounds. U. S.
Environmental Protection Agency, Cincinnati, Ohio.

46. Rangaswamy, V., and W. Altekar. 1994. Ketohexokinase (ATP: D-
fi'uctose l-phosphotransferase) from a halophilic archaebacterium,

Haloarcular vallismortis: purification and properties. J. Bacteriol. 176:
5505-5512.

47. Ratledge, C. 1994. Biodegradation of oils, fats and fatty acids. 89-
141. In: C. Ratledge (ed.), Biochemistry of rrricrobial degradation. Kluwer
academic publishers, Netherlands.

48. Reistad, R. 1970. On the composition and nature of the bulk protein of
the extremely halophilic bacteria. Arch. Microbiol. 71: 353-360.

49. Shimmin, L., and P. P. Dennis. 1989. Characterization of the L11, L1,
L10 and L12 equivalent ribosomal protein gene cluster of the halophilic
archaebacteria Halobacterium cutirubrum. EMBO J. 8: 1225-1235.

 

25

50. Webley, D. M., R. B. Duff, and V. C. Farmer. 1955. Beta-oxidation of
fatty acids by Nocardia opaca. J. Gen. Microbiol. 13: 361-369.

51. Wieland, F. 1988. The cell surfaces of halobacteria. Vol. 2: 55-65. In:
F. Rodriguez-Valera (ed.), Halophilic bacteria. CRC Press, Inc., Florida.

52. Wierenga, R. K., K. H. Kalk, J. M. van der laan, J. Drenth, J.
Hofsteenge, W. J. Weijer, P. A. Jekel, J. J. Beintema, F. Muller, and W. J.
H. van Berkel. 1982. The structure of p-hydroxybenzoate hydroxylase.
11-19. In V. Massey et al. (ed.), F lavins and flavoproteins. Elsevier
Science Publishers, New York.

53. Woese, C. R., O. Kandler, and M. L. Wheelis. 1990. Toward a natural
system of organisms: proposal for the domains Archaea, Bacteria, and
Eucarya. Proc. Natl. Acad. Sci. USA. 87: 4576-4579.

54. Woese, C. R., and G. E. Fox. 1977. Phylogenetic structure of the
procaryotic domain: the primary kingdoms. Proc. Natl. Acad. Sci. USA.
74: 5088-5090.

55. Yamaguchi, M., and H. Fujisawa. 1982. Subunit structure of
oxygenase component in benzoate 1,2-dioxygenase system from
Pseudomonas arvilla C-l. J. Biol. Chem. 257: 12497-12502.

56. Yang, S. Y., and H. Schulz. 1983. The large subunit of the fatty acid
oxidation complex from Escherichia coli is a multifunctional polypeptide.
J. Biol. Chem. 258: 9780-9785.

57. Zillig, W., P. Palm, H. Klenk, D. Langer, U. Hudepohl, J. Hain, M.
Lanzendorfer, and 1. H012. 1993. Transcription in archaea. 367-391. In:
M. Kates et al. (ed.), The biochemistry of Archaea (Archaebacteria).
Elsevier Science Publishers, New York.

Chapter 2
DEGRADATION OF 3-PHENYLPROPIONIC ACID BY

HALOFERAX SP. D1227

Abstract

Haloferax sp. D1227, isolated from soil contaminated with highly saline
oil brine, is the first halophilic archaeon to demonstrate the utilization of
aromatic compounds (i.e., 3-phenylpropionic acid, cinnamic acid and
benzoic acid) as sole carbon and energy sources for growth. The
degradation of 3-phenylpropionic acid in this strain was studied to examine
the strategies utilized by Archaea to metabolize aromatic compounds.
Based on the following findings: (1) the extracellular accumulation of
cinnamic acid, benzoic acid, 3-hydroxybenzoic acid, and gentisic acid in
cultures of Hf D1227 grown on 3-phenylpropionic acid; (2) the presence of
an 3-phenylpropiony1CoA dehydrogenase; (3) the ATP, CoA and NAD-
dependent conversion of cinnamic acid to benzoleoA; and (4) the presence

of gentisate 1,2-dioxygenase, we propose that Hf D1227 metabolizes

26

27

3-phenylpropionic acid by initial 2-carbon shortening of the side chain to
benzoleoA via a mechanism similar to fatty acid B—oxidation, followed by
aromatic degradation using a genti sate pathway. The upper aliphatic
pathway from 3-phenylpropionic acid to benzoic acid is separately

regulated from the lower gentisate pathway.

Introduction

Haloferax sp. D1227, isolated from soil contaminated with highly saline
oil brine near Grand Rapids, Michigan, is a halophilic archaeon requiring 2
M NaCl for optimal growth. To date, Hf D1227 is the only reported
archaeon capable of aerobic metabolism of aromatic compounds (i.e.,
benzoic acid, cinnamic acid, and 3-phenylpropionic acid) as sole carbon and
energy sources for grth (12). Although the pathways for degradation of
aromatic compounds in bacteria and fungi have been elucidated in detail (5,
13, 25), little is known about the aromatic catabolism by Archaea. Since the
recognition that Archaea, consisting of methanogens, extreme thermophiles
and extreme halophiles, represents a third domain of life phylogenetically
distinct fi'om Bacteria and Eucarya (33), biochemical and genetic research

has elucidated some unusual features of these organisms and provided new

28

insights into their evolutionary relationships with bacterial and eucaryotic
organisms (11, 18, 24). The isolation of Hf D1227 provides an
opportunity to investigate the strategies utilized by Archaea for catabolism
of aromatic compounds and to compare an aromatic degradation pathway in
Archaea with those in Bacteria and Eucarya.

3-Phenylpropionic acid is a member of the phenylpropanoid family
comprising a wide variety of C6-C3 compounds synthesized by plants from
phenylalanine. Phenylpropanoids are important in plant physiology for
synthesis of lignin, flavonoids, insect repellents, UV protectants, and signal
molecules (15). It has been shown that the microbial degradation of 3-
phenylpropionic acid can occur via two different routes. In one route
demonstrated in a species of Achromobacter and two strains of
Pseudomonas (2, 6, 9), the aromatic ring is first oxidized and opened,
followed by the degradation of the resulting aliphatic segment. An
alternative route for the degradation of 3-phenylpropionic acid is catabolism
of the side chain followed by aromatic ring fission. This latter pathway was
suggested by Webley et a1. (30) based on their observation of the transient

accumulation of cinnamic acid and benzoic acid by Nocardia opoca grown

29

with 3-phenylpropionic acid. They proposed that 3-phenylpropionic acid
was metabolized by two carbon shortening of the side chain first, via a
mechanism similar to fatty acid B-oxidation, resulting in benzoleoA. The
enzymes involved in this oxidation were not investigated, nor was the
degradation of benzoleoA. Under aerobic conditions, benzoate is usually
transformed into a few key intermediates including catechol,
protocatechuate and gentisate, followed by aromatic ring cleavage by ring-
fission dioxygenases (13). Recently, Altenschmidt et al. (1) demonstrated a
new aerobic benzoate degradation pathway involving CoA derivatives.
Their research showed that Pseudomonas KB740, a facultative denitrifying
Pseudomonas species, metabolized benzoate to 3-hydroxybenzoleoA via
benzoleoA with further 5-hydroxy1ation to gentisate.

In this paper, we present results which indicate that the degradation of 3-
phenylpropionic acid by Hf. D1227 is initiated by B-oxidation of the side
chain to produce benzoleoA, which is subsequently metabolized via a

gentisate pathway.

30

Materials and Methods

Materials. All chemicals used in this study were reagent grade. 3-
phenylpropionic acid, cinnamic acid and benzoic acid were purchased from
Aldrich Chemical Co. (Milwaukee, WI.). Catechol, protocatechuic acid,
gentisic acid, salicylic acid, 3-hydroxybenzoic acid, 4-hydroxybenzoic acid,
oleic acid, lauric acid, palrrritic acid, Triton X-100, ATP, CoenzymeA and
2,2'-dipyridyl were obtained from Sigma chemical Co. (St. Louis, MO.).
HPLC grade acetonitrile was purchased from EM Science (Gibbstown, NJ .).

Microorganism, media, and growth conditions. Haloferax sp. D1227
has been described (12). For grth of Hf D1227 on various carbon
sources, mineral salts medium (BS3) (12) of the following composition (in
g/l) was used: (NH 4)ZSO 4, 0.33; KCl, 6.0; MgC12-6HZO, 12.1; MgSO4.
7HZO, 14.8; KHZPO4, 0.34; CaClZ-HZO, 0.36; and NaCl, 100. 1 ml/l of a

trace element solution (32) was also added prior to sterilization and the pH
was adjusted to 6.9 with KOH. Filter-sterilized grth substrates were
added afier sterilization. For rich medium (BSYT), BS3 mineral salts
medium was supplemented with 3 g/l yeast extract and 3 g/l tryptone. For

solid media, 15 g/l Bacto-agar was added prior to autoclaving. Cells were

31

cultured aerobically in Erlenmeyer flasks containing 1/4 volume of BS3
mineral salt medium supplemented with various growth substrates. For
grth on fatty acids, the sole carbon source was 1 mM oleic acid, lauric
acid, or palmitic acid with or without 0.4% (v/v) Triton X-100 to disperse
fatty acids (34). Each flask was inoculated with a mid-log phase culture
grown in BSYT rich medium to give an initial OD600nm approximately 0.02.
Flasks were incubated at 37°C on a rotary shaker at 200 rpm. Growth was
monitored by measuring OD600nm on a Gilford DU spectrophotometer except
that growth on fatty acids was determined by viable count plating.

HPLC analysis. Aromatic compounds were analyzed by reverse-phase
HPLC on a Waters model chromatography (Waters, Millipore Corp.,
Milford, Mass.) equipped with a Waters 486 tunable absorbance detector set

at 254nm, a Waters 746 integrator and Waters model 501 solvent delivery

system. Separation was achieved on a Nova-PakTM C18 column (150x39
mm) at a flow rate of 0.8 ml/min. The injection volume was 20 ul. Free
aromatic acids were eluted using 2% acetonitrile in 200 mM ammonium
acetate buffer (pH 6.5). CoA thioesters were eluted using 15% acetonitrile

in 200 mM ammonium acetate buffer (pH 5.5). For CoA derivatives for

32

which standards were unavailable, 3 ul of 10N NaOH was added to 300 pl
samples to hydrolyze CoA thioesters to their corresponding free acids (31).
Aromatic acids, benzoleoA, and 3-hydroxybenzoleoA were identified by
comparison of retention times with those of authentic standards.

Resting cell experiments with added 2,2'-dipyridyl. Cells grown on 3
mM 3-phenylpropionic acid, cinnamic acid, benzoic acid, or pyruvate were
harvested when the culture (250 ml) reached mid-log phase at an OD600nm of
approximately 0.7. Following centrifugation at 10,600Xg for 30 rrrin, cells
were washed twice with 100 mM potassium phosphate buffer (pH 7.0)
containing 2 M KCl and resuspended in 25 m1 of the same buffer. Each 125
ml Erlenmeyer flask contained in a total volume of 10 ml: 100 mM
potassium phosphate buffer (pH 7.0) containing 2 M KCl, 4 ml of the cell
suspension, 5 mM 2,2'-dipyridyl, and 1 mM 3-phenylpropionic acid,
cinnamic acid, benzoic acid, or 3-hydroxybenzoic acid. Cells were
incubated at 37°C on a rotary shaker at 200 rpm. Samples were filtered
through 0.2 pm nylon filters (Scientific Resources Inc., Eatontown, NJ .),

and analyzed immediately by HPLC.

33

Preparation of cell-free extracts. Hf D1227 was grown in 250 ml BS3
medium containing 3 mM 3-phenylpropionic acid, cinnamic acid, benzoic
acid or pyruvate. Cells in mid-log phase were harvested by centrifugation at
10,600xg for 30 min, washed twice with 30 m1 100 mM potassium
phosphate buffer (pH 7.0) containing 2 M KCl, and resuspended in 10 m1 of
the same buffer. Using a 4710 Series Ultrasonic Homogenizer (Cole-
Parmer Instrument Co., Chicago, IL.), 2.5 m1 aliquots of the suspension
were sonicated four times (30 seconds at 50 W followed by 2 min of
cooling) in an ice bath. Unbroken cells and cell debris were removed by
ultracentrifugation (245,300xg for 1 hour at 4 °C). The supematants were
then diafiltrated using Centricon-IO (MW cut-off : 10,000. Arnicon, Inc.,
Beverly, MA.) at 4°C with three volumes of 100 mM potassium phosphate
buffer (pH 7.0) containing 2 M KCl to remove small molecules in the
extracts. The concentrated extracts were diluted with the same buffer to 10
ml for enzyme analysis.

Enzyme assays. All enzyme assays were performed at 37°C, unless
stated otherwise. 3-Phenylpropionic acid-CoA ligase, cinnamic acid-

CoA ligase, benzoic acid-CoA ligase, and 3-hydroxybenzoic acid-CoA

34

ligase. The assay mixture for ligase activity contained in a total volume of
0.5 ml: 100 mM potassium phosphate buffer (pH 7.0) containing 2 M KCl,
0.5 mM 3-phenylpropionic acid, cinnarrric acid, benzoic acid or 3-
hydroxybenzoic acid, 2 mM ATP, 2 mM Coenzyme A, 10 mM MgCl2 and
50 ul of cell-free extract. The ligase activities were determined by
measuring the rate of the appearance of CoA thioester products using
HPLC.

3-PhenylpropionleoA dehydrogenase. In order to assay 3-
phenylpropionleoA dehydrogenase activity, 3-phenylpropionleoA,
which is not commercially available, was synthesized from 3-
phenylpropionic acid with Hf D1227 cell-free extracts utilizing the 3-
phenylpropionic acid-CoA ligase reaction described above. Since the
reaction catalyzed by 3-phenylpropionleoA dehydrogenase required no
added cofactor, cinnamleoA started to appear in the ligase reaction
mixture upon the formation of 3-phenylpropionleoA. The 3-
phenylpropionleoA dehydrogenase activity was determined by measuring

the rate of cinnamleoA production using HPLC.

35

Conversion of cinnamleoA to benzoleoA. Since cinnamleoA is
not commercially available, it was synthesized from cinnamic acid with Hf
D1227 cell-free extracts utilizing cinnamic acid-CoA ligase reaction
described above. Once the formation of cinnamleoA ceased, NAD was
added to the reaction mixture (0.5 ml) to a final concentration of 1 mM to
initiate the conversion of cinnamleoA to benzoleoA. The simultaneous
disappearance of cinnamleoA and appearance of benzoleoA were
measured using HPLC.

BenzoleoA-3-hydroxylase, 3-hydroxybenzoleoA-6-hydroxylase,
benzoate-B-hydroxylase, and 3-hydroxybenzoate-6-hydroxylase.
BenzoleoA-3-hydroxylase, 3-hydroxybenzoleoA-6-hydroxylase,
benzoate -3-hydroxy1ase, and 3-hydroxybenzoate-6-hydroxylase activities
were assayed using the methods modified from those described by Niemetz
et al.(23), Kiemer et al. (19), Wang et al. (29), Suarez et a1. (26), van Berkel
et al. (27), and Groseclose et a1. (14). The reaction mixture in a total
volume of 1 ml contained: 100 mM potassium phosphate buffer (pH 7.0)
containing 2 M KCl, or 100 mM Tris-HCl buffer (pH 7 .0) containing 2 M

KCl; 0.25 mM benzoleoA, 3-hydroxybenzoleoA, benzoate, or

36

3-hydroxybenzoate; 0.1 mM NADH or NADPH or in combination with 0.1
mM FAD; 200 pl supernatant or resuspended pellet of sonicated or 0.1%
Triton X-100 treated cells. The stabilizers, 0.1 mM DTT, 0.1% glycerol,
and 0.1 mM EDTA, were also added to the reaction mixture, either
individually or in combination. The enzyme activities were determined by
measuring the appearance of hydroxylation products on HPLC.

Gentisate 1,2-dioxygenase. Gentisate 1,2-dioxygenase was assayed at
room temperature by measuring the formation of maleylpyruvate at 334 nm
(20) with a Perkin Elmer double beam 124 spectrophotometer. The reaction
mixture (2 m1) contained 0.25 mM gentisic acid, 50 1.11 cell-free extract, and
100 mM potassium phosphate buffer (pH 7.0) containing 2 M KCl. The
reference cuvette contained the same reaction mixture except gentisic acid
was omitted. The molar absorption coefficient for maleylpyruvate (a =
10,800) (7) was used to calculate enzyme activity. This assay was
confirmed by measurement of the disappearance of gentisate using HPLC.

cis-trans Isomerase. Since maleylpyruvate is not commercially
available, it was synthesized from gentisic acid by gentisate 1,2-

dioxygenase as described above. When the formation of maleylpyruvate

37

ceased, glutathione was added to the reaction mixture to a final
concentration of 1 mM. The activity of cis-trans isomerase was determined
by measuring the disappearance of maleylpyruvate at 334 nm (20).

Protein determination. Protein concentrations of crude enzyme extracts
were determined by the method of Bradford (3) with bovine serum albumin
dissolved in 100 mM potassium phosphate buffer (pH 7.0) containing 2 M
KC1 as the standard.

Nucleotide sequence analysis of small-subunit (SSU) rDNA. To
prepare genomic DNA of Hf D1227, cells were grown in 100 m1 BSYT
medium and harvested at mid-log phase by centrifugation at 10,600x g for
30 min. After washing twice with 100 mM potassium phosphate buffer (pH
7.0) containing 2 M KC1, the pellet was resuspended in 50 ml distilled water
for cell lysis. After removal of cell debris by centrifugation at 47,800xg for
30 min, the supernatant was extracted once with an equal volume of phenol,
and twice with an equal volume of phenol/chloroform. DNA in the aqueous
phase was precipitated with 2 volumes of ethanol and 0.1 volume of 3 M
sodium acetate (pH 5.2) at —16°C overnight. After centrifugation, the pellet

was rinsed with 1 ml of 70% ethanol, air dried, and resuspended in 1 ml of

38

H20 (22). This preparation was then treated with RNase at a final

concentration of 25 ug/ml and used as template DNA for PCR. The PCR
reaction mixture in a total volume of 100 u] contained: 2 u] of template

DNA (ca. 100 ng), 20 mM Tris-HCl (pH 8.4), 50 mM KC1, 1.5 mM MgC12,

0.25 mM of each dNTP, 2.5 U of Taq DNA polymerase, and 30 pmol of
each primer. Primers ARCH21BF [5'-TCCGCTTGATCC(C/T)GCC(A/G)
G-3'] and 1492R [5‘-GGTTACCTTGT TACGACTT-3'] have been
described (21). PCR reaction was canied out on Perkin Elmer GeneAmp
PCR System 2400 (Perkin Elmer, Norwalk, CT.) and thermal cycling
conditions were 94°C for 5 min followed by 35 cycles of 94°C for 30 sec,
55°C for 45 sec, and 72°C for 2.1 min. A final extension was performed at
72°C for 6.1 min. PCR product was purified with GENECLEAN II Kit
(B10 10] Inc., La Jolla, CA.) following the protocol recommended by the
manufacturer. Nucleotide sequencing was carried out at the Nucleic Acid
Sequencing Facility of Michigan State University with an ABI Prism
sequencer (Applied Biosystems). The following sequencing primers were
used: ARCH 21BF, 269R [5'-TACCGAT TATCGGCACGGTG-3‘], 356F [

5'-AGGCGCGAAAC CTT TACACT-3'], 59212 [5'-CGAAGGTTCATCGG

39

GAAATC-3'], 960R [5'-ATGTCCGGCGTT GAGTCCA A-3'], 1196R [5'-
ATTCGGGGCATACTG ACCTA-3'], l352F[5'-CGCATTTCAATAGAGT
GCGG-3'], and 1492R. An edited, contiguous sequence was constructed
from the data manually. A list of the known sequences that were most
similar to the SSU rRNA sequence of Hf D1227 was obtained using the
"Similarity Rank" routine at the Ribosomal Database Project (University of
Illinois, Urbana-Champaign). Sequences were then aligned manually with
Genome database Environment version 2.2 operating on a Sun SPARC
station. Similarity matrix was constructed with the Jukes and Cantor
correction for base changes. Phylogenetic trees were constructed from the
same alignment by distance and maximum-parsimony methods (the later
was bootstrapped with SEQBOOT). These analyses were run with PHYLIP

version 3.55c.

RESULTS
Small-subunit (SSU) rRNA sequence analysis. To confirm the earlier
identification of Hf D1227 as a Haloferax species (12), the small subunit

rRNA encoding gene from Hf. D1227 was amplified by PCR using

40

Archaea-specific forward primer ARCH21BF and "universal" reverse
primer 1492R. Sequence corresponding to E. coli SSU rRNA nucleotide
position 26 through 1491 was obtained. Hf D1227 showed 98.4% and
98.2% sequence similarities to Haloferax volcanii and Haloferax
mediterranei respectively. The unrooted phylogenetic tree constructed by
maximum-parsimony analysis grouped Hf. D1227 within the genus
Haloferax, and this grouping was supported by bootstrap value of 100% for
the node from which this strain and the other members of the genus radiate
(Figure 2.1). The distance method also grouped Hf D1227 within the genus
Haloferax (tree not shown). Out of the 26 nucleotide differences between
Hf. D1227 and its closest relative Hf volcanii, there were four dual-
compensatory differences corresponding to E. coli positions 128 and 233,
141 and 222,184 and 191, 835 and 851.

Aerobic growth of Hf. D1227 on aromatic compounds and fatty acids.
In addition to grth on 3-phenylpropionic acid, cinnamic acid and benzoic
acid previously described (12), Hf. D1227 also demonstrated the ability to
utilize 3-hydroxybenzoic acid (1 mM) as a sole carbon source. No growth

was observed with 4-hydroxybenzoic acid (1 mM), salicylic acid (1 mM),

41

.8886 588296 8 oococotfi fimd a $8qu82 83 2:. 89:88 80868983858
a no woman. REQQBE 28% 08 88:3 339 88% mo notion ouocoworfim fiN oSwE

EEQEEB Eatoaoeefiem

oocuuommmv o\.. m. 53.50%3 Extfiuueofium

N NQVZKLLQE «SUDOUQNQ:

 

 

 

833.88 gooooEm

 

 

 

 

83888.4 SackeoEm

 

 

 

 

3268.8qu Sackefiem

 

 

 

§u~3000 39090§0§32

 

 

 

88838 Eatfiogoeohez
.8§.\9§§QS EEQEQEE

 

 

 

 

 

 

838285: EEQEQBE

82a .8... E883:

 

 

 

 

 

823.8588 8.883%

 

Nd 3.. .8 .% REQQBE
888%

BBKQEE
uncombhmeon 8.8onme

 

 

 

 

 

 

 

 

 

 

 

.8563?» REQSGE

 

42

catechol (0.1 mM), protocatechuic acid (0.1 mM), gentisic acid (0.1 mM),
oleic acid (1 mM), lauric acid (1 mM), and palmitic acid (1 mM). Lack of
grth on gentisic acid was attributed to the inhibitory effect of either
gentisic acid or its autoxidation products, since growth of Hf D1227 on
benzoate was inhibited when gentisic acid was added to the medium.

The metabolites detected by HPLC analysis in the cultures of Hf D1227
grown on 3 mM 3-phenylpropionic acid, cinnamic acid, or benzoic acid
(Table 2.1) suggested that cinnamic acid, benzoic acid, 3-hydroxybenzoic
acid and gentisic acid were intermediates in 3-phenylpropionic acid
degradation.

Resting cell experiments with added 2,2'-dipyridyl. To observe
metabolites detected during growth in more detail, resting cell experiments
were carried out with added 2,2'-dipyridyl, which is an inhibitor of gentisate
1,2-dioxygenase (See enzyme activities), thus blocking further degradation
of gentisic acid. When 2,2'-dipyridy1 and 3-phenylpropionic acid were
incubated with resting cell suspensions of Hf D1227 pregrown on 3-
phenylpropionic acid, accumulations of the following metabolites were

observed (in the order of decreasing concentration): gentisic acid,

43

Table 2.1 Aromatic metabolites observed in 48 hour Hf D1227 cultures
grown on various substrates.

 

 

 

 

 

 

 

 

 

 

 

 

Grth substrate Aromatic metabolites Concentration

(3mM) observed in 48 hour culture (uM)
3-phenylpropionic cinnamic acid 0.061
acid benzoic acid 0.048
3-hydroxybenzoic acid 0.079

gentisic acid 0.035

cinnamic acid benzoic acid 0.086
3-hydroxybenzoic acid 0.095

gentisic acid 0.074

benzoic acid 3-hydroxybenzoic acid 0.104
gentisic acid 0.073

pyruvate none /

 

 

 

 

44

3-hydroxybenzoic acid, benzoic acid, and cinnamic acid (Figure 2.2). The
detection of these metabolites in high concentrations provided additional
evidence that Hf D1227 metabolized 3-phenylpropionic acid via cinnamic
acid, benzoic acid, 3-hydroxybenzoic acid and gentisic acid. When the
intermediates involved in the 3-phenylpropionic acid degradation pathway
and 2,2'-dipyridyl were added to 3-phenylpropionic acid-grown cell
suspension, the anticipated metabolites from the proposed pathway were
seen, i.e. benzoic acid, 3-hydroxybenzoic acid and gentisic acid
accumulated when cinnamic acid was added as the substrate; 3-
hydroxybenzoic acid and gentisic acid accumulated with benzoic acid as the
substrate; gentisic acid accumulated with 3-hydroxybenzoic acid as the
substrate (data not shown).

Enzyme activities. The following enzyme activities were found in the
soluble fraction of Hf D1227 cells. 3-Phenylpropionic acid-CoA ligase,
cinnamic acid-CoA ligase, benzoic acid-CoA ligase, and 3-hydroxy
benzoic acid-CoA ligase. Crude extracts prepared from cells grown on 3-

phenylpropionic acid demonstrated ATP and CoA-dependent ligase

activities which converted 3-phenylpropionic acid, cinnamic acid, benzoic

3-Phenylpropionic acid/Gentisic acid

45

 

 

 

 

1000 100
900 . 90 .
M A
800 ~ 80 it 2
8 3
700 - 7O .3 '53
O o
600 — 60 a 3
D m
A O
a 500 . 50 g E
V 400 ~ ~ 40 8 a
.2 g
300 — 30 §§
200 - . . 20 E z
W O
100 _ .__.. M 10
e Oflzée’.’. 0
0 c. _ 4" . 1 1 1 1 . . . f 0
0 30 60 90 120 150 180
Time (min)

Figure 2.2 Accumulation of intermediates when 3-
phenylpropionic acid and 2,2'-dipyridyl were added to 3-

phenylpropionic acid-grown cells

+ 3-phenylpropionic acid
-<>- benzoic acid

-O— gentisic acid

+ cinnarrric acid

+ 3-hydroxybenzoic acid

46

acid, and 3-hydroxybenzoic acid to their corresponding CoA derivatives.
Stimulation of ligase activities was observed with the addition of 10 mM
MgClz to the reaction mixtures (Table 2.2). These ligase activities were
also detected in cells grown on cinnamic acid, benzoic acid, or pyruvate. It
is not yet known whether these reactions are catalyzed by the same ligase or
different ones, and also whether they are constituous or inducible.
3-PhenylpropionleoA dehydrogenase. The activity of 3-
phenylpropionleoA dehydrogenase catalyzing the conversion of 3-
phenylpropionleoA to cinnamleoA was present in extracts of 3-
phenylpropionic acid-grown cells with a specific activity of 526:19 nmol
cinnamleoA produced/min mg protein and also in extracts of cinnamic
acid-grown cells with a specific enzyme activity of 711i24 nmol
cinnamleoA produced /min-mg protein. This enzyme activity was not
detectable in cells grown on benzoic acid or pyruvate. The reaction
catalyzed by the 3-phenylpropiony1CoA dehydrogenase proceeded without
any added cofactor. When 1 mM FAD, NAD, or NADP was added to the
reaction mixture, only FAD addition stimulated enzyme activity (50%). No

conversion of 3-phenylpropionic acid to cinnamic acid was observed

47

Table 2.2 Stimulation effect of MgClz on ligase activity

 

 

 

 

 

 

 

acid-CoA ligase

 

 

 

Ligase Specific enzyme activity (uM Stimulation
thioester product/min. mg protein) (fold)
0 mM MgClz 10 mM MgCl2
3-Phenylpropionic 1 9:03 7.3i0.9 3 .5
acid-CoA ligase,
cinnarrric acid- 1.1i0.2 67:07 6.0
CoA ligase
benzoic acid-CoA 3.6:0.5 15.8i0.4 4.0
ligase
3-hydroxybenzoic 0.4i0.05 1 .0i0.07 2.5

 

Note: Crude extracts prepared from 3-phenylpropionic acid-grown Hf

D1227 cells were used in this experiment.

 

48

without both ATP and Coenzyme A added to the reaction mixture. Flushing
the reaction mixture with nitrogen gas caused 34 % loss of enzyme activity,
suggesting the possible involvement of O2 in the reaction. Since 3-
phenylpropionleoA dehydrogenase activity was demonstrated only in the
soluble fraction of Hf D1227, membrane-associated electron transfer
proteins appear not to be involved in this dehydrogenation reaction.

Conversion of cinnamleoA to benzoleoA. Conversion of
cinnamleoA to benzoleoA was observed with cell-free extracts prepared
from cells grown on either 3-phenylpropionic acid, or cinnamic acid. This
conversion was NAD-dependent over a broad concentration range with
maximal conversion at 1 mM NAD (Figure 2.3). NADP and FAD were
ineffective. Extracts of benzoic acid or pyruvate-grown cells did not show
the ability to transform cinnamleoA to benzoleoA, and no conversion of
cinnamic acid to benzoic acid was detected without both ATP and
Coenzyme A added to the reaction mixture.

BenzoleoA-3-hydroxylase, 3-hydroxybenzoleoA-6-hydroxylase,
benzoate-3-hydroxylase, and 3-hydroxybenzoate-6-hydroxylase.

BenzoleoA, 3-hydroxybenzoleoA, benzoate and 3-hydroxybenzoate

Appearance of benzoleoA (mM)

49

0.35

 

0.30

0.25

0.20

0.15

0.10

0.05

 

 

 

000 I r 1 r
0 0.001 0.01 0.1 1 10

 

NAD concentration (mM)

Figure 2.3 Effect of NAD on the conversion of
cinnamleoA to benzoleoA

50

hydroxylation enzymes were not detected using the assay methods
described in Materials and Methods in cells grown on 3-phenylpropionic
acid, cinnamic acid, or benzoic acid.

Gentisate 1,2-dioxygenase. Gentisate 1,2-dioxygenase activity was
demonstrated in extracts from cells grown with 3-phenylpropionic acid,
cinnamic acid, or benzoic acid with a specific activity of 14:02, 1.5i0.3,
or l.2i0.3 umol maleylpyruvate produced/min-mg protein respectively.
This enzyme activity was not detectable in pyruvate-grown cells. Unlike
reported eubacterial gentisate 1,2-dioxygenases (7, 16, 19), no stimulating
effect of FeZ+ on enzyme activity was observed, although 2,2'-dipyridyl was
an inhibitor for the enzyme activity (Figure 2.4). The optimal salt
concentration for the activity was 2 M KC1 or NaCl. Detailed studies of the
purified gentisate 1,2-dioxygenase and its encoding gene will be presented
in chapter 3.

cis-trans Isomerase. A glutathione-dependent cis-trans isomerase with a
specific activity of 47.6i3.9 pmol/min ~mg protein was present in cells
grown on benzoic acid. When the ring fission reaction of gentisate

catalyzed by gentisate 1,2-dioxygenase ceased, the reaction mixture showed

Specific enzyme activity

51

 

(mein. mg protein)

 

 

 

 

2,2'- dipyridyl (mM)

Figure 2.4 Inhibition of gentisate 1,2-dioxygenase
activity by 2,2'-dipyridyl

52

a product peak at 334 nm corresponding to the absorption maximum of
maleylpyruvate. A rapid decrease of the 334 nm absorbance was observed
when 1 mM glutathione was added to the mixture. This was attributed to
the presence of a cis-trans isomerase which interconverts maleylpyruvate
and fumarylpyruvate upon the addition of glutathione (20).

Pathway induction. To study the induction of the 3-phenylpropionic
acid degradation pathway, 2,2'-dipyridyl and suspensions of Hf D1227
resting cells pregrown on 3-phenylpropionic acid, cinnamic acid, benzoic
acid, or pyruvate were incubated with 3-phenylpropionic acid or benzoic
acid, and the appearance of metabolites in the culture was examined. When
3-phenylpropionic acid and 2,2'-dipyridyl were incubated with suspension
of cells pregrown with either 3-phenylpropionic acid or cinnamic acid,
metabolites of 3-phenylpropionic acid degradation were detected in the
culture. However, when either a pyruvate or benzoic acid-grown cell
suspension was used, no metabolites of 3-pheny1propionic acid degradation
were detected (Table 2.3). When benzoic acid and 2,2'-dipyridyl were
added to suspensions of cells grown on pyruvate or benzoic acid, 3-

hydroxybenzoic acid and gentisic acid accumulated in benzoic acid-grown

53

Table 2.3 Accumulation of intermediates when 3-phenylpropionic acid and
2,2'-dipyridyl were added to cells pregrown on various substrates

 

 

 

 

 

 

 

Grth substrate Accumulation of intermediates in 3 hours ( 11M)
(3mM)
cinnamic benzoic 3-hydroxy gentisic
acid acid benzoic acid acid
3-phenylpropionic 1.6103 1.9:02 8.1i1.2 1 15.-1:14
acid
cinnamic acid 0.5i0.07 2.9104 52.5i3.3 317:26
benzoic acid 0 0 0 0
pyruvate 0 0 0 0

 

 

 

 

 

 

54

cells, but not in pyruvate-grown cells (data not shown). These data indicate
that the upper aliphatic pathway from 3-phenylpropionic acid to benzoic
acid can be induced by either 3-phenylpropionic acid or cinnarrric acid, but
not benzoate. The lower gentisate pathway from benzoic acid to gentisic

acid is induced by benzoate.

Discussion

Although the extracellular metabolites observed during growth and in
resting cell experiments suggest aerobic metabolism of 3-phenylpropionic
acid via cinnamic acid, benzoic acid, 3-hydroxybenzoic acid, and gentisic
acid, the enzyme data indicate its catabolism via CoA thioesters, which are
not released extracellularly due to membrane impermeability. The presence
of thioesterase activities converting 3-phenylpropionleoA, cinnamleoA,
benzoleoA, and 3-hydroxybenzoleoA to their corresponding free acids
was demonstrated in Hf D1227 cells grown on 3-phenylpropionic acid,
cinnamic acid, benzoic acid or pyruvate (data not shown). The
accumulation of benzoic acid in the cultures of Hf D1227 grown on 3-

phenylpropionic acid suggests that 2-carbon scission of the

55

3-phenylpropionic acid side chain resulting in benzoic acid precedes attack
on the aromatic nucleus. The enzyme results indicate that the
transformation of 3-phenylpropionic acid to benzoic acid occurs via a
mechanism similar to fatty acid B—oxidation. Although the anticipated but
unstable intermediates 3-hydroxy-3-phenylpropionleoA and 3-keto-3-
phenylpropionleoA (17) were not detected in the incubation of
cinnamleoA and NAD with cell-free extracts, NAD-dependent conversion
of cinnamleoA to benzoleoA was observed.

The accumulation of 3-hydroxybenzoic acid and gentisic acid in benzoic
acid-grown cells suggests that benzoic acid is converted to gentisic acid via
3-hydroxybenzoic acid. Although it is clear that gentisic acid serves as the
substrate for gentisate 1,2-dioxygenase, we have not yet detected either
benzoate or 3-hydroxybenzoate hydroxylase activity. Thus whether benzoic
acid is transformed into gentisic acid via benzoleoA and 3-hydroxy
benzoleoA as demonstrated by Altenschmidt et a]. (1) or via free acids
remains to be clarified. The proposed pathway for 3-phenylpropionic acid
degradation in Hf D1227 is shown in Figure 2.5.

The enzyme steps in the aliphatic pathway are separately regulated from

56

O
HrCHrCOOH Hz-CHrC-CO" A
C-———
———->
phenylpropionlc acid phenylpropionleoA
H=CH-COOH HFCHiCOA
<————
———>
clnnamlc acid clnnamleoA
l a
E EHOH-CHz-C-CoA]
3-hydroxy-3-phenylproplonleoA
E %Hr0—COA]
3-keto-3-phenyipropionleoA
Io]
(50H “'— 5M
-——>
benzoic acid benzoleoA
ii’
<—
0H
3-hydroxybenzoic acid 3-hydroxybenzoleoA
................ i
........... OOH
....... § 0
OH
gentisic acid

m8:

maleylpyruvate

iOOH
/
\ I
HOOC H
fumarylpyruvate

l

Figure 2.5 Proposed pathway for the degradation of 3-phenylpropionic acid by Hf. D1227

57

the aromatic gentisate pathway. Although the upper pathway steps strongly
resemble those of fatty acid B-oxidation, their induction by
phenylpropanoids (i.e. 3-phenylpropionic acid and cinnamic acid) and lack
of fatty acid utilization by Hf. D1227 suggest that they are distinct.

While the importance of CoA thioesters in the anaerobic degradation of
aromatic compounds has long been recognized (10), the involvement of
CoA derivatives in aerobic aromatic pathways has only recently been
described (1, 4, 8, 28). Our results indicate the participation of CoA
thioesters in the 2-carbon shortening of the side chain of 3-phenylpropionic
acid that had been postulated in eubacteria, but for which no enzyme
activities have been described (30).

In summary, Haloferax sp. D1227 metabolizes 3-phenylpropionic acid by
2-carbon scission of the aliphatic side chain via a B-oxidation mechanism to
produce benzoleoA, which is further degraded via a gentisate pathway.
Although lack of pathway details precludes in-depth comparisons, overall
similarities between the Hf D1227 pathway and that of eubacteria suggest
the possibility of general pathway existence prior to the separation of

Archaea and Bacteria.

58

Acknowledgments

We wish to thank Georg Fuchs for kindly providing 3-hydroxy
benzoleoA. We are extremely grateful to Mike Roberts and Jared
Leadbetter for help with molecular phylogeny. We also thank Olga
Maltseva, Tae-Kyou Cheong and David Emerson for many helpful

discussions. This work was supported by AURIG from Michigan State

University.

References

l. Altenschrrridt, U., B. Oswald, E. Steiner, H. Herrmann, and G. Fuchs.
1993. New aerobic benzoate oxidation pathway via benzoyl-coenzymeA
and 3-hydroxybenzoyl-coenzymeA in a denitrifying Pseudomonas. J.
Bacteriol. 175: 4851-4858.

2. Blakley, E. R. and F. J. Simpson. 1964. The microbial metabolism of
cinnamic acid. Can. J. Microbiol. 10: 175-185.

3. Bradford, M. M. 1976. A rapid and sensitive method for the
quantitation of microgram quantities of protein utilizing the principle of
protein-dye binding. Anal. Biochem. 72: 248-254.

4. Buder, R., and G. Fuchs. 1989. 2-Aminobenzoyl-CoA monooxygenase
/reductase, a novel type of flavoenzyme. Purification and some properties
of the enzyme. Eur. J. Biochem. 185: 637-643.

5. Cain, R. B. 1980. The uptake and catabolism of lignin-related aromatic
compounds and their regulation in microorganisms. 21-56. In: T. K. Kirk

59

et al. (ed.), Lignin biodegradation: microbiology, chemistry, and potential
applications. CRC press, Inc., Florida.

6. Coulson C. B., and W. C. Evans. 1959. Microbiological degradation of
trans-cinnamic acid by soil Pseudomonads. Chem. & Ind. 543-544.

7. Crawford, R. L., S. W. Hutton, and P. J. Chapman. 1975. Purification

and properties of gentisate 1,2-dioxygenase from Moraxella osloensis. J.
Bacteriol. 121: 794-799.

8. Crooks, G. P., and S. D. Copley. 1993. A surprising effect of leaving
group on the nucleophilic aromatic substitution reaction catalyzed by 4-
chlorobenzoyl-CoA dehalogenase. J. Am. Chem. Soc. 115: 6422-6423.

9. Dagley, S., P. J. Chapman, and D. T. Gibson. 1965. The metabolism of
B-phenylpropionic acid by an Achromobacter. Biochem. J. 97: 643-650.

10. Dangel, W., R. Barckrnan, A. Lack, M. Mohamed, J. Koch, B. Oswald,
B. Seyfiied, A. Tschech, and G. Fuchs. 1991. Differential expression of
enzyme activities initiating anoxic metabolism of various aromatic
compounds via benzoleoA. Arch. Microbiol. 155: 256-262.

11. Darnell, J. E., and W. F. Doolittle. 1986. Speculations on the early
cause of evolution. Proc. Natl. Acad. Sci. USA. 83: 1271-1275.

12. Emerson, D., S. Chauhan, P. Oriel, and J. A. Breznak. 1994. Haloferax
sp. D1227, a halophilic Archaeon capable of growth on aromatic
compounds. Arch. Microbiol. 16]: 445-452.

13. Gibson, D. T. and V. Subramanian. 1984. Microbial degradation of
aromatic hydrocarbons. 181-252. In: D. T. Gibson (ed.), Microbial
degradation of organic compounds. Marcel Dekker Inc., New York.

14. Groseclose, E. E., and D. W. Ribbons. 1973. 3-Hydroxybenzoate 6-
hydroxylase from Pseudomonas aeruginosa. Biochem. Biophys. Res.
Cornmun. 55: 897-903.

60

15. Hahlbrock, K., and D. Scheel. 1989. Physiology and molecular

biology of phenylpropanoid metabolism. Annu. Rev. Plant Physiol. Plant
Mol. Biol. 40: 347-369.

16. Harpel, M. R., and J. D. Lipscomb. 1990. Gentisate 1,2-dioxygenase
from Pseudomonas: purification, characterization and comparison of the

enzymes from Pseudomonas testosteroni and Pseudomonas acidovorans. J.
Biol. Chem. 265: 6301-6311.

17. Hilton, M. D., and W. J. Cain. 1990. Bioconversion of cinnamic acid
to acetophenone by a Pseudomonod: microbial production of a natural
flavor compound. Appl. Environ. Microbiol. 56: 623-627.

18. Juez, G. 1988. Taxonomy of extremely halophilic Archaebacteria. In:
F. Rodriguez-Valera (ed.), Halophilic bacteria. CRC press, Inc., Florida.

19. Kiemer, P., B. Tshisuaka, S. F etzer, and F. Lingens. 1996. Degradation
of benzoate via benzoyl-coenzyme A and gentisate by Bacillus

stearothermophilus PKl, and purification of gentisate 1,2-dioxygenase.
Biol. Fertil. Soils. 23: 307-313.

20. Lack, L. 1959. The enzymic oxidation of gentisic acid. Biochem.
Biophys. Acta. 34: 117-123.

21. Leadbetter, J. R., and J. A. Breznak. 1996. Physiological ecology of
Methanobrevibacter cuticularis sp. nov. and Methanobrevibacter curvatus

sp. nov., isolated from the hindgut of the termite Reticulitermes flavipes.
Appl. Environ. Microbiol. 62: 3620-3631.

22. Maniatis, T., E. F. Fritsch, and J. Sambrook. 1982. Molecular cloning:
a laboratory manual. Cold Spring Harbor Laboratory Press, Cold Spring
Harbor, New York.

23. Niemetz, R., U. Altenschmidt, S. Brucker, G. Fuchs. 1995. Benzoyl-
coenzyme-A 3-monooxygenase, a flavin-dependent hydroxylase.
Purification, some properties and its role in aerobic benzoate oxidation via
gentisate in a denitrifying bacterium. Eur. J. Biochem. 227: 161-168.

61

24. Olsen, G. J ., and C. R. Woese. 1997. Archaeal genomics: an overview.
Cell. 89: 991-994.

25. Rochkind, M. L., J. W. Blackburn, and G. S. Sayler. 1987. Metabolism
of non-chlorinated compounds. 45-68. In: M. L. Rochkind et al. (ed.),
Microbial decomposition of chlorinated aromatic compounds. U. S.
Environmental Protection Agency, Cincinnati, Ohio.

26. Suarez, M., E. Ferrer, A. Garrido-Pertierra, and M. Martin. 1995.
Purification and Characterization of the 3-hydroxybenzoate 6-hydroxylase
from Klebsiella pneumoniae. FEMS Microbiol. Lett. 126: 283-290.

27. van Berkel, W. J. H., W. J. J. van den T wee]. 1991. Purification and
characterization of 3-hydroxyphenylacetate 6-hydroxylase: a novel FAD-

dependent monooxygenase from a F lavobacterium species. Eur. J.
Biochem. 201: 585-592.

28. Vitovski, S. 1993. Phenylacetate-coenzyme A ligase is induced during
grth on phenylacetic acid in different bacteria of several genera. FEMS
Microbiol. Lett. 108: 1-6.

29. Wang, L-H., R. Y. Hamzah, Y. Yu, S-C. Tu. 1987. Pseudomonas
cepacia 3-hydroxybcnzoate 6-hydroxylase: induction, purification and
characterization. Biochemistry. 26: 1099-1104.

30. Webley, D. M., R. B. Duff, and V. C. Farmer. 1955. Beta-oxidation of
fatty acids by Nocardia opaca. J. Gen. Microbiol. 13: 361-369.

31. Webster, L. T., J. J. Mieyal, and U. A. Siddiqui. 1974. Benzoyl and
hydroxybenzoyl esters of coenzyme A. Ultraviolet characterization and
reaction mechanism. J. Biol. Chem. 249: 2641-2645.

32. Widdel, F., and F. Bak. 1992. Gram-negative mesophilic sulfate-
reducing bacteria. 3342-3378. In: A. Balows et al. (ed.), The prokaryotes.
Springer, New York.

33. Woese, C. R., O. Kandler, and M. L. Wheelis. 1990. Toward a natural

62

system of organisms: proposal for the domains Archaea, Bacteria, and
Eucarya. Proc. Natl. Acad. Sci. USA. 87: 4576-4579.

34. Yang, S. Y., and H. Schulz. 1983. The large subunit of the fatty acid
oxidation complex from Escherichia coli is a multifunctional polypeptide.
J. Biol. Chem. 258: 9780-9785.

Chapter 3

GENTISATE 1,2-DIOXYGENASE FROM HALOFERAX SP. D1227

Abstract

Gentisate 1,2-dioxygenase from extreme halophile Haloferax sp. D1227
was purified using a three-step procedure. The enzyme was found to be a
homotetramer of 42,000:i:l ,000 Da subunits with a native molecular weight
of 174,000-_|-6,000 Da. The optimal salt concentration, temperature and pH
for enzyme activity were 2 M KC1 or NaCl, 45°C and pH 7.2, respectively.
The gene encoding Hf. D1227 gentisate 1,2-dioxygenase was cloned,
sequenced and expressed in Hf Volcanii. The deduced amino acid
sequence exhibited a 9.2% excess acidic over basic amino acids typical of
halophilic enzyme. Four nova] histidine clusters and a possible extradiol

dioxygenases fingerprint region were identified.

63

Introduction

Under aerobic conditions, aromatic compounds are transformed by
monooxygenases and dioxygenases into dihydroxylated central
intermediates including gentisate, catechol, and protochatechuate, which are
subsequently cleaved by ring-fission dioxygenases (12). Intradio]
dioxygenases such as catechol 1,2-dioxygenase and protocatechuate 3,4-
dioxygenase cleave between the two hydroxyl groups (artho cleavage) and
typically contain nonheme F e31, whereas extradiol dioxygenases such as
catechol 2,3-dioxygenase and protocatechuate 4,5-dioxygenase cleave at a
bond proximal to one of the two hydroxyl groups (meta cleavage) and
contain Fe“. Phylogenetic analyses have indicated that while enzymes
within each class are evolutionarily related, intradiol and extradiol
dioxygenases have arisen from different ancestors (7, 13, 21). Gentisate
1,2-dioxygenase can't be classified as either intradiol or extradiol, since the
cleavage of gentisate occurs between the carboxyl substituent and the
proximal hydroxyl group. Gentisate 1,2-dioxygenases from Moraxella
osloensis, Pseudomonas testosteroni, Pseudomonas acidovorans, Bacillus

stearothermophilus PK], and Rhodococcus erythropolis S-l , have been

65

purified and characterized (2, 14, 16, 24). The enzymes fi'om P. testosteroni
and P. acidovorans have been demonstrated to contain active site F e2+ with
coordination sites for both the substrate and 02 suggesting that gentisate
1,2-dioxygenases in these organisms are mechanistically more similar to
extradiol dioxygenases (l4).

Haloferax sp. D1227, an extreme halophile isolated from soil
contaminated with highly saline oil brine near Grand Rapids, Michigan, is
the only reported aerobic archaeon utilizing aromatic compounds (i.e.,
benzoic acid, cinnamic acid, and 3—phenylpropionic acid) as sole carbon and
energy sources for growth (8). During our study of the 3-phenylpropionic
acid metabolism by Hf D1227 (10), gentisate 1,2-dioxygenase activity was
demonstrated in cell-free extracts prepared from cells grown on 3-
phenylpropionic acid, cinnamic acid or benzoic acid. Examination of
gentisate 1,2-dioxygenase from this extreme halophile was therefore
undertaken. In this communication, we report the purification and
characterization of gentisate 1,2-dioxygenase from Hf D1227, as well as the
cloning, sequencing and expression of the gene encoding Hf D1227

gentisate 1,2-dioxygenase.

66

Materials and Methods

Materials. All chemicals used in this study were reagent grade. Benzoic
acid and cetyltrimethylammonium bromide (CTAB) were purchased from
Aldrich Chemical Co. (Milwaukee, WI.). Catechol, protocatechuic acid,
gentisic acid, Fe(NH4)2(SO4)2, 2,2’-dipyridyl, ascorbic acid, SDS and PVDF
membrane were obtained from Sigma chemical Co. (St. Louis, MO.).
Phenyl-Sepharose CL-4B, Bio-Gel Hydroxyapatite and SuperoseTMIZ
prepacked column were from Pharrnacia Fine Chemicals (Uppsala,
Sweden). N,N,N’,N’,-tetramethylethylenediamine (TEMED), acrylamide,
bisacrylamide, ammonium persulfate, restriction enzymes and T4 DNA
ligase were purchased from Boehringer Mannheim Corporation
(Indianapolis, IN .). X-Gal, isopropylthio-B-D-galactoside (IPTG) and T aq
DNA polymerase were obtained from Life Technologies (Grand Island,
NY.). GENECLEAN II Kit was obtained from BIO 101 Inc.(La Jolla, CA.).

Microorganisms, plasmid vector, media, and growth conditions.
Haloferax sp. D1227 has been described (8). Haloferax volcanii WFDl 1
was obtained from Dr. W. F. Doolittle (Dalhousie University, Halifax,

Canada). pBluescript SK(+) was purchased from Strategene (La J olla,

67

CA.). Plasmid vector pMDS30 was a gift from Dr. M. L. Dyall-Smith
(University of Melbourne, Parkville, Australia).

For grth of Hf D1227 on benzoic acid, BS3 mineral salts medium (8)
of the following composition (in g/l) was used: (NH 4)ZSO 4, 0.33; KC], 6.0;
MgClz-6HZO, 12.1; MgSO4-7HZO, 14.8; KHZPO4, 0.34; CaClZonO, 0.36;
and NaCl, 100. 1 ml/l of a trace element solution (25) was also added prior
to sterilization, and the pH was adjusted to 6.9 with KOH. F ilter-sterilized
benzoic acid was added after sterilization. For rich medium (BSYT), BS3

mineral salts medium was supplemented with 3 g/l yeast extract and 3 g/l

tryptone. E. coli DH5 a was grown on LB medium (18). For solid media,
15 g/l Bacto-agar was added prior to autoclaving.

Cells were cultured aerobically in Erlenmeyer flasks containing 1/4
volume of medium. Flasks were incubated at 37°C on a rotary shaker at
200 rpm. Growth was monitored by measuring OD600nm on a Gilford DU
spectrophotometer.

Enzyme assay and protein determination. Unless stated otherwise,
Gentisate 1,2-dioxygenase was assayed at room temperature by measuring

the formation of maleylpyruvate at 334 nm for 30 seconds with a Perkin

68

Elmer double beam 124 spectrophotometer (17). The reaction mixture (2
ml) contained 0.25 mM gentisic acid, 50 u] enzyme, and 100 mM potassium
phosphate buffer (pH 7.0) containing 2 M KCl. The reference cuvette
contained the same reaction buffer except gentisic acid. One unit of enzyme
activity was defined as the amount of enzyme that catalyzed the formation
of 1 pmol maleylpyruvate/min. The molar absorption coefficient for
maleylpyruvate (e = 10,800) (2) was used to calculate enzyme activity.
Protein concentrations of the enzyme preparations were determined by
the method of Bradford (1) with bovine serum albumin dissolved in 100
mM potassium phosphate buffer (pH 7.0) containing 2 M KC1 as the
standard.
Preparation of cell-free extracts. Mid-log phase cells grown on 3 mM
benzoic acid were harvested by centrifirgation at 10,600xg for 30 min.
After washing twice with 1/5 culture volume of 100 mM potassium
phosphate buffer (pH 7.0) containing 2 M KC1, cells were resuspended in
10 ml of the same buffer. 2.5 ml aliquots of the suspension were sonicated
four times (30 seconds at 50 W followed by 2 min of cooling) in an ice bath

using a 4710 Series Ultrasonic Homogenizer (Cole-Parmer Instrument Co.,

69

Chicago, IL.). After removing unbroken cells and cell debris by
centrifugation at 47,800xg for 30 min (4°C), 5 mM ascorbic acid was added
to the supernatant to stabilize the enzyme.

Enzyme purification. 10 ml cell-free extract was loaded onto a phenyl-
Sepharose CL-4B column (6 cmx1.7 cm) equilibrated with 100 mM
potassium phosphate buffer (pH 7.0) containing 2 M KCl and 5 mM
ascorbic acid. The sample was eluted with the same buffer at a flow rate of
14 ml/hr. Fractions (0.55 ml) with high activity were pooled, diluted with
equal volume of 10 mM potassium phosphate buffer (pH 7.0) containing 2
M KCl and 5 mM ascorbic acid, and loaded onto a hydroxyapatite column
(3 cmx 1 .7 cm) equilibrated with the same buffer. The sample was eluted
with 60 mM potassium phosphate buffer (pH 7.0) containing 2 M KC] and 5
mM ascorbic acid at a flow rate of 7 ml/hr. Fractions (0.55 ml) with high
activity were pooled and concentrated to 200p] using a Centricon— 1 0
concentrator (MW cut-off : 10,000. Amicon, Inc., Beverly, MA.). The
concentrated sample was then loaded onto SuperoseTMIZ prepacked column
(30 cmxl cm) equilibrated with 100 mM potassium phosphate buffer (pH

7.0) containing 2 M KC] and 5 mM ascorbic acid and eluted with the same

70

buffer at a flow rate of 6 ml/hr. Fractions (0.55 ml) with the highest
activity were pooled and stored at 4°C. A11 chromatographic steps were
performed at 4°C except gel filtration which was carried out at room
temperature on a Pharmacia FPLC system.

Estimation of native molecular mass and subunit molecular weight.
The molecular weight of the native gentisate 1,2-dioxygenase was
determined by gel filtration chromatography as described in the purification
procedure. The column was calibrated with bovine thyroglobulin (M,
669,000); horse spleen apoferritin (M, 443,000); sweet potato B-amylase (M,
200,000); yeast alcohol dehydrogenase (M, 150,000); bovine serum albumin
(M, 66,000); and bovine erythrocytes carbonic anhydrase (M, 29,000) in the
MW-GF-1000 kit (Sigma).

The subunit molecular weight of gentisate 1,2-dioxygenase was obtained
by CTAB-PAGE performed as described by Eley et al. (6). A 4%
concentrating gel and a 12% separating gel were used. The samples were
desalted by diafiltration with 100 mM potassium phosphate buffer (pH 7.0)
using Centricon-lO prior to electrophoresis. Proteins were stained with

Coomassie brilliant blue R-250 (l 1). Molecular weight standard SDS-6H

71

from Sigma was used. CTAB-PAGE was done at 30 mA and room
temperature with cold water cooling of the buffer chamber.

N-terminal amino acid sequencing and peptide sequencing. The
purified enzyme was run on SDS-PAGE and electroblotted onto PVDF
membrane by the method of Matsudaira (19). The protein band stained with
Coomassie brilliant blue R-250 was excised and submitted for N-terminal
amino acid sequencing and peptide sequencing at MSU Macromolecular
Structure Facility. For peptide sequencing, enzyme on the membrane was
digested with trypsin at 37 °C for 20 hours and the peptides were separated
by HPLC using a C18 column (0.8 mm x 250 mm) with a gradient of 5%
acetonitrile to 50% acetonitrile. Both N-terminal arrrino acid sequencing
and peptide sequencing were performed on a 494 Procise protein sequencer
(Applied Biosystems, Weiterstadt, Germany) using automated Edman
degradation.

Preparation of Hf D1227 genomic DNA. Mid-log phase cells of Hf
D1227 grown in 100 m] BSYT medium were harvested by centrifugation at
10,600x g for 30 min. After washing twice with 100 mM potassium

phosphate buffer (pH 7.0) containing 2 M KC1, the pellet was resuspended

72

in 50 ml distilled water for cell lysis. After removal of cell debris by
centrifirgation at 47,800xg for 30 min, the supernatant was extracted once
with an equal volume of phenol, and twice with an equal volume of
phenol/chloroform. DNA in the aqueous phase was precipitated with 2
volume of ethanol and 0.] volume of 3 M sodium acetate (pH 5.2) at —16°C
overnight. After centrifirgation, the pellet was rinsed with 1 m1 of 70%

ethanol, air dried, and resuspended in 1 ml of H20. This preparation was

then treated with RNase at a final concentration of 25 ug/ml.

Cloning and sequencing of Hfi D1227 gentisate 1,2-dioxygenase gene.
A part of the open reading frame for the Hf. D1227 gentisate 1,2-
dioxygenase gene was first obtained by PCR amplification of the Hf D1227
genomic DNA using two degenerate primers derived from the purified Hf.
D1227 gentisate 1,2-dioxygenase. The entire nucleotide sequence of the
coding and flanking regions of the gene was then obtained by chromosomal
walking using PCR techniques. The overall strategy for cloning the Hf
D1227 gentisate 1,2-dioxygenase gene is depicted in Figure 3.1. For this, a
226 bp fragment (segment a) of Hf D1227 gentisate 1,2-dioxygenase gene

was first obtained by PCR amplification of the Hf D1227 genomic DNA

73

 

r
I l

 

—6
Sad P045

 

 

I‘
V

T3 SaCI I

I 7

segment b: 900 bp

J
I segment a: 226bp1

4..—
P067

 

‘_._

P044 SalI KpnI
—-——. ‘__.__...--

PO66 T,
L SalI
If 1' ‘I
segment c: 644 bp
9071 T,
Kpnl

 

t . 1

segment (1: 264 bp

Figure 3.] Strategy for cloning the coding and flanking regions of Hf
D1227gentisate 1,2-dioxygenase gene. The coding region is enclosed in the
box. Solid lines represent Hf D1227 sequences and dashed lines represent

pBluescript SK(+) sequences

74

using forward primer PO45 [5’-GCIGA(A/G)CA(A/G)GA(A/G)CCIAA
(A/G)GA-3 ’] and reverse primer P044 [5 ’-TAICCIGT(A/G)TTIACIGG
(A/T/C/G)AC-3’] corresponding to N-terminal sequences AEQEPKE and
YGTNVPV from the intact enzyme and a tryptic peptide of the enzyme
respectively. Segment a was sequenced using P044 and P045 as
sequencing primers.

A 900 bp fragment (segment b) upstream and overlapping segment a was
obtained by PCR amplification of the ligation mixture of SacI-digested
genomic DNA and pBluescript SK(+) with primers T3 from pBluescript
SK(+) [5’-AATTAACCCTCACTAAAGGG-3 ’] and P067 [5’-GCCGAA
TTGGTT TCCGAAGT-3’] which is 88 bp downstream of the 5’end of
segment a. Part of segment b was sequenced using PO67 as the sequencing
primer. Segment c, a 644 bp fragment downstream and overlapping
segment a, was obtained by PCR amplification of the ligation mixture of
Sail-digested genomic DNA and pBluescript SK(+) with T7 primer from
pBluescript SK(+) [5’-GTAATACGACTCACTATAGG GC-3’] and P066
[5’-TCTGGAAGTGGGAAGACATC-3’] which is 82 bp upstream of the

3’end of segment a. Segment c was sequenced using T 7 and P066 as

75

sequencing primers. A 264 bp fragment (segment (1) downstream and
overlapping segment c was obtained by PCR amplification of the ligation
mixture of KpnI-digested genomic DNA and pBluescript SK(+) with
primers T7 and P071 [5’-TGTTTCCGACGATGTCGTTC-3’] which is 54
bp upstream of the 3’end of segment c. The sequence of segment d was
determined using T7 and P071 as sequencing primers.

The PCR reaction mixtures in a total volume of 100 pl contained: 2 pl of
DNA template (ca. 100 ng), 20 mM Tris-HCl (pH 8.4), 50 mM KC1, 1.5

mM MgC12, 0.25 mM of each dNTP, 2.5U of T aq DNA polymerase, and 30

pmol of each primer. PCR reactions were carried out on Perkin Elmer
GeneAmp PCR System 2400 (Perkin Elmer, Norwalk, CT.) and thermal
cycling conditions were 94°C for 5 min followed by 35 cycles of 94°C for
l min, 55°C for 1 min, and 72°C for 3 min. A final extension was
performed at 72°C for 6 min. PCR products were purified with
GENECLEAN II Kit following the protocol recommended by the
manufacturer. Nucleotide sequencing was done at the Nucleic Acid
Sequencing Facility of Michigan State University with an ABI Prism

sequencer (Applied Biosystems).

76

Expression of H12 D1227 gentisate 1,2-dioxygenase gene in H]:
volcanii WFD 11. A 1353 bp fragment encoding gentisate 1,2-dioxygenase
gene and its flanking regions was amplified from the genomic DNA of Hf
D1227 using PO92 [5’-GCGGAAAGCTTTGG GAGTAC-3’] and P093
[5’-TAGGTACCTACCCGGCCTGG-3’]. PO92 corresponds to nucleotides
10 to 29 of the sequence shown in Figure 3.4 (see below) except that a G at
position 21 was changed to a T in order to introduce a Hind III restriction
site. PO93 corresponds to nucleotides 1346 to 1363 of the sequence with
two random nucleotides T and A added to the 5’end for effective digestion
at the KpnI site. PCR reaction and product purification were carried out as
described above.

After double digestion with Hind III and KpnI, this 1353 bp fragment was

ligated to plasmid pMDS30 which had been cut with the same two enzymes.

Transformation of the resulting plasmid into E. coli DH5a and the plasmid
isolation from E. coli transformants were performed using standard methods
(18). Transformation of the resulting plasmid into Haloferax volcanii WFD
11 was carried out using the PEG method described by Holmes et al. (15).

Transformants were selected on rich BSYT plates supplemented with 15%

77

sucrose and 0.3% novobiocin. Confirmed to contain the 1353 bp insert by
PCR amplification with P092 and P093, one colony was inoculated into 10
ml BS3 medium containing 5 mM pyruvate and 0.3% novobiocin, and the
gentisate 1,2-dioxygenase activity was measured as described in Enzyme

assay.

Results

Purification of gentisate 1,2-dioxygenase from Haloferax sp. D1227.
The soluble gentisate 1,2-dioxygenase from Hf D1227 was purified 207-
fold from cell-free extracts of benzoic acid-grown cells in three steps with
54% yield to give a preparation with a specific activity of 187 U/mg (Table
3.1). This preparation was electrophoretically homogeneous, showing one
band on an CTAB-PAGE gel (Figure 3.2). All purification steps were
performed in the presence of 2 M KC1, since Hf D1227 gentisate 1,2-
dioxygenase lost its activity irreversibly in low salt concentrations.
Ascorbic acid, which proved to be an stabilizer for the enzyme, was added

to all the purification buffers.

78

Table 3.] Purification of gentisate 1,2-dioxygenase from Hf D1227

 

 

 

 

 

 

 

 

 

 

 

Total Total Specific Yield Purification
Step protein activity activity (%) (fold)
(mg) (U) (U/mg)
Crude extract 114 104 0.9 100 1
Phenyl Sepharose 1 9 100 5.3 96
Cl-4B
Hydroxyapatite 1 .7 84 49.4 80 55
Superose 12 0.3 56 186.7 54 207

 

 

 

- «—-116kDa
- "'—97kDa
m'.
w ’-
=‘r "" - 4—66kDa
u. E
. -
- =' ° - “45“)!
- ‘ -
- - -
-
“3 ; - *29KD8
.-
- - 3
.4.

Figure 3.2 CTAB polyacrylamide electrophoresis of Hf. D1227 gentisate
1,2-dioxygenase at different stages of purification. 1. marker proteins; 2.

crude extract; 3. phenyl Sepharose Cl-4B; 4. hydroxyapatite; 5. superose
12.

80

Molecular weight and subunit structure. The molecular weight of the
native purified Hf D1227 gentisate 1,2-dioxygenase was estimated to be
174,000i6,000 by gel filtration. The subunit molecular weight of the
purified enzyme estimated by CTAB-PAGE was 42,000i1,000. These
results suggest that Hf D1227 gentisate 1,2-dioxygenase is a tetrarner
composed of four equally sized 42 kDa subunits. CTAB-PAGE instead of
SDS-PAGE was used in subunit molecular weight estimation in this study
because it has been demonstrated (20) that electmphoresis with this cationic
detergent is a more accurate method to determine the subunit molecular
weight of halophilic proteins, as halophilic proteins only bind a small
amount of SDS, resulting in a reduced mobility and an overestimation of the
molecular weight.

N-terminal amino acid sequencing and peptide sequencing. Amino
acid residues at the N-terminus of Hf D1227 gentisate 1,2-dioxygenase
were determined to be A-E-Q-E-P-K-E-L-L-E-M-S-T-D-T—E-R-L-L-E-E-N-
D-L-R-P-L-W-E-V-B—K—D-F-G-N-Q-F-G-G-. The N-terminal sequence of
one tryptic peptide of this enzyme was also obtained: V-A-V-P-V-N-T-G-

Y-R-. This peptide sequence along with N-terminal sequence were used in

81

gene cloning and sequencing (see below). Among the five purified
gentisate 1,2-dioxygenases, only N-terminal sequences of P. testosterom'
and P. acidovorans enzymes have been reported, and these have no
sequence similarity with that of Hf D1227 gentisate 1,2-dioxygenase.
Effect of salt, temperature and pH on the activity and stability of the
purified Hfl D1227 gentisate 1,2-dioxygenase. The salt dependence of Hf
D1227 gentisate 1,2-dioxygenase activity is shown in Figure 3.3, exhibiting
the highest activity in the presence of 2 M KCl. NaCl could substitute for
KC], yielding the same 2 M optimum. The optimal pH was pH 7.2 and the
optimal temperature for the activity was 45°C (data not shown). The
purified Hf. D1227 gentisate 1,2-dioxygenase lost activity rapidly and
irreversibly in salt concentrations below 2 M KC1, with 40% of activity
remaining when the enzyme was kept in 1.5 M KC1 for 30 min in ice and
assayed. This enzyme was also heat labile, incubation at 55°C for 10 min
caused 96% loss of activity. The enzyme was stable over a narrow pH
range (pH 6.5- 8.0). The enzyme in 100mM potassium phosphate buffer
(pH 7.0) containing 2 M KC1 was stable for up to three days at 4°C when 5

mM ascorbic acid or DTT was added.

82

190

 

180 _

170 ,

160

150 .

140 _

Specific activity (U/mg)

130 .

 

 

120 1 . - . 1
0.0 0.5 1.0 1.5 2.0 2.5 3.0 3.5 4.0

 

KC] concentration (M)

Figure 3.3 Effect of salt concentration on Hf. D1227 gentisate
1,2-dioxygenase activity

83

Catalytic properties and product identification. The me of the
gentisate 1,2-dioxygenase reaction was 187 i 24 U/mg and the Km value for
gentisic acid was 95 i 8 uM. Flushing the reaction mixture with nitrogen
gas caused 89% loss of enzyme activity, suggesting the involvement of O2
in the reaction. Catechol and protecatechuate were not substrates for this
enzyme. 2,2-dipyridyl was an inhibitor for the enzyme activity with an K,
value of 1.2 mM.

As a result of enzymatic gentisate oxidation, a compound with an
absorbance maximum at 334 nm was observed, which was lost on
acidification. These properties are characteristic of the expected
maleylpyruvate product (1 7).

Cloning and sequencing of Hf D1227 gentisate 1,2-dioxygenase gene.
The gene encoding Hf D1227 gentisate 1,2-dioxygenase was cloned and
sequenced as described in Materials and Methods. The determined
nucleotide sequence and the deduced amino acid sequence are shown in
Figure 3.4. The open reading frame of this gene contains 1074 bp which
encodes a protein of 358 amino acids with a molecular weight of 40,547.

The deduced protein sequence of Hf D1227 gentisate 1,2-dioxygenase had

84

1 CCGAACTGGGCGGAAAGCTCTGGGAGTACGCCATTCAGGGGCAGATGGTCTGCGAGGAC
61 GCCGCGAGCTGTACCCCTCGAAGGAGATATTCGACGCGCAGGTGAGCCAGTAGCGACCC

12 1 GGTGGCGGGTGACGGGTGACGGTGTCCGCCCGTAGCGGCGCCACGCGACGAGAATAGAC
l 8 l CATACA@GAAGAATCC'I‘CGGGAAC'I'I'I'ACACCATGGCTGAACAAGAGCCGAAGGA
M A E O E P K E

24 1 ACT CCT AGAGATGAGTACCGACACGGAGCGTCT CCT CGAGGAGAACGACCT GCGACCT CT
LLEMSLDTEBLJEENDLRPL

301 CT GGGAAGTCGAAAAGGACTT CGGAAACCAATTCGGCGGGTTCGAGGCGGACATCT GGAA
WEVEKDFGNOFGGFEADIWK

361 GTGGGAAGACATCCAGGCGTCGATCGACGCCATCGAGCGGGACGTCCCCATCGCGGACCT
WEDIQASIDAIERDVPIADL

421 CCCGCCGGGGTTCCAGCGACGCGTCGCCGTGCCCGTCAACACGGGATACCGAAACGCCAT
PPGFQRRVAVPVNIGYJNAI

48 1 CTCGAACACGATTTACGTCGGCGTCCAGACCGTCTCGCCCGGCGAGACCGCGCCGGCACA
SNTIYVGVQTVSPGETAPAH

541 CCGACACGGGGCGAACGCGC'ITCGGTTCACTATCGACGGGAGCGAGGACATGAAGACCGT
RHGANALRFTIDGSEDMKTV

601 CGTCGCGGGCGAGGAGTTCCCGATGCGGGACAACGACCTCATCACGACGCCGCAGTGGGA
VAGEEFPMRDNDLITTPQWE

661 GTGGCACGACCACGTCAACGACGGCGACGAGACGGCGGCGTGGCTCGACGTGCTCGACCT
WHDHVNDGDETAAWLDVLDL

72] CCCGCTCGTCCTCGACTCGCTCAACGCGCGCAACACCTTCGAGAACCACGAACTCGACCG
PLVLDSLNARNTFENHELDR

7 81 CCAGCCCGTGACGAAGTCGCAGGGCT ACT GGGAGTCGCAGTACGGACGCGCCCGCCCGTT
QPVTKSQGYWESQYGRARPF

841 GAGGACACGAAGGAAGACGGGATTCCCGGACCCTTCGAGGGGAACTGCGCCGCGACGCC
EDTKEDGIPGPFEGNCAATP

901 GCCGTACCGATTCAGCTGGAAAGACACCCTCCAGACCCTCCGACAGCGAGCCGAAAACGA
PYRFSWKDTLQTLRQRAEND

961 CGACCCGGACCCGCACGACGGCT ACAGTCT CT CGTACGTCAACCCCGCGACCGGCCAGCC
DPDPHDGYSLSYVNPATGQP

102 1 CCGCTGTI'TCCGACGATGTCGTTCCGCGCGCAGTTGCTTCAAGAAGAGACCGACCCGCA
PLFPTMSFRAQLLQEETDPg

85

108 1 CIT CCACAACGCCGTCGACGCGTAC'IT CGTCATCGAGGGCGAGGGCGCGACGCACGTCGG
FHNAVDAXFVIEGEGATHVG

1 l41CGACGACGTGCTCGAATGGAGCGAACGCGACATCT'I‘CGTGA'I'I‘CCGCCGGACGAGATTCA
DDVLEWSERDIFVIPPDEIH

1201 CCACCACGACCCCGACGGCGAAGCGATTCTCCTCGGGATGACCGACCGCCCGGTGTTCGA
HHDPDGEAILLGMTDRPVFE

l261GGCGTTCAACTI‘CI‘ACGCCGAGGCCGAACCGTAGGCGCGCGGGGTCCAG'I'I‘CCGGACCAC
A F N F Y A E A E P "

13 2 l TCTGTTCGCGCCTCGGCGGATACCGCCAGGCCGGGTAGGTACC

Figure 3.4 Nucleotide sequence of Hf D1227 gentisate 1,2-dioxygenase
gene and the deduced amino acid sequence. Amino acid sequences
determined by protein sequencing are underlined. A putative promoter
sequence is boxed. The histidine clusters are in bold letters. The putative
conserved residues are double-underlined

86

a 9.2% excess acidic over basic amino acids with an isoelectric point of
4.15. A putative promoter sequence TTAT similar to the archaeal box A
consensus TTA(T/A) usually located 271-4 bp upstream ATG start codon
(4), was observed 25 bp upstream from the ATG translation initiation
codon. F our histidine clusters H-X-H were found in the sequence, whose
function is unclear, although it has been shown that histidine residues are
involved in both iron ligand binding and catalytic function in eubacterial
extradiol dioxygenases (7). When compared with sequences of ring-
cleavage dioxygenases, a segment of Hf D1227 gentisate 1,2-dioxygenase
sequence encoding three residues (His, Tyr and Glu) conserved in the
eubacterial extradiol dioxygenase fingerprint region (7) was identified
(Figure 3.4).

Expression of Hfi D1227 gentisate 1,2-dioxygenase gene in H]:
volcanii WFD 11. The gentisate 1,2-dioxygenase activity of the Hf
volcanii transforrnant was 4.810.5U/ g protein, which was 0.54% of the
activity in Hf D1227. No activity was detected in Hf volcanii containing
only shuttle vector pMSD30 without insert, and no increase of enzyme

activity was detected when 0.5 mM benzoic acid was added as an inducer to

87

the growth medium.

Discussion

Gentisate 1,2-dioxygenase from Hf D1227 is similar in molecular weight
and subunit structure to those from P. testosteroni, P. acidovorans, and M.
osloensis (2, 14), all being tetramers composed of four equal-sized subunits
of molecular weight around 40 kDa. The gentisate 1,2-dioxygenase from
Bacillus stearothermophilus is a homohexamer with a subunit molecular
weight of 40 kDa and that of Rhodococcus is a homooctamer with a subunit
molecular weight of 43 kDa (16, 24). Gentisate 1,2-dioxygenase from Hf
D1227 demonstrates a similar K,,1 value for gentisate with those of the two
Pseudomonas enzymes and has a similar pH optimum (around pH 7.0) to
four of the purified eubacteria enzymes. The exception is the Rhodococcus
enzyme, exhibiting the maximal activity at pH 8.5.

Hf. D1227 gentisate 1,2-dioxygenase differs from the eubacterial
counterparts in the requirement of 2 M KC1 or NaCl for enzyme activity and
stability. By possessing additional acidic residues (glutamic acid and

aspartic acid), halophilic proteins are highly adapted to function in a high

88

salt milieu (3, 5, 9). Compared to the chemical neutrality of bulk protein for
non-halophiles (23), Hf D1227 gentisate 1,2-dioxygenase contains a 9.2%
excess acidic over basic amino acids, which is typical of halophilic proteins
(22, 23). The acidic amino acid content of the non-halophilic gentisate 1,2-
dioxygenases has not been determined.

Four of the purified eubacteria gentisate 1,2-dioxygenases required
exogenously added Fe2+ for enzyme activity. Although the gentisate 1,2-

dioxygenase from H]? D1227 did not require added F e2+ for activity,
Fe(NH4)2(SO4)2 could partially restore the lost activity of older Hf D1227
enzyme preparations and 2,2’-dipyridyl was an inhibitor for this enzyme
(10), suggesting the presence of Fe2+ in the enzyme.

Sequence comparison by BLAST showed no GenBank sequence similar
to the Hf D1227 gentisate 1,2-dioxygenase gene. This is not surprising
given this enzyme’s archaeal origin and the absence of eubacterial gentisate
1,2-dioxygenase sequences in the database. Although the expression of
gentisate 1,2-dioxygenase in Hf volcanii verified the function of the cloned
gene, the expression level was quite low. Several factors might contribute

to this low activity. First, the pMDS30 vector used does not have a

89

promoter adjacent to the Hf D1227 gentisate 1,2-dioxygenase gene insert,
and the control elements associated with the gentisate pathway were
apparently not included with the cloned gene since induction of expression
by benzoate was not observed. Secondly, the stability of the gentisate 1,2-
dioxygenase and its mRNA in Hf volcanii transformants is unknown.

In summary, gentisate 1,2-dioxygenase from extreme halophile
Haloferax sp. D1227 is similar to its eubacterial counterparts in terms of
subunit size and metal participation. The enzyme also possesses the
properties characteristic of halophilic enzymes including the requirement of
2 M KC1 or NaCl for activity and stability, and an excess of acidic over
basic amino acids. It will be of interest to determine whether the four
histidine clusters in the H]? D1227 gentisate 1,2-dioxygenase gene are
features shared by eubacterial gentisate 1,2-dioxygenases. Although a
segment of the Hf D1227 gentisate 1,2-dioxygenase contains the histidine,
tyrosine and glutamic acid residues conserved in the eubacterial extradiol
dioxygenase fingerprint region, further genetic and structural studies will be
required to confirm that these residues are involved in metal binding and

active site firnction of the dioxygenase.

90

Acknowledgments

We wish to thank Michael L. Dyall-Smith for plasmid vectors and
assistance. We are extremely grateful to Robert Hausinger for help with
purification. We also thank Olga Maltseva and Tae-Kyou Cheong for many
helpful discussions. This work was partially supported by the MSU

Biotechnology Research Center.

References

1. Bradford, M. M. 1976. A rapid and sensitive method for the
quantitation of microgram quantities of protein utilizing the principle of
protein-dye binding. Anal. Biochem. 72: 248-254.

2. Crawford, R. L., S. W. Hutton, and P. J. Chapman. 1975. Purification

and properties of gentisate 1,2—dioxygenase from Moraxella osloensis. J.
Bacteriol. 121: 794-799.

3. Dennis, P. P., and L. C. Shimmin. 1997. Evolutionary divergence and
salinity-mediated selection in halophilic archaea. Microbiol. Mol. Biol.
Rev. 61: 90-104.

4. Dennis, P. P. 1993. The molecular biology of halophilic archaebacteria.
274-279. In: R. H. Vreeland et al. (ed.), The biology of halophilic bacteria.
CRC Press, Inc., Florida.

5. Dym, 0., M. Mevarech, and J. L. Sussman. 1995. Structural features

that stabilize halophilic malate dehydrogenase from an archaebacterium.
Science. 267: 1344-1346.

91

6. Eley, M. H., P. C. Burns, C. C. Kannapell, and P. S. Campbell. 1979.
Cetyltrimethylammonium bromide polyacrylamide gel electrophoresis:

estimation of protein subunit molecular weights using cationic detergents.
Anal. Biochem. 92: 411-419.

7. Eltis, L. D., and J. T. Bolin. 1996. Evolutionary relationships among
extradiol dioxygenases. J. Bacteriol. 178: 5930-5937.

8. Emerson, D., S. Chauhan, P. Oriel, and J. A. Breznak. 1994.
Haloferax sp. D1227, a halophilic Archaeon capable of grth on aromatic
compounds. Arch. Microbiol. 161: 445-452.

9. Frolow, F ., M. Hare], J. L. Sussman, M. Mevarech, and M. Shoham.
1996. Insights into protein adaptation to a saturated salt environment from
the crystal structure of a halophilic 2Fe-ZS ferredoxin. Nat. Struct. Biol.

3: 452-458.

10. Fu, W., and P. Oriel. 1997. Degradation of 3-phenylpropionic acid by
Haloferax sp. D1227. Submitted.

1]. Garfin, D. E. 1990. One-dimensional gel electrophoresis. 182: 425-
441. In: Methods in enzymology. Academic Press, Inc., New York.

12. Gibson, D. T. and V. Subramanian. 1984. Microbial degradation of
aromatic hydrocarbons. 181-252. In: D. T. Gibson (ed.), Microbial
degradation of organic compounds. Marcel Dekker Inc., New York.

13. Harayama, S., M. Kok, and E. L. Neidle. 1992. Functional and
evolutionary relationships among diverse dioxygenases. Annu. Rev.
Microbiol. 46: 565-601.

14. Harpel, M. R., and J. D. Lipscomb. 1990. Gentisate 1,2-dioxygenase
from Pseudomonas: purification, characterization and comparison of the

enzymes from Pseudomonas testosteroni and Pseudomonas acidovorans. J.
Biol. Chem. 265: 6301-6311.

92

15. Holmes, M. L., and M. L. Dyall-Smith. 1990. A plasmid vector with a
selectable marker for halophilic archaebacteria. J. Bacteriol. 172: 756-761.

16. Kiemer, P., B. Tshisuaka, S. Fetzer, and F. Lingens. 1996.
Degradation of benzoate via benzoyl-coenzyme A and gentisate by Bacillus

stearothermophilus PKl , and purification of gentisate 1,2-dioxygenase.
Biol. Fertil. Soils. 23: 307-313.

17. Lack, L. 1959. The enzymic oxidation of gentisic acid. Biochem.
Biophys. Acta. 34: 117-123.

18. Maniatis, T., E. F. Fritsch, and J. Sambrook. 1982. Molecular cloning:

a laboratory manual. Cold Spring Harbor Laboratory Press, Cold Spring
Harbor, New York.

19. Matsudaira, P. 1987. Sequence from picomole quantities of proteins
electroblotted onto polyvinylidene difluoride membranes. J. Biol. Chem.
262 :10035-10038.

20. Monstadt, G., and A. W. Holldorf. 1990. Arginine deminase from
Halobacterium salinarium: purification and properties. Biochem. J. 273:
739-745.

21. Neidle, E. L., C. Hartnett, S. Bonitz, and L. N. Omston. 1988. DNA
sequence of the Acinetobacter calcoaceticus catechol 1,2-dioxygenase I
structural gene catA: evidence for evolutionary divergence of intradiol

dioxygenases by acquisition of DNA sequence repetitions. J. Bacteriol.
170: 4874-4880.

22. Prub, B., H. E. Meyer, and A. W. Holldorf. 1993. Characterization of
the glyceraldehyde 3-phosphate dehydrogenase fiom the extremely

halophilic archaebacterium Haloarcula vallismortis. Arch. Microbiol. 160:
5-11.

23. Reistad, R. 1970. On the composition and nature of the bulk protein of
the extremely halophilic bacteria. Arch. Microbiol. 71: 353-360.

93

24. Suemori, A., R. Kurane, and N. Tomizuka. 1993. Purification and
properties of gentisate 1,2-dioxygenase from Rhodococcus erythropolis S-l.
Biosci. Biotechnol. Biochem. 57: 1781-1783.

25. Widdel, F., and F. Bak. 1992. Gram-negative mesophilic sulfate-
reducing bacteria. 3342-3378. In: A. Balows et al. (ed.), The prokaryotes.
Springer, New York.

Chapter 4
SUMMARY AND RECOMMENDATIONS

FOR FUTURE RESEARCH

Summary of the dissertation research findings

The first aromatic degradation pathway in Archaea is elucidated in this
study. Haloferax sp. D1227 metabolizes 3-phenylpropionic acid by 2-carbon
scission of the aliphatic side chain via a B-oxidation mechanism to produce
benzoleoA, which is further degraded via a gentisate pathway. The upper
aliphatic pathway from 3-phenylpropionic acid to benzoic acid is separately
regulated from the lower aromatic gentisate pathway. Although it has been
proposed by Webley et al. (2) that Nocardia opoca metabolizes 3-
phenylpropionic acid to benzoate via a mechanism similar to fatty acid [3-
oxidation based on their observation of the transient accumulation of
cinnamic acid and benzoic acid in the 3-phenylpropionic acid-grown
cultures, our study is the first to demonstrate the enzymes involved in the

pathway.

94

95

The purified Hf D1227 gentisate 1,2-dioxygenase involved in the 3-
phenylpropionic acid degradation is similar to its eubacterial counterparts in
terms of subunit size and metal participation. The enzyme also has
properties characteristic of halophilic enzymes including the requirement of
2 M KC1 or NaCl for activity and stability, and an excess of acidic over
basic amino acids. The gene encoding Hf D1227 gentisate 1,2-dioxygenase
was cloned, sequenced and expressed in Hf volcanii. The gene contains a
sequence resembling the extradiol dioxygenases fingerprint region and four
histidine clusters (H-X-H) whose function and distribution among eubacteria

gentisate 1,2-dioxygenases remain unknown.

Future research

One area of firture research could be to increase the expression level of
Hf. D1227 gentisate 1,2-dioxygenase gene in Hf volcanni. pMDS30 used in
this study does not have a vector promoter adjacent to the Hf D1227
gentisate 1,2-dioxygenase gene insert, so the endogenous promoter of the
gene was used in the expression. It may be feasible to use those archaeal
expression vectors containing promoters next to the polylinker regions, such

as pWL107, pWL106, and pWL105 (1) to increase the expression.

96

Another area of research interest is the possible roles of the histidine
clusters and the conserved histidine, tyrosine and glutarrric acid residues in
metal binding and active site function of Hf D1227 gentisate 1,2-
dioxygenase, for which site-directed mutagenesis should be performed.
When the gene sequences of eubacteria gentisate 1,2-dioxygenases become
available, they should be compared with our Hf D1227 sequence.

The gentisate 1,2-dioxygenase activity was induced by benzoate in Hf.
D1227, however, no benzoate induction was demonstrated in Hf volcanii
transformants, indicating control elements were not complete in the cloned
gene. So, sequences adjacent to the cloned gene should be determined to
detect the possible open reading frames and regulatory elements. This may
also uncover the benzoate and 3-hydroxybenzoate hydroxylase activities
possibly involved in 3-phenylpropionic acid degradation by Hf D1227.

It will also be of interest to study the enzymes involved in the upper
aliphatic pathway of 3-phenylpropionic acid degradation by Hf D1227 and

their relationships with fatty acid B-oxidation enzymes.

97

Reference
1. Lam, W. 1989. Unpublished data.

2. Webley, D. M., R. B. Duff, and V. C. Farmer. 1955. Beta-oxidation of
fatty acids by Nocardia opaca. J. Gen. Microbial. 13: 361-369.

APPENDIX

APPENDIX

Degradation of Pinene by Bacillus pallidus BR 425

Abstract

An aerobic thermophile has been isolated from an or- and B-pinene
enrichment culture possessing the ability to grow on the monoterpenes or-
and B-pinene, pinocarveol, limonene, carveol, and carvone as sole carbon
sources. The isolate which was designated BR425 has been tentatively
identified as Bacillus pallidus using 16S ribosomal RNA gene sequencing
and organism morphology. Biphasic incubations of BR425 cells with or-
pinene, B-pinene, and limonene yielded carveol as a common metabolite,

and an unusual enzyme activity in BR425 crude extracts catalyzing
conversion of pinene to carveol utilizing FAD and potassium ferricyanide as

electron acceptors was observed.

Introduction
Pinene, the major constituents of turpentine, are bicyclic monoterpenes

which are produced in significant quantities by plants of the Pinaceae

98

99

family. Because of their volatility, pinene emission from conifer forests and
during pulping operations constitute a major source of biogenic
hydrocarbons (11, 17). The metabolism of pinene by microorganisms has
been little studied, as they have limited water solubility, and are membrane-
destructive to procaryotic and eucaryotic microorganisms (1). In early
reports in which a number of pinene metabolites were identified, catabolism
of or-pinene by Pseudomonas strain PL was suggested to proceed by
isomerization of the pinene to limonene with subsequent oxidation to perillic
acid prior to ring cleavage and further catabolism utilizing a B-oxidation
pathway (16). An alternative pathway through limonene proposed by
Gibbon and Pirt (7) has been questioned (18). A third pathway and some of
the participating enzymes for Nocardia strain P183 (8) and Pseudomonas
fluorescens strain NCIMB 11671 (2) have been described, in which or-
pinene is directly oxidized to pinene epoxide prior to ring cleavage. Recent
evidence indicates the presence of an alternative pinene pathway in the
pinene-epoxidizing Pseudomonasfluorescens strain NCIMB 1167], the
metabolites for which have not yet been identified (5). Few of the enzymes
participating in pinene catabolism or their encoding genes have so far been

characterized.

100

Because of their important environmental roles including composting and
waste treatment at elevated temperature, our laboratory has been engaged in
exploring the metabolic diversity of aerobic thermophilic bacteria. In this
report, we provide initial information on the degradation of or- and B-pinene
by a newly isolated Bacillus thermophile, which has been tentatively

identified as a strain of B. Pallidus.

Materials and Methods

Reagents and Media. The (R)-enantiomer of or-pinene was selected for
microbial degradation studies as this isomer predominates in North America
pines. All monoterpenes were purchased from Aldrich Company
(Milwaukee, Wisconsin), examined for purity using GC/MS (see below),
and filter sterilized prior to use. M9 minimal salts medium (13) contained
(per liter) NazPO4, 6g; KH2P04, 3g; NaCl, 0.5g; NH4C1, lg; pH 7.4. After
autoclaving and cooling, 2 ml of 1M MgSO4 were added. LB medium has
been described (13).

Organism isolation and growth. Pinene-degrading thermophiles were
isolated from samples of dried wound exudate from a stand of white pine

tree near Midland, Michigan. Enrichments were carried out in a 125 ml

101

bottle containing 50 m1 M9 minimal salts and 0.05 ml or-pinene, and were
incubated at 60 °C in a gyrorotatory water bath shake. After 72 hours
incubation, samples of the enrichment culture were diluted and grown at 60
°C on M9 salt plates in Petri dishes containing or-pinene in small glass tubes
attached to the cover. Isolates were repeatedly transferred on these plates
and retained as putative pinene utilizers. One isolate, designated BR425,
exhibited vigorous or-pinene-dependent growth, and was chosen for further
study.

For examination of growth rates and metabolite production, isolate BR425
was grown in 100 ml serum bottles containing 49 m1 M9 salts and pinene at
the desired concentration and sealed with Teflon-coated stoppers and
aluminum caps. Bottles were incubated in a gyrorotatory water bath shake.
Colony-forming units were enumerated by plating on LB agar at 60 °C.

Extraction and analysis of pinene biotransformation products. For
analysis of pinene biotransforrnation products, liquid cultures of BR425
grown for the desired time were centrifuged at 12,000x g for 20 minutes at
4 °C, acidified to pH 2.0 with HCl and extracted three times with equal
volumes of ether. The ether fraction was evaporated to 25 ml and separated

into neutral and acidic fractions by extraction (3 x 0.6 volumes) with 5%

102

NaOH. The ether fraction was concentrated with nitrogen, neutralized with
5% HCl, and analyzed by GC/MS. The NaOH fiaction was acidified to pH
2.0 and reextracted (3 x 0.3 volumes) with ether. Following concentration,
the ether fraction was analyzed using a Hewlett-Packard HP 5890 gas
chromatograph with a MSDHP 5970 detector and a fused silica capillary
column (0.25mmI.D. x 30M DB-wax). Conditions used were: helium
carrier gas, injection port and detector port at 240 °C, column temperature
from 40-240 °C at 7 °C/min with 2 min initial hold time.

Two-phase biotransformation studies. For two-phase
biotransformation studies by suspensions of pinene-grown BR425 cells, 50
ml of BR425 cells grown in LB medium were washed with M9 medium and
resuspended to 109 cells per ml in 0.05 M KZHPO4 buffer, pH 7.0 with 0.6
m1 of pinene or limonene added. Following shaking in a gyrorotatory water
bath at 50 °C, samples of the monoterpene phase were injected into the
GC/MS for directe analysis.

16S ribosomal RNA sequence analysis. Partial analysis of the BR425
16S ribosomal RNA gene sequence was obtained using chromosomal DNA
prepared as described by Maniatis et al. (8). PCR amplification and

sequencing of the 16S ribosomal gene was carried out as described by

103

Maltseva et al. (12) using the Michigan State University Automated
Sequencing Facility. Sequences were analyzed using the ribosomal database
accessed using the GCG program (Genetics Computer Group, Madison,
WI).

Assay of pinene oxide lyase. For assay of pinene oxide lyase, cells
grown in LB broth into late exponential phase were disrupted using brief
bursts of sonification at 4 °C followed by removal of cell debris by
centrifugation for 30 minutes at 12,000xg. Enzyme assay utilized the
spectrophotometric procedure of Griffiths et al. (8).

Pinene hydroxylase measurement in crude extracts. Pinene
hydroxylase activity in crude extracts was measured utilizing BR425 cells
grown in M9 salts containing 0.1% glycerol and 0.2% or-pinene at 50 °C to
late exponential phase. Cells were centrifuged and resuspended in 5 ml
0.05M Tris buffer, pH 8.0 to which 5 pl phenylmethylsulfonylfloride added
to inhibit proteases. Cells were sonicated using a Cole-Farmer 4710
ultrasonic homogenizer for 3 minutes in 30 second bursts with cooling, and
centrifuged at 12,000xg for 30 nrinutes to provide a crude enzyme extract.

Pinene hydroxylase was assayed in a mixture to which 200 pl crude enzyme

was added to 800 pl containing 50 mM a-pinene, 1 mM FAD, 1 mM

104

potassium ferricyanide in 50 mM Tris buffer, pH 8.0. Carveol production
was routinely measured with HPLC using 60% acetonitrile and 40% water
eluant and a 3.9 x150mm Nova-Pak C18 column (Waters). For enzyme
assay under anoxic conditions, the crude enzyme extract and enzyme
cocktail were separately sparged with nitrogen gas for 20 minutes prior to
rrrixing and assay. For GC/MS analysis of enzyme reaction products, the
filtered enzyme reaction mixture was extracted three times with ether and

analyzed following ether evaporation using nitrogen gas.

Results

Characterization of isolate BR425. Isolate BR425 was plated from an
or-pinene enrichment culture of white pine sap exudate and demonstrated
good growth on M9 plates with either or- or B-pinene vapor, producing small
creamy colonies in two days. Although the isolate exhibited a growth
optimum at 55 °C (data not shown), a slower grth temperature of 50 °C
was utilized for convenience. Microscopic examination indicated large rod-
shaped bacteria with terminal endospores. Analysis of the 16S ribosomal

gene sequence data approximately 350 nucleotides from each end yielded

Bacillus pallidus (emb/zZ6930/BP16SRNA) as the best sequence match,

105

with the only sequence differences being G for A at position 6 and T for C at

positon 573.

Growth of BR425 on monoterpenes. In previous studies with
thermophile isolates growing on the monoterpene limonene, we found a
limited concentraton range in which grouth was not inhibited (3). For isolate
BR425 growing on a-pinene, a similar limited range was observed (Figure
A1), with grth inhibition occurring above 0.2% concentration. The
BR425 isolate was also shown to grow on a wide range but not all
monoterpenes tested as shown in Table A1.

Metabolites produced during growth on or-pinene. Ether extracts of
BR425 culture supematants grown on or-pinene were examined at various
times of culture with the results shown in Table A2. Significant amounts of
the neutral metabolites pinocarveol, pinocarvone, carveol and carvone were
observed together with lesser amounts of myrtenol, myrtenal, limonene, and
B-pinene in two independent experiments. None of these metabolites were
observed in the absence of the thermophile. For most metabolites, the
highest concentrations were observed in the late exponential phase. In
addition to these neutral metabolites, an acid metabolite of molecular mass

168 was observed showing the fragmentation pattern shown in Figure A2.

106

 

 

 

Growth (CFU x105)
8
l

 

 

 

0 l l i l
0 24 48 72

Time (h)

Figure A] Growth of Bacillus pallidus BR425 in varied concentrations of
(R)-or-pinene. I , no pinene; A, 0.05% pinene; 0, 0.1% pinene, 1:1 , 0.2%
pinene; o , 0.3% pinene

107

Table A1 Growth of B. pallidus on monoterpenes

 

 

 

 

 

 

 

 

 

 

 

Monoterpenes Optima] growth concentration
(R)-or-pinene 0.2%

(R)-Limonene 0.2%

(S)—or-pinene 0. 1%

(S)-B-pinene 0.1%

(8)-carveol 0. 1%

(S)-pinocarveol 0.05%

(S)-carvone 0.01%

Myrtenol, myrtenal, or-terpineol N0 growth

pinene epoxide

 

 

Table A2 Metabolites formed during grth of B. pallidus BR425 on or-

108

 

 

 

 

 

 

 

 

 

pinene

Metabolites (mg/1) 0 hr 24 hr 48 hr 72 hr
B-pinene 0 0.01 0.03 0.02
Limonene 0 0.08 0.18 0.06
Pinocarveol 0 1 .02 0.82 0.52
Pinocarvone 0 0.87 0.94 0.82
Myrtenol 0 0.58 0.52 0.32
Myrtenal 0 0.46 0.47 0.44
Carveol 0 4.38 6.32 1.48

Carvone 0 1.28 2.69 0.81

 

 

 

 

 

 

 

m9

N\2

 

0:028 co 0888-8 co £38m mmvmm 80¢ 2:382: Bus 08 868mm :2558m8m ~< 23wE

owH ova ONH oofl om om ov

ow

ov

om

om

ooH

‘CDSC’UMCUQ’

m ma41012M > m

110

This acid metabolite was also observed during grth in carvone, and was
not identified in a library search. This fragmentation pattern is similar but
not identical with acid metabolites of the same mass number previously
reported during grth of Pseudomonas strains on pinene (8, 19).
Metabolites produced during two-phase incubations with a-pinene
and limonene. In previous work with a limonene-degrading E. coli
recombinant (l4), incubations with neat limonene provided a convenient
method for identification of neutral terpenoid metabolites, as the organic
phase insures saturation of substrate in the aqueous phase, while facilitating
removal and concentration of metabolites in the organic phase which can be
utilized for direct identification using GC/MS eliminating the need for
aqueous extraction. Experiments using incubations using BR425 cell
suspensions with a-pinene, B-pinene, and limonene are shown in Table A3.
For incubations with or-pinene and B-pinene, similar metabolites were
identified to those obtained by ether extraction during grth on or-pinene,
except that metabolites were found in higher concentrations, and carvone
was only observed in trace quantities. With limonene as substrate, carveol
and carvone were found in significant quantities, together with a smaller

amount of or-terpineol.

111

Table A3 Metabolites formed during two-phase incubation of BR425

suspended cells

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Metabolites Organic Phase

(mg/1) or-pinene B-pinene limonene
or-pinene ND ND
B-pinene 4.38 ND
Limonene 1 .3 1 l . 1

Pinocarveol 20.5 1 9.3 ND
Pinocarvone 7.2 6.8 ND
Myrtenol 23 .4 17.] ND
Myrtenal 19.5 12.5 ND
or-terpineol ND ND 1 .4
Carveol 14.3 9.6 10.2
Carvone trace trace 7.5

 

ND: not detected

 

112

or-pinene hydroxylase activity. In experiments with a cloned limonene
degradation pathway in our laboratory, we have discovered a limonene
hydroxylase with novel gene sequence whose action is unusual in utilizing
the electron acceptor FAD and whose catalysis is not inhibited under anoxic
conditions. A preliminary search for a similar hydroxylating activity for
pinene in BR425 extracts indicated a or-pinene hydroxylase activity with
added FAD and/or potassium ferricyanide shown in Table A4. Under
anoxic conditions produced with nitrogen sparging of enzyme and reaction
mixture prior to reaction, no inhibition of activity was observed. Although
other peaks including those for carvone were observed, GC/MS analysis of
ether extracts of the crude enzyme reaction products verified that carveol

was the principle reaction product (data not shown).

Discussion

On the basis of morphology, grth characteristics, and partial 16S
ribosomal RNA gene sequence, we have tentatively identified thermophilic
isolate BR425 as a new strain of Bacillus pallidus, with confirmation
awaiting more extensive examination. Bacillus pallidus was first described

by Scholz, et a] (15) as a dominant member of yeast factory sewage and in

113

Table A4 Stimulation of B. pallidus BR425 pinene hydroxylase activity
by electron acceptors

 

 

 

Additions Enzyme activity (U/ml)
0 0

0.1 mM FAD 0.04

1 mM FAD 0.47

1 mM FAD and 0.25 mM fenicyanide 0.60

1 mM FAD and 0.5 mM ferricyanide 0.94

1 mM FAD and 1 mM ferricyanide 2.1

1 mM FAD and 1 mM ferricyanide“ 2.3

 

 

 

* anoxic conditions by nitrogen flushing

114

municipal sludge, and a recent isolate has been found to degrade both
aromatic and aliphatic nitriles (6). In contrast to a previously isolated
thermophile B. stearothermophilus BR388 which could degrade limonene
but not pinene (4), isolate BR425 demonstrated a fairly broad monoterpene
substrate range, but with limited growth due to limited monoterpene
solubility and monoterpene toxicity.

In this and other studies of monoterpene degradation (16, 18), large
numbers of metabolites have made monoterpene pathway determinations
difficult. In BR425, the presence of both oxidized bicyclic and monocylic
intermediates suggests either the existence of separate pathways for these
intermediates, mechanisms for interconversion of these structural isomers, or
oxidation steps with broad specificity. While cloning of the pathway genes
and examination of participating enzymes will be required and are underway
to resolve this question, our working hypothesis is that carveol and carvone
are central intermediates in a common branched pathway shown in Figure
A3. This hypothesis is based on the ability of BR425 to grow on carvone
and carveol, production of these intermediates during growth on both
limonene and pinenes, and the common but as yet unidentified intermediate

found during grth on either pinene or carvone. Since growth was not

115

§ <83 55

a- pinene fl - pinene

Limonene YOH
H3 CH2 OH
HO
4—

Pinocorveol Myrtenol

Carveol

l i 7

CH, CH, CH=O
O o
+—

 

 

Pinoccrrvone Myrtenol

Carvone

1

Figure A3 Hypothetical pinene degradation pathway for Bacillus pallidus
BR425

116

observed on pinene oxide, no pinene oxide was observed in pinene
metabolite examinations, and no pinene oxide lyase activity was observed in
BR425 cell extracts (data not shown), we conclude that formation of carveol
from pinene oxide by BR425 is unlikely.

The pinene hydroxylase activity observed in initial experiments with
BR425 crude extracts differs from that expected for a monooxygenase, in
that electron acceptors rather than electron donors serve as cofactors, and
because enzyme inhibition was observed under anoxic conditions. Although
detailed studies including H2180 incorporation are required to characterize
this hydroxylase activity, these initial results suggest that the pinene
hydroxylase may resemble p-cresol hydroxylase (9) and nictotinarnide
hydroxylase (10), which hydroxylate by hydrolysis of an oxidized
intermediate rather than by introduction of molecular oxygen. It will be of
interest to determine the extent of similarity of this enzyme with the
limonene hydroxylase of B. stearothermophilus BR3 88, which catabolizes

limonene through a perillic acid pathway.

117

Acknowledgments

We gratefully acknowledge support for this research by the MSU Center
for Plant Lipids and Starches, State of Michigan Research Excellence Fund.
Mass spectral data were obtained at the MSU-NIH Mass Spectrometry

Facility supported in part by a grant (RR00480) from NIH National Center

for Research Resources.

References

l. Andews, R.E., L.W. Parks, and K.D. Spence. 1980. Some effects of

Douglas fir terpenes on certain microorganisms. Appl. Environ. Microbiol.
40: 301-304.

2. Best, D.J., N.C. Floyd, A. Magalhaes, A. Burfield, and RM. Rhodes.
1987. Initial steps in the degradation of alpha-pinene by Pseudomomas
fluorescens NCIMB 1167]. Biocatalysis 1: 147-159.

3. Chang, HQ, and P. Oriel. 1994. Bioproduction of perillyl alcohol and
trlated monoterpenes by isolates of Bacillus stearothermophilus. J. Food
Sci. 3: 660-662.

4. Chang, H.C., J. Gage, and P. Oriel. 1995. Cloning and expression of a
limonene degradation pathway from Bacillus stearothermophilus in
Escherichia coli. J. Food Sci. 60: 551-553.

5. Colcousi, A., K. M. Saquib, and DJ. Leak. 1996. Mutants of
Pseudomonasfluorescens NCIMB 11671 defective in the catabolism of or-
pinene. Appl. Microbiol. Biotechnol. 45: 822-830.

118

6. Cramp, R., M. Gilmour, and DA. Coman. 1977. Novel thermophilic

bacteria producing nitrile-degrading enzymes. Microbiology 143: 2313-
2320.

7. Gibbon, G.H., and SJ. Pirt. 1971. Degradation of or-pinene by
Pseudomonas PXl. FEBS Lett. 18: 103-105.

8. Griffiths, E.T., S.M. Bociek, P.C. Harries, R. Jeffcoat, D.J. Sissons, and
P. W. Trudgill. 1987. Bacterial metabolism of or-pinene: pathway from or-

pinene oxide to acyclic intermediates in Nocardia sp. Strain P18.3. J.
Bacteriol. 169: 4972-4979.

9. Hopper, D.J., and D.G. Taylor. 1977. The purification and properties of
p-cresol (acceptor) oxidoreductase (hydroxylating), a flavocytochrome from
Pseudomonas putida. Biochem. J. 167: 155-162.

10. Jones, M.V., and DE. Hughes. 1972. The oxidation of nicotinic acid
by Pseudomonas Chester. Biochem. J. 129: 755-761.

1 1. Lindskag, A., and A. Potter. Terpene emission and ozone stress.
Chemosphere. 30: 1171-1181.

12. Maltseva, 0., C. McGowan, R. Fulthorpe, and P. Oriel. 1006.
Degradation of 2,4-dichlorophenoxyacetic acid by an alkaliphilic haliphile.
Microbiology. 142: 1115-1122.

13. Maniatis, T., E...F Fritsch, and F. Sambrook. 1982. Molecular cloning.
Cold Spring harbor Laboratory, New York.

14. Savithiry, N., T.K. Cheong, and P. Oriel. 1997. Production of or-
terpineol from Escherichia coli cells expressing thermostable limonene
hydratase. Applied Biochemistry and Biotechnology (in press).

15. Schholz, T., W. Demharter, R. Hensel, and O. Kandler. 1987. Bacillus
pallidus ssp. nov., a new thermophilic species from sewage. System. Appl.
Microbiol. 9: 91-96.

 

119

16. Shukla, DP, and PK. Bhattacharyya. 1968. Microbial transformations
of terpenes: part XI-pathways of degradation of or- and B—pinenes in a soil
Pseudomonad (PL-strain). Ind. J. Biochem. 5: 92-99.

17. Stromvall, A.M., and G. Petersson. 1993. Photooxidant-forrrring
monoterpenes in air plumes from Kaft pulp industries. Environ. Pollution.
79: 219-2

18. Trudgill, P.W. 1990. Microbial metabolism of terpenes-recent
developments. Biodegradation. 1: 93-1005.

19. Tudroszen, N.J., D.P. Kelly, and NF. Millis. 1977. or-pinene
metabolism by Pseudomonas putida. Biochem. J. 168: 315-318.