I. f
.::
Y
, L I.
.7 :

 

. magi .

. :23“ .0. .

k» x a. 4. 3 Km
‘ , V .a

.v x .
3 a. 5..
ha ... w
Ilvl

m an.

in.

v:

 

.3. 1 t1:
:9... r . .

...

 

.5. nu: uh.
9%

"V“!

I
r.
9
1 . 5.. L ‘
f w. apkfa. : ‘ _ 4.
, ‘ ‘ . lurk .KH’H . . .l
, ‘ ‘ ion.” aivutrudnila .

 

l

A mmuafi

 

I

 

 

l! [\l

 

THESlS

NIVERSITY LIBRARIES

HHHH HHIHHH |H IHH HI

31293 01714

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

This is to certify that the

thesis entitled

The Initial Steps in 2,4-D Degradation in
Soil Bacteria

presented by

Timothy Martin Sassanella

has been accepted towards fulfillment
of the requirements for

M. S . Microbiology
degree in

 

 

//
”sf/pk
Major%fessor

 

Date 5/13/97

 

0-7639 MS U is an Affirmative Action/Equal Opportunity Institution

 

 

LIBRARY

Mlchlgan State

Unlvorslty

 

 

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

 

DATE DUE

MTE DUE

DATE DUE

 

'HPR—l—l—EBfi—fl
12 O 1.: *4

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

ma (:lCIWDpGS-p.“

THE INITIAL STEPS OF 2,4-DICHLOROPHENOXYACETIC ACID
(2,4-D) DEGRADATION IN SOIL BACTERIAL ISOLATES

By

Timothy Martin Sassanella

A THESIS

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

MASTER OF SCIENCE

Department of Microbiology

1 998

ABSTRACT

THE INITIAL STEPS OF 2,4-DICHLOROPHENOXYACETIC ACID
(2,4-D) DEGRADATION IN SOIL BACTERIAL ISOLATES

By

Timothy M. Sassanella

The initial step of the canonical 2,4-D degradation pathway is carried
out by 2,4-D/alpha-ketoglutarate dioxygenase (deA), a member of a
diverse superfamily of mechanistically related enzymes. Phylogenetic
analysis of these enzymes indicated that this superfamily is probably
polyphyletic. A continuous, quantitative assay using 4-nitrophenoxyacetic
acid (4-NPAA) to detect deA-like activities was developed for the
screening of intact bacterial cells and cell lysates. A survey using this assay
indicated substantial diversity among a diverse collection of 2,4-D
degrading soil bacteria. Nitrobacter winogradskyi M1 was examined
further, and was found to carry a plasmid borne atypical deA-like activity.
Differential response to the 4-NPAA assay of several engineered constructs

indicated that the permeability of 4-NPAA varied among soil bacteria, and

was confirmed using l4C-2,4-D uptake and incorporation assays.
Transposon mutagenesis of the 2,4-D degradation plasmid pJP4 indicated
that the plasmid encodes an unknown factor that influences the permeation

of 2,4-D in some strains.

Copyright by
Timothy Martin Sassanella
1998

To my parents and family, whose unflagging support made me the person I
am today, to Brenda Knotts, for being the true friend that she is, and to my
best friend and partner - Dean.

l

ACKNOWLEDGMENTS

A thesis is never generated in isolation, so I would like to extend my
thanks to a few people. Thanks to all of the graduate students that I have
spent hours speaking with about this and that, especially those of the
incoming class of 1991 - all of whom were more than just colleagues.
Thanks to my friends and associates that participated in the ROME lab,
where a great deal of intriguing and thoughtful work was done that inspired
my personal work. Thanks to Dr. Tiedje and the NSF Center for Microbial
Ecology for many years of financial and institutional support. I would like
to thank my committee members: Dr. Pat Oriel, Dr. Robert Hausinger, Dr.
Mike Thomashow, and Dr. Frans DeBruijn for guiding my work.

Finally, I would like to thank my mentor Dr. Micheal Bagdasarian,
who has many qualities that I may only aspire to have, for his time, patience,

and wisdom.

vi

TABLE OF CONTENTS

LISTOF TABLES ix
LIST OF FIGURES x
INTRODUCTION 1

References ................................................................. 1 3

CHAPTER 1
THE PHYLOGENY OF THE a-KETOGLUTARATE-
DEPENDENT DIOXYGENASE SUPERFAMILY... 21

Abstract .................................................................... 21

Introduction ............................................................... 22

Results and Discussion .................................................. 26

Summary .................................................................. 59

References ................................................................ 61
CHAPTER 2

USE OF 4-NITROPHENOXYACETIC ACID FOR THE DETECTION
AND QUANTIFICATION OF 2,4-Dlor-KETOGLUTARATE
DIOXYGENASE ACTIVITY IN 2,4-D DEGRADING

Abstract .................................................................... 76
Introduction ............................................................... 77
Results and Discussion .................................................. 79
Summary .................................................................. 89
References ................................................................. 90

vii

 

CHAPTER 3

2,4-D DEGRADING STRAIN CHARACTERIZATION AND THE
CONJUGAL CAPTURE OF A PLASMID FROM NITROBACTER
WINOGRADSKYI M1 THAT RESCUES 2,4-
DICHLOROPHENOXYACETIC ACID DEGRADATION

IN A TFDA’ ALCALIGENES EUTROPHUS STRAIN 93

Introduction ............................................................... 93
Results and Discussion ................................................... 95
General Comments ................................................ 95
Phylogenetic Identification of the or-
proteobacterial Strains ..................................... 95
Strain Characterization ........................................... 102
Genetic Testing of the or-proteobacterial strains ............ 106
Conjugal Capture of M1 Plasmids .............................. 108
Growth and Genetic Characteristics of M1
and P strains ............................................... 108
Attempts to Isolate and Subclone the deA-like Activity... 109
Summary .................................................................. 112
References ................................................................ 1 13
CHAPTER 4
VARIABLE 2,4-D PERMEABILITY IN BACTERIA... 118
Abstract .................................................................... 118
Introduction ............................................................... 1 19
Results and Discussion .................................................. 123
Identification of Putative Permeation Effects ................. 123
The Mutagenesis of pJP4 and the Subsequent
Screening for Permease Mutants ....................... 127
Hybridization Experiments ...................................... 127
2,4-D Incorporation Assay ...................................... 128
Partial Subcloning of the Transposon Insertion Site ........ 128 -
Evolutionary and Ecological Implications .................... 131
References ................................................................. 134

CONCLUSION 137

APPENDIX A
Detailed Procedures for Chapter 1 ..................................... 139

viii

APPENDIX B

Detailed Procedures for Chapter 2 ..................................... 142
APPENDIX C

Detailed Procedures for Chapter 3 ..................................... 145
APPENDIX D

Detailed Procedures for Chapter 4 ..................................... 150

ix

LIST OF TABLES

Table 1.1 - Abbreviated names, accession numbers, enzyme names,
and references for strains used in this study ........ ‘ ....... 27

Table 1.2 - Putative iron binding regions containing strictly
conserved histidine residues from the groups of the or-
ketoglutarate-dependent dioxygenase superfamily .......... 35

Table 2.1 - Control and engineered strains and their response to
the 4-NPAA assays ............................................ 80

Table 2.2 - 4-NPAA conversion to 4-NP by a diverse set of
natural isolates that degrade 2,4-D ............................ 81

Table 3.1 - Strains and plasmids used to investigate novel
deA—like activities ............................................. 96

Table 3.2 - Carbon source utilization of four B-proteobacterial
2,4-D degrading strains and several strains used
for genetic manipulation ....................................... 98

Table 3.3 - Hybridization and enzyme activity characteristics
‘ of four B-proteobacterial 2,4-D degrading strains ...... 103

Table 4.1 - Response of various engineered host/plasmid
combinations using 4-NPAA as an alternate
substrate to assay for deA activity ........................... 123

LIST OF FIGURES

Figure A - The canonical pathway for 2,4-D degradation as

occurs in Alcaligenes eutrophus JMP134 ..................... 4

Figure B - Genetic structure of the canonical 2,4-D
degradation pathway of plasmid pJP4 isolated

from Alcaligenes eutrophus JMP134 ......................

Figure C - Reaction mediated by 2,4-D/or-ketoglutarate

-dependent dioxygenase (deA) ............................

Figure D - Two of the known reactions catalyzed by 2,4-D/

or-ketoglutarate-dependent dioxygenase (deA) .........

Figure 1.1 - Multiple sequence alignment of the or-keto
acid-dependent cluster of the or-ketoglutarate-

dependent dioxygenase superfamily ......................

Figure 1.2 - Neighbor-joining tree of the multiple sequence
aligned a-keto acid cluster of the a-ketoglutarate-

dependent dioxygenase superfamily .......................

Figure 1.3 - Multiple sequence alignment of the
4-hydroxyphenylpyruvate dioxygenase cluster
of the or-ketoglutarate-dependent dioxygenase

superfamily ....................................................

Figure 1.4 - Neighbor-joining tree of the multiple sequence
aligned 4-hydroxyphenylpyruvate (4-hppd)
dioxygenase cluster of the or-ketoglutarate-

dependent dioxygenase superfamily .......................

xi

..... 5

9

10

..... 30

37

41

..... 45

Figure 1.5 - Multiple sequence alignment of the gamma-
butryobetaine hydroxylase cluster of the

or-ketoglutarate-dependent dioxygenase
superfamily ....................................................... 46

Figure 1.6 - Multiple sequence alignment of the 2,4-D/
or-ketoglutarate-dependent dioxygenase
cluster of the or-ketoglutarate-dependent
dioxygenase superfamily ........................................ 50

Figure 1.7 - Neighbor-joining tree of the multiple
sequence aligned 2,4-dichlorophenoxyacetic
acid/or-ketoglutarate-dependent (deA)
dioxygenase cluster of the or-ketoglutarate-
dependent dioxygenase superfamily ............................ 53

Figure 1.8 - Multiple sequence alignment of the 2,4-D/
or-ketoglutarate-dependent dioxygenase
cluster and the gamma-butyrobetaine
hydroxylase cluster of the or-ketoglutarate-
dependent dioxygenase superfamily ............................ 54

Figure 2.1 - The 4-NPAA petri plate assay ................................... 84

Figure 2.2 - 4-NPAA cell lysate assay using Alcalz'genes
eutrophus JMP134 and Burkholderia cepacia RASC ....... 85

Figure 2.3 — 4-NPAA cell lysate assay using Alcaligenes
eutrophus JMP134 indicating the 4-NPAA
assay is quantitative ............................................... 86

Figure 3.1 - Phylogenetic tree representing the neighbor-

joining analysis of the 16S rDNA sequence
of strain M1 and close relatives ................................. 100

xii

Figure 3.2 - Effect of known deA co-factors on the conversion
of 4-NPAA to 4-NP by the crude lysates of 2,4-D
degrading bacteria ................................................. 105

Figure 3.3 - Compartmentalization of deA-like activity
in several 2,4-D degrading strains. ............................ 107

Figure 3.4 - Schematic diagram of the CHEF gel analysis of
the intact genetic elements found in strains used
in the conjugal capture of a deA-like gene
from strain Ml ................................................... 110

Figure 4.1 - Variability in the permeation or utilization of
2,4-D of environmental strains and putative
permease mutants is indicated by the 4-NPAA
whole cell assay .................................................. 125

Figure 4.2 - Determination of 2,4-D permeation in various

environmental and engineered strains using a

2,4-D uptake assay ............................................... 126
Figure 4.3 - 2,4-D incorporation assay ....................................... 129

Figure 4.4 - Physical map of plasmid me2 ................................. 130

xiii

INTRODUCTION

Microorganisms play a significant part in the removal and

detoxification of xenobiotic compounds in nature (Alexander,l981;

Chaudhry,199l; Ghosal, 1985; Haggblom, 1992; van der Meer, 1992).

Degradation of these compounds, which include halogenated aromatics, is
often slow due to limited degradation by unfavorable physiochemical
conditions (Evans, 1959; World Health Organization, 1984 and 1989) and
the inability of most microorganisms to metabolize them due to unusual
chemical structure. Microbial communities have been shown to adapt,
developing the ability to utilize many halogenated aromatic compounds
upon exposure to xenobiotic compounds (Pemberton, 1981). This
presumably occurs by induction of appropriate members of the community,
shifts in the population that increase the number of degrading bacteria
present in the community (Spain,1983), or by mutation, gene transfer, or
other mechanisms of bacterial evolution that result in a state where
degradation is possible. In this last case, a novel catabolic function is
created from a specific ‘parental’ enzyme or enzymes. Genes expressing
these enzymes would be expected to continue to evolve rapidly under
selection of the xenobiotic, particularly if the compound can serve as a

source of carbon and energy.

The herbicide 2,4-d-ichlorophenoxyacetic acid (2,4-D), a synthetic
auxin, is one of the least chemically complex members of the chlorinated
aromatic family that includes PCP, PCBs, and dioxins. 2,4-D has been in
wide commercial and agricultural use for approximately forty years, with 40
million pounds being applied to 54.8 million acres of US. agricultural land
during 1971 alone (Bovey, 1980). Unexpectedly, this compound, though
halogenated and moderately toxic, has not been found to accumulate in soil or
water (Perkins, 1988). Individual environmental isolates as well as microbial
communities from around the world have been found to mineralize 2,4-D
readily (Friedrich, 1983; Fulthorpe 1995 and 1996; Pemberton, I981; Sinton,
1986; Spain, 1983). The diverse geographic and temporal application of 2,4-
D to soils, combined with the widespread degradability of this xenobiotic,
should provide a maximal diversity of microorganisms and catabolic systems
in which the evolution of novel catabolic genes and pathways can be studied.

The canonical pathway for the complete mineralization of 2,4-D has
been well characterized (Don, 1985; F ukumori, 1993a and 1993b; Harker,
1989; Kaphammer, 1990a and 1990b; Kasberg, 1995; Perkins, 1988 and
1990; Pieper, 1988; Streiber, 1987; You, 1995). The genes are encoded on the
90 kb, IncP, broad-host—range plasmid pJP4 originally isolated from

Alcaligenes eutrophus (Don, 1985). There are six known structural and two

identical regulatory genes responsible for the conversion of 2,4-D to b-
ketoadipate (Don, 1985; Kukor, 1989; Kasberg, 1995). The initial three steps
of the pathway are performed by three oxygenases: deA, an a-ketoglutarate-
dependent dioxygenase that converts 2,4-D to 2,4-DCP (Fukumori, 1993a and
1993b); deB, a dichlorophenol monooxygenase that converts 2,4-DCP to
3,5-dichlorocateChol (Perkins, 1988); and deC, an intradiol chlorocatechol
dioxygenase that begins a typical modified ortho-cleavage pathway through
2-chloromaleylacetate to TCA cycle compounds (Perkins, 1990; Figure A).
The pathway genes are clustered in a 20 kb region of pJP4. The fimctional
and regulatory genes are found in three clusters that are separated by regions
that contain non-functional partial duplications of pathway genes (Figure B).
The overall genetic structure of the pathway, including the abundance of
partial fragments and the presence of insertion sequences, suggests that this
pathway was assembled from diverse genetic elements. The genes tfdA and
that encoding 2,4-dichlorophenol hydroxylase (deB) seem to encode the
most atypical enzymes of the group, and appear to have been recruited to
expand the chlorocatechol degradation pathway (Harker, 1989; Kaphammer,
1990a and 1990b; Perkins, 1990). An alternate 2,4-D pathway has also been
found that retains a deA-like side chain cleavage function as in the canonical

pathway, but where the 2,4-dichlorophenol is degraded in a similar manner as

 

 

2,4-dichlorophenoxyacetate 2,4-dichlorophenol 8,5-dichlorocatechol
/—C00.
0 OH - OH
CI CI . HO CI
0 ——>— © »- ©
tidA tidB
CI —" Cl — . Cl
Succinate tde
l o I
COOH l COOH
| —<—— / ~4— \ |
I 39E: / 9:99
o HOOC 0'

2-chloromaleylacetate cis-2-chlorodienelactone 2,4-dichloro-cis,cis-muconate

Figure A. The canonical pathway for 2,4-D degradation as occurs in
Alcaligenes eutrophus JMP134. . .

tfdA :de :de tdeII tfdA/I 131104 tde tdeDEF :de

@le >C>

 

 

 

Figure B. Genetic structure of the canonical 2,4-D degradation
pathway of plasmid pJP4 isolated from Alcaligenes eutrophus JMP134.
The known 2,4-D degradation genes are located in an approximately 20 kb
region of the 90 kb plasmid. T fdAII and T de11 are highly similar non-
functioning truncated copies of the primary pathway gene. deT is a non-
functional partial copy of a LysR regulatory gene that is closely related to
the identical pathway regulators tde and tde. Compiled from Don, 1985,
Harker, 1989, and Levaeu, 1996.

pentachlorophenol (O. Maltseva, personal communication). An analogous
compound, 2,4,5-trichlorophenoxyacetic acid (2,4,5-T), is also degraded by
the pentachlorophenol pathway. The chlorophenol is produced by a NAPH-
dependent monooxygenase (TfiA) that performs an analogous reaction to that
of deA (Haughland, 1991, Xun,l995) and will cleave both 2,4-D and 2,4,5-
T, though at a markedly reduced rate with the dichlorinated substrate
(Fukumori, 1993b; Haughland, 1991). Though they share a similar
biochemical function, deA and TftA are evolutionarily distinct.

Various surveys of 2,4-D degrading isolates indicate that there are
many plasmids that closely resemble the pathway encoded on pJP4 among
Gram-negative soil bacteria that can use 2,4-D as a sole carbon and energy
source (Tonso, 1996; Top, 1996). Hybridization studies indicate that 2,4-D
degradation pathways utilizing different but closely related (>60% identity)
genes have been assembled independently in different organisms (F ulthorpe,
1995). This mosaic pattern for the assembly of catabolic pathways is
consistent with the sequence information for pJP4. These experiments also
indicate that there is significant diversity among these bacteria, with a number
of strains failing to hybridize to probes for zfdA, tde, or tde. Many of the

strains tested, including Alcaligenes eutrophus JMP134, were [3-

proteobacterial isolates, and those strains that are phylogenetically more

distant rarely hybridized with the canonical genes (McGowan, 1995). Further,
plasmid capture experiments using a tfdA' deletion mutant of pJP4 indicated
that there is a wide variety of plasmids in soil that can complement the lack of
deA activity in the A. eutrophus construct (Top, 1995). A number of these
plasmids carry only a deA-like activity (Top, 1996), suggesting that there
may be significant genetic diversity found among incomplete 2,4-D pathways.
The utilization of a-ketoglutarate places deA into the highly diverse
superfamily of a-ketoglutarate-dependent dioxygenases, a group of
oxygenases that are thought to share a common mechanism of catalysis
(Prescott, 1993). A number of sequences of enzymes with this requirement
from both eucaryotic and procaryotic organisms are known, and the
phylogenetic relationships within these sequences and among other enzymes
of similar function are unclear. Members of this group are as diverse as
eucaryotic prolyl-4-hydroxylase (Helaakoski, 1989), bacterial gamma-
butyrobetaine hydroxylase (Englard, 1985), and fungal isopenicillin N
synthase (Aharonowitz, 1992). deA, several slightly divergent variants, and
a putative or-ketoglutarate—dependent dioxygenase tauD (Leisinger, 1997)
show conservation of a number of residues, particularly histidine residues
which have been implicated as essential for activity (Fukumori,'1993b).

Though clearly related to each other, these enzymes exhibit distant

similarity to another cluster of enzymes that are gamma-butyrobetaine
hydroxylases, but not with other members of the superfamily (Prescott,
1993; this work, Chapter 1).

deA converts 2,4-D, 02, and a-ketoglutarate to 2,4-dichlorophenol,
glyoxylate, carbon dioxide, and succinate (Fukumori, 1993b; Figure C).
This enzyme is a non-heme iron dependent dioxygenase has utilizes reducing
agent, usually ascorbate, that is thought to help maintain the metallocenter in
its active state. This enzyme has the highest affinities for 2,4-D and or-
ketoglutarate among a series of phenoxyacids and or-ketoacids examined
(Fukumori, 1993a and 1993b), but may utilize other similar compounds to a
lesser degree (Figure D). This property was exploited in the development of a
chromogenic assay for deA-like activity that has significant advantages to
previous methods, and allows rapid, large scale screening of bacteria (Chapter
2).

Two virtually identical zfdA—like genes have recently been cloned from
Burkholderia species, both of which have greater than 75% homology to {fdA
and 91% identity at the amino acid level (Suwa, 1996; Matheson, 1996). The
product of tfdAmsc has been partially characterized biochemically, and it
functions in a similar fashion to the benchmark enzyme. The three-

dimensional structure of these enzymes is unknown. Recent genome sequence

/—-COO .
o (300' 0H .
O . 00
Cl TFDA CI coo C
O, . . ___. . . . g .. co,
. Fe'2 CHO ,
COO ‘ COO
Cl Cl
2,4-D a-keto 2,4-DCP glyoxylate succinate
glutarate .

Figure C. Reaction mediated by 2,4-D/0t-ketoglutarate dioxygenase
(deA). This diagram is from F ukumori, 1993b.

10

cnzcoon

/
O OH
C1 C1
r O
Cl C1
2 , 4—D I 2 , 4—DCP
CH2COOH
/
O OH
——+
N02 N02
4—NPA - p—nitrophenol-

Figure D. Two of the known reactions catalyzed by 2,4-D/0t-
ketoglutarate dioxygenase (deA). 2,4-dichlorophenoxyacetic acid (2,4-D)
is converted to 2,4-dichlorophenol (2,4-DCP) preferentially by deA. The
conversion of 4-nitrophenoxyacetic acid produces 4—nitrophenol, an
intensely yellow compound. This chromogenic compound can not be
utilized by the rest of the 2,4-D pathway and accumulates, a property which
was exploited for the development of an assay for deA (Chapter 2).

 

11

projects have produced several complete sequences that are similar enough to
tfdA to infer relatedness. Three sequences - scox] from Saccharomyces
cereviseae, and mtox] and mtox2 from Mycobacterium tuberculosis - share
more than 27% identity with tfdA, but have not been biochemically
characterized. The gene product of tauD from E. coli is taurine dioxygenase,
and shares 31% identity with tfdA. It is part of a sulfur scavenging pathway,
and has yet to be well characterized (Leisinger, 1997). Since the
phylogenetically related tauD gene product can catalyze alternate
hydroxylation reactions, yet the unrelated tmo gene product will hydroxylate
2,4-D at C-2 in a deA-like reaction, then is not unfeasible that there are
{fdA-like genes that are less homologous than have been examined, and there
may be other phylogenetically distinct deA-like enzymes in soil bacteria.
More than 50 Gram-negative bacterial 2,4-D degrading strains were examined
for alternate activities, and the identification of two strains expressing
biochemically distinct deA-like enzymes is covered in Chapter 3.

Though the characterization of the canonical 2,4-D pathway has been
extensive, another degradation factor has been overlooked. Since bacterial
cytoplasmic membranes are relatively impermeable to most solutes,
permeases are required for the transport of most substances used for catabolic

metabolism (Nikaido, 1985). This transport step would be a crucial initial step

12

in the biodegradation of xenobiotic compounds. The transport of 2,4-D into
bacterial cells was found vary significantly among bacterial strains, and the
expression of a putative permease encoded on the plasmid pJP4 is critical
factor in the degradation of 2,4-D in some environmental strains. The

transport of 2,4-D into some bacterial cells is examined in Chapter 4.

 

13

REFERENCES

Aharonowitz, Y., Cohen, G., and Martin, J.F. 1992. Penicillin and
cephalosporin biosynthetic genes: structure, organization, regulation,
and evolution. Ann. Rev. Microbiol. 46, 461-95.

Alexander, M. 1981. Biodegradation of chemicals of environmental concern.
Science 211, 132-8.

Bovey, R.W. and A. L. Young. 1980. The Science of 2,4,5-T and Associated

 

Phenoxy Herbicides. John Wiley & Sons, New York.

 

Chaudhry, G. R., and S. Chapalamadugu. 1991. Biodegradation of
halogenated organic compounds. Microbiol. Rev. 55: 59-79.

Don, RH. and J .M. Pemberton. 1985. Genetic and physical map of the 2,4-
dichlorophenoxyacetic acid degradative plasmid pJP4. J. Bacteriol.
161:466-468.

Don, R.H., A.J. Weightman, H.J. Knackmuss, and K.N. Timmis. 1985.
Transposon mutagenesis and cloning analysis of the pathways for
degradation of 2,4-dichlorophenoxyacetic acid and 3-chlorobenzoate
in Alcaligenes eutrophus JMPl34(pJP4). J. Bacteriol. 161:85-90.

Englard, S., Balnchard, J .S., and C. F. Middlefort. 1985. Gamma-

14

butyrobetaine hydroxylase: stereochemical course of the hydroxylation
reaction. Biochemistry. 24: 1 1 10-1 1 16.

Evans, WC. and B.S.W. Smith. 1959. The photochemical inactivation and
microbial metabolism of the chlorophenoxyacetic acid herbicides.
Proc. Biochem. Soc. 26, p. 126.

Friedrich, 8., Meyer, M., and H. G. Schlegel. 1983. Transfer and expression
of the herbicide-degrading plasmid pJP4 in aerobic autotrophic
bacteria. Arch. Microbiol. 134, 92-7.

F ukumori, F ., and RP. Hausinger. 1993a. Alcaligenes eutrophus JMP134
“2,4-dichlorophenoxyacetic acid monooxygenase” is an a-
ketoglutarate-dependent dioxygenase. J. Bacteriol. 175:2083-2086.

Fukumori, F., and R. P. Hausinger. 1993b. Purification and characterization
of 2,4-dichlorophenoxyacetate/a-ketoglutarate dioxygenase. J. Biol.
Chem. 268:2431 124317.

F ulthorpe, R.R., C. McGowan, O.V. Maltseva, W.E. Holben, and J.M.
Tiedje. 1995. 2,4-dichlorophenoxyacetic acid-degrading bacteria
contain mosaics of catabolic genes. Appl. Environ. Microbiol. 61:
3274-3281.

Fulthorpe, R.R., A. N. Rhodes, and J. M. Tiedje. 1996. Pristine soils

I5

mineralize 3-chlorobenzoate and 2,4-dichlorophenoxyacetate via
different microbial populations. Appl. Environ. Microbiol. 62:1159-
1166.

Ghosal, D., You, I. S., Chatterjee, D. K., and A. M. Chakrabarty. 1985.
Microbial degradation of halogenated compounds. Science. 228: 135-
142.

Harker, A.R., Olsen, RH, and R. J. Seidler. 1989. Phenoxyacetic acid

 

degradation by the 2,4-dichlorophenoxyacetic acid (TFD) pathway of i
plasmid pJP4: mapping and characterization of the TFD regulatory
gene, tde. J. Bacteriol. 171, 314-20.

Haggblom, M. M. 1992. Microbial breakdown of halogenated aromatic
pesticides and related compounds. FEMS Microbiol. Rev. 103: 29-72.

Haughland, R.A., Sangodkar, U.M.X., Sferra, RR, and A. M. Chakrabarty.
1991. Cloning and characterization of a chromosomal DNA region
required for growth on 2,4,5-T by Peeudomonas cepacia AC1 100.
Gene 100, 65-73.

Helaakoski, T., Vuori, K., Myllyla, R., Kivirikko, K. I., and T. Pihlajaniemi.
1989. Molecular cloning of the alpha-subunit of human prolyl 4-

hydroxylase: the complete cDNA-derived amino acid sequence and

evidence for alternative splicing. Proc. Natl. Acad. Sci. USA. 86:
4392-4396.

Ka, J .O., W.E. Holben, and J .M. Tiedje. 1994. Genetic and phenotypic
diversity of 2,4-dichlorophenoxyacetic acid (2,4-D)-degrading
bacteria isolated from 2,4-D-treated field soils. Appl. Environ.
Microbiol. 60:1106-11 15.

Kaphammer, B., Kukor, J .J ., R. H. Olsen. 1990a. Regulation of deDEF by
tde of the 2,4-dichlorophenoxyacetic acid degradation plasmid pJ P4.
J. Bacteriol. 172, 2280-6.

Kaphammer, B., Kukor, J .J ., R. H. Olsen. 1990b. Cloning and
characterization of zde, the repressor/activator of tde, from the 2,4-
dichlorophenoxyacetic acid degradation plasmid pJP4. J. Bacteriol.
172, 5856-62.

Kasberg, T., Daubaras, D. L., Chakrabarty, A. M., Kinzelt, D., and W.
Reineke. 1995. Evidence the operons tcb, tfd, and clc encode
maleylacetate reductase, the fourth enzyme of the modified ortho
pathway. J. Bacteriol. 177(13): 3885-3889.

Kukor, J .J ., Olsen, RH, and J. S. Siak. 1989. Recruitment of a

chromosomally encoded maleylacetate reductase for degradation of

I7

2,4-dichlorophenoxyacetic acid by plasmid pJP4. J. Bacteriol. 171,
33 85-90.

Levaeu, J. H. J ., and J. R. van der Meer. 1996. The tde gene product can
successfully take over the role of the insertion element-inactivated
deT protein as a transcriptional activator of the tdeDEF gene
cluster, which encodes chlorocatechol degradation in Ralstonia
eutropha JMP134 (pJP4). J. Bacteriol. 178(23): 6824-6832.

Matheson, V. G., Forney, L. J., Suwa, Y., Nakatsu, C. H., Sexstone, A. J .,
and W. E. Holben. 1996. Evidence for acquisition in nature of a
chromosomal 2,4-dichlorophenoxyacetic acid/or-ketoglutarate
dioxygenase gene by different Burkholderia spp. Appl. Environ.
Microbiol. 62(7): 2457-2463.

Nikaido, H. and M. Vaara. 1985. Molecular basis of bacterial outer
membrane permeability. Microbiol. Rev. 49: 1-32.

Perkins, E. J ., and P. F. Lurquin. 1988. Duplication of a 2,4-
dichlorophenoxyacetic acid monooxygenase gene in Alcaligenes
eutrophus JMPl34(pJP4). J. Bacteriol. 170:5669-5672.

Perkins, E.J., M.P. Gordon, 0. Caceres, and PF. Lurquin. 1990.

18

Organization and sequence analysis of the 2,4-dichlorophenol
hydroxylase and dichlorocatechol oxidative operons of plasmid pJ P4.
J. Bacteriol. 172:2351-2359.

Pemberton, J .M. 1981. Genetic engineering and biological detoxification of
environmental pollutants. Residue Rev. 78, 1-11.

Pieper, D.H., Reineke, W., Engesser, K.H., and H. J. Knackmuss. 1988.
Metabolism of 2,4-dichlorophenoxyacetic acid, 4-chloro-2-
methylphenoxyacetic acid, and 2-methylphenoxyacetic acid by
Alcaligenes eutrophus JIVIP134. Arch. Microbiol. 150: 95-102.

Prescott, A. G. 1993. A dilemma of dioxygenases (or where biochemistry
and molecular biology fail to meet). J. Exp. Bot. 44:849-861.

Sinton, G. L., L. T. Fan, L. E. Erickson, and S. M. Lee. 1986.
Biodegradation of 2,4—D and related xenobiotic compounds. Enzyme
Microbiol. Techno]. 82395-403.

Spain, J .C. and P. A. van Veld. 1983. Adaption of natural microbial
communities to degradation of xenobiotic compunds: effects of
concentration, exposure, time, inoculum, and chemical structure. Appl.
Env. Microbiol. 45, 428-35.

Streber, W.R., KN. Timmis, and M.H. Zenk. 1987. Analysis, cloning, and

 

19

high level expression of 2,4-dichlorophenoxyacetic acid
monooxygenase gene tfdA of Alcaligenes eutrophus JMP134. J.
Bacteriol. 169:2950—2955.

Suwa, Y., A. D. Wright, F. F ukumori, K. A. Nummy, R. P. Hausinger, W. E.
Holben, and L. J. F omey. 1996. Characterization of a chromosomally
encoded 2,4-dichlorophenoxyacetate (2,4-D)/ rat-ketoglutarate
dioxygenase. Appl. Environ. Microbiol. 62: 2464-2469.

Tonso, N.L., V.G. Matheson, and WE. Holben. 1995. Polyphasic
characterization of a suite of bacterial isolates capable of degrading
2,4-D. Microbial Ecol. 30:3-24.

Top, E. M., Holben, W. E., and L. J. Fomey. 1995. Characterization of
diverse 2,4-D degradative plasmids isolated from soil by
complementation. Appl. Environ. Microbiol. 61 : 1691-1698.

Top, E. M., O. M. Maltseva, and L. J. Fomey. 1996. Capture of a catabolic
plasmid that encodes only 2,4-dichlorophenoxyacetic acid/or -
ketoglutarate dioxygenase (deA) by genetic complementation. Appl.
Environ. Microbiol. 62(7): 2470-2476.

van der Meer, J .R., de Vos, W.M., Harayama, S., A.J.B. Zehnder. 1992.

Microbiol. Rev. 56: 677-94.

20

World Health Organization. 1984. ZA-dichlorophenoxyacetic acid (2,4-D).
World Health Organization, Geneva, Switzerland.

World Health Organization. 1989. 2,4-dichloropheno>gacetic acid (2,4-D) -

environmental aspects. World Health Organization, Geneva,
Switzerland.

You, I. S., and D. Ghosal. 1995. Genetic and molecular analysis of a
regulatory region of the herbicide 2,4-dichlorophenoxyacetic acid
catabolic plasmid pJP4. Molec. Microbiol. 16:321-331.

Xun, L., and K. B. Wagnon. 1995. Purification and properties of component
B of 2,4,5-trichlorophenoxyacetic acid oxygenase from Pseudomonas

cepacia AC1 100. App]. Env. Microbiol. 61: 3499-3502.

Chapter 1:

The Phylogenetic Analysis of the AIpha-ketoglutarate-dependent

Dioxygenase Superfamily

ABSTRACT

Genetically, the or-ketoglutarate-dependent dioxygenase superfamily
is an extremely diverse group of enzymes that are thought to share a similar
biochemical mechanism. Using protein sequences of the known members of
the superfamily it was possible to perform a series of analyses that identified
clusters of related enzymes using current phylogenetic techniques. This
superfamily contains at least five distinct groups of enzymes that retain a
number of conserved residues and regions of high similarity. Since the
sequences of the distal members of the groups can share as little as 15%
identity, this analysis provides a clearer picture of the evolutionary
relationships of this superfamily than is revealed by general identity or

similarity analysis.

21

22

INTRODUCTION

The or-ketoglutarate-dependent dioxygenase superfamily is formed of
a highly diverse group of non-heme iron-(II) enzymes most of which are
thought to share a common or highly similar biochemical mechanism of
catalysis (Prescott,l993). These enzymes typically incorporate oxygen into
organic substrates, producing intermediates that may retain the oxygen
atoms or rearrange to produce one or more products. This mechanism is
initiated by the binding of ferrous iron, oxygen, a substrate, and, in most
cases, or-ketoglutarate. In the case of many of these enzymes including
deA, oxidative decarboxylation of the or-ketoglutarate occurs to yield
carbon dioxide, succinate, and a reactive iron-oxygen species. The insertion
of oxygen into the other substrate the occurs (reviewed in Prescott, 1993).
Members of this superfamily have been cloned fi'om bacteria, fungi, plants,
and animals. They exhibit moderate to very low sequence identity at the
DNA and protein sequence level, raising the possibility that this group is
polyphyletic. A clarification of the relatedness of the members of this group
could be useful in further biochemical and evolutionary investigations.

Multiple sequence alignment of divergent sequences allows the

detection of conserved residues and regions of proteins. These regions often

 

23

adopt a similar three-dimensional conformation, presenting specific residues
for structural considerations or for the binding of substrates and co-factors.
All of the enzymes in this superfamily are non-heme, iron metalloproteins
(Prescott, 1993). Typical iron ligands are charged at physiological pH,
including histidine, lysine, aspartic acid, glutamic acid, and cysteine, though
the tyrosine residue is uncharged. Conservation of these residues may
indicate that these amino acids are important for iron binding. Further, these
enzymes bind at least one other substrate, and may require other conserved
residues to maintain steric constraints and provide polar and hydrophobic
interactions essential for catalysis. Histidine residues have been implicated
as iron ligands in 2,4-D/0t-ketoglutarate-dependent dioxygenase (Fukumori,
1993b, Hausinger, 1997), flavanone 3B-hydroxylase (Britsch, 1993),
isopenicillin N synthase (Borovok, 1996, Jiang, 1991, Ming, 1991, Randall,

1993, Sim, 1994, Tan, 1996), and lysyl hydroxylase and prolyl hydroxylase

(Myllyléi, 1992), cysteine residues in clavamate synthase (Marsh, 1992), and

tyrosine residues in 4-hydroxyphenylpyruvate dioxygenase (Lindstedt,
1992). The only enzyme of the superfamily whose crystal structure has been
determined, isopenicillin N synthase, has two histidyl and one asparaginyl

iron ligands (Fujishima,1994, Roach, 1995). Substrate binding studies have

24

been done for for several of the proteins, and positively charged residues
essential for or-ketoacid binding have been suggested (Fukumori,1993b,
Majammaa, 1985, Pascal, 1985, Ng, 1991).

After initial sequence searches, it was noted that some a-
ketoglutarate-dependent dioxygenases show significant protein sequence
identity to another group of non-heme iron (11) dioxygenases which bind
ascorbate but not tat-ketoglutarate (McGarvey, 1992). It was also noted that
at least some a-ketoglutarate-dependent enzymes bind both substrates
(Fukumori, 1993, Prescott, 1993). In reactions where both compounds may
be bound, ascorbate is thought to reduce the oxidized iron metallocenter to
the ferrous state after the normal dioxygenation reaction in the case of the
ascorbate-dependent group, or after the uncoupled decarboxylation of a-

ketoglutarate in the case of the a-ketoglutarate-dependent enzymes

(Myllyléi, 1984). It has been proposed that the binding sites of these co-

substrates overlap (Myllyléi, 1984). The mechanism of the ascorbate-

dependent dioxygenases is proposed to be similar to that of the a-
ketoglutarate-dependent dioxygenases, forming a logical bridge for the

inclusion of these enzymes in any biochemical analysis of the superfamily.

25

In previous biochemical studies of a few members this superfamily of
enzymes, there has been some indication of the existence of conserved
histidines and possibly binding motifs, that may be involved in iron and/or

substrate binding (Borovok,1996, Bradley,1986, Fujishima,1994, Jiang,

1991, Lindsteadt, 1982, Ming, 1991, Myllyléi, 1992, Ng, 1991, Randall, F

1993, Roach, 1995, Ruetschi, 1993, Tan, 1996, Tiow-Sian, 1994). Since

there has been a recent proliferation in uncharacterized sequences due to

 

genome sequencing projects, the structural or phylogenetic relationships of
these enzymes to other enzymes that may be isofunctional or may have
homologous or analogous regions or domains could be useful. By
examining the members of the superfamily, it may be possible to identify
shared residues or binding motifs for iron, the or-keto acid used as a
cosubstrate, or one or more of the variable regions responsible for binding
the second substrates. By examining the sequences using phylogenetic
techniques, the superfamily can be assessed for internal clusters of related
sequences and compared directly for the identification of universally
conserved residues or highly conserved regions. This information could

then be used to inform further biochemical analyses.

26

RESULTS AND DISCUSSION

Since the a-ketoglutarate/ascorbate-dependent dioxygenase
superfamily is so diverse in both genetic and protein sequence, direct
comparison of the sequences is fairly ineffective in identifying potential
evolutionary relationships. A detailed series of sequence alignments and
phylogenetic analyses was used to group these enzymes into related clusters
for further analysis. The initial step of the analysis of these sequences
consisted of a series of multiple sequence alignments that identified groups
of sequences within the superfamily that are most closely related (>3 0%
identity), and typically consisted of those enzymes which utilize the same
substrates. The sequences within each of the groups were aligned and a
preliminary neighbor-joining analysis was used to determine the most
divergent members of the highly related cluster (i.e. - the isopenicillin N
synthase group with nine members). This most divergent pair was then used
in the later analyses to reduce the size of the data set due to computational
considerations. A number of sequences did not show significant similarity
to any of the other sequences of the superfamily.

Approximately seventy protein sequences (Table 1.1), several of

which were representative of a larger highly related group, were then

27

Table 1.1 Abbreviated names, accession numbers, enzyme names, and
references for strains used in this study. 4-hppd is 4-
hydroxyphenylpyruvate, ACC is l-aminocycloprOpane-l-carboxylate, and ?
denotes that function has not been biochemically confirmed.

  
   

          

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

hppd 1562148 LHomo sapiens 4-hppd dioxygenase fRuetschi, I993
4hppdb iP80064 {Pseudomonas Sp. 4-hppd dioxygenase Reutschi, 1992
4hppdca Eg1555806 Streptomyces sp. 4-hppd dioxygenase unpublished
4hppdh 1P32754 lHomo sapiens 4-hppd dioxygenase Reutschi, I993
4hppdm P49429 Mus musculus 4-hppd dioxygenase Endo, 1995
4hppdp 002110 Sus scrofa 4-hppd dioxygenase Endo, 1992
4hppdr P32755 Rattus norvegicus 4-hppd dioxygenase Gershwin, I987
pgbg X05130 Homo sapiens prolyl 4-hydroxylase alpha subunit jPihlajaniemi, 1987
phah X78949 Rattus norvcgicus prolyl 4-hydroxylase alpha subunit junpublished
P4l2l3 Zea mays Janthrocyanidin synthase 'Menssen, I990
Ebb $39897 Bos bovis IasparginyI-beta-hydroxylase Jia, 1992
Ecapp 000985 Malus domestica ACC oxidasc Dong, I992
[accara 006588 Ambadopsis thaliana ~ ACC oxidasc Gomez-Lint, 1993
Eceavo PI9464 Persea americana ACC oxidasc McGarvey, I992
[accbra P09052 Brassica juncea ACC oxidasc Pua, 1992
Ecccam P31528 Dianthus caryophyllus ACC oxidasc ,Wang, 1991
Iacckiw M97961 Actinida deliciosa LACC oxidasc iMacDiarmid, unpublished
Fecmel 004644 Cucumis melo YACC oxidasc Balague, 1993
ccorc 1.07912 Doritaenopsis sp. ACC oxidasc Nadeau, unpublsihed
Encepea P31239 Pisum sativum ACC oxidasc Peck, unpublished
hecpetl 1008506 Petunia hybrida ACC oxidasc Tang, I993
[accpet3 jgossm Petunia hybrida ACC oxidasc Tang, 1993
[accpet4 T008508 Petunia hybrida ACC oxidasc Tang, 1993
Eccpss ID13182 Pseudomonas syringac ACC oxidasc —_ I I lFukuda, I992
croml IP07920 Lycopersicon csculcntum ACC oxidasc lHoldsworth, I987 - T
ctom3 P2415? Lycopersieon csculcntum ACC oxidasc ‘Spanu, 1991
Eetom4 P05116 Lyeopersieon csculcntum ACC oxidasc Holdsworth, 1987
CS] L06213 Streptomyces clavuligerus claviminate synthase Marsh, 1992
csZ L06214 Streptomyces clavuligerus claviminate synthase Marsh, 1992
cs4 X84101 Streptomyces clavuligerus claviminate synthase Hodgeson, unpublished
dacs M63809 Streptomyces clavuligerus deacetylcephalosporin synthase TKovacevic, 1991
daocs M32324 Streptomyccs clavuligerus “deaetetoxycephalosporin synthase Kovacevic, I989
daocsc P11935 Cephalosporium acremonium deacteoxycephalosporin C synthase Samson, 1987
daocsn 003047 Nocardia lactamdurans deacteoxycephalosporin C synthase Coque, 1993
daocss Pl8548 ‘Streptomyces clavulgerus deacteoxycephalosporin C synthase Kovacevic, 1989
E8 XI3437 Lyeopersicurn csculcntum ethylene responsive gene Deikman, I988
8__2 S4997 5 Arabadopsis thaliana ethylene responsive gene Trentman, unpublished
coxl 064043 Escherichia coli taurine dioxygenase Echelard, 1988
EggdoZ #851766 jolmum melongena dioxygenase? __ j'l‘oguri, unpublished

 

Table 1.1 continued.

 

 

S27339HY

i4- -hppd d1oxygenase'7

28

 

Hummel 1992 F I

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Tetrahymena thermophila

fpro LA60235— ”@Emusculus 4- -hppd dioxygenase? Schofield, I991
[ga4 L37126 {Arabadopsis thaliana , growth enzyme Chiang, I995

bb P80193 Pseudomonas sp. AK-l gamma-butyrobetaine hydroxylase Ruetschi, 1993
lgibl 1X73314 Cucurbita maxima giberellin 20-oxidase Lange, 1994
lgibz TX8338I Arabadopsis thaliana giberellin 20-oxidase Phillips, unpublished
kib3 X83380gArabadopsis thaliana giberellin 20-oxidase Phillips, unpublished
[gib4 X83382 EArabadopsis thaliana giberellin 20-oxidase Phillips, unpublished
IgibS U33330 Spinacia oleracea giberellin 20-oxidase Wu, unpublished
h6h M62719 iTIyoscyamus niger ihyoscyamine hydroxylase Matsuda, I99]
h6h2 U20596 iSolanum lycopersicum jhyoscyamine hydroxylase Milligan, I995
ids $4797 I lHordeum vulgare ilow iron dioxygenase Okumura, I994
idsB ,D10058 iHordeum vulgare iron deficiency protein Nakanishi, 1994
isca EX03 I48 J'Cephalosporium acremonium isopenicillin-N synthase Samson, 1985
isen $05326 iEmericella nidulans isopenicillin-N synthase Wiegel, 1988
isfs 3P16020 [Flavobacterium Sp isopenicillin-N synthase Shiffman, 1990
isnl iP27744 ’Nocardia lactamdurans isopenicillin-N synthase Coque, I991
isp7 ”WPTO902 Schizosaccharomyces pombe lsexual development protein Saw, 1994
isfipcfiu "11308703 - lPenicillium chrysogenum isopenicillin-N synthase Carr, 1986
issc TM19421 Streptomyces clavuligerus isopenicillin-N synthase Leskiw, I988
issj P18286 :Streptomyces jumonjinensis isopenicillin-N synthase Shiffman, I988
issl P12438 [Streptomyces lipmanii isopenicillin-N synthase Wiegel, 1988
lb] M59183 IGallus domestica lysyl hydroxylase Myllyla, 199]
Iig 869666 [Legionella pneumophila legiolysin Wintemeyer, 1994
melA M59289 i_S_hewanella colwelliana 4-hppd dioxygenase Fuqua, I99]
mtoxl 277165 IMycobactcrium tuberculosis unknown MT genome sequencing
mtox2 274410 Etycobacterium tuberculosis unknown MT genome sequencing
n3dbar X58138 iHordeum vulgare 'Lflavone 3-dioxygenase Meldgaard, 1992
n3dca X69664 :Malus sp. _{naringenin-3-dioxygenase Britsch, 1993
n3dcam 005964 I Dianthus caryophillus inaringenin-3-dioxygenase Britsch, I993
n3dgr VII-090m “IT/This vinifera -.__._ "iflavone 3-hydroxylase -iTTrS-parvoli, I994
n3dpet 1X60572 Petunia hybrida lflavanone 3-beta-hydroxylase Britsch, 1992
n3dst 005965 Matthiola reana flavanone 3-hydroxylase Britsch 1993
Ecoxl $50963 LSaccharomyces cerevisiae unknown Wedler, unpublished
hrg S4426] [Ambadopsis thaliana unknown Callard, unpublished
trcl JC4220 lCoccidioides immitis T-cell reactive protein Wyckoff, I995
trc2 JC4220 iCoccidioides immitis T-cell reactive protein Wyckoff, 1995
yhc] lP8OI93 HPseudomonas sp. AK-I 1gamma-butyrobetaine hydroxylase Ruetschi, I993

 

 

 

29

analyzed using multiple sequence alignments and neighbor-joining analysis
producing five distinct and significant clusters containing most of the
sequences of the superfamily.

The a-ketoglutarate/ascorbate-dependent non-heme dioxygenase
group. This is the largest cluster and contains all of the ascorbate-dependent
dioxygenases, as well as many of the a-ketoglutarate-dependent
dioxygenases. Multiple sequence alignments of this group revealed the
conservation of a number of residues, including two invariant histidines, an
invariant arginine, and invariant aspartic acid, and significant regions of
similarity considering the breadth of evolutionary divergence of the
organisms and the number of sequences involved. Each of the noted
residues is found in one of two regions of high similarity across the group,
both occurring in the c-terminal region (Figure 1.1). The invariant histidines
are found in two motifs: [histidine-X-aspartic acid-3X-aliphatic residue-X-
two aliphatic residues] and [asparagine-7X-histidine-8-1OX-arginine-X-
serine] (Table 1.2). The [histidine-X-aspartic acid-SO-7OX-histidine] motif
was previously noted in sequences including the isopenicillin N synthase
subgroup (Boroviok, 1996). It is known that these histidines are iron ligands

in isopenicillin N synthase (Roach,l995). The most divergent sequences in

30

Figure 1.1 Multiple sequence alignment of the a-keto acid-dependent
cluster of the a-ketoglutarate-dependent dioxygenase superfamily. This
alignment does not include all of sequences of this cluster, but includes the
most divergent amino acid sequences when there are many closely related
sequences that form consistent subgroups (i.e. - the ACC oxidases). Two
strictly conserved histidines and one aspartic acid residue implicated in
ACC oxidase activity and known to be iron ligands in isopenicillin N
synthase are located in two regions of high similarity noted under the black
bar.

ispc l GDNHEEKHKV
issc : GTDAAAKKRV
ids
eggdoa
hbhE
hbh
srg
accpetB -LNG--SERDA
accpetH CG--AERDA
accpetl = -VNG--VERAA
accpea = -LNT

n3dcarn : IDG--EKRGEI
n3dcon = IDDXXGKRGEI
n3dca : IDGC--RRAEI
e&_a :
gib3 = ---SaDSTL
gibM : --—SGDSTL
giba = SCLASEA
gibS : AVSKA
gibl :
gaH
isp?
daocss

---AKGDERPA

-L

 

 

 
    
  
   

YNK

nn
I‘r

ispc
issc
ids : ...................
eggdo? -AKDAANNTA -------------------
hbhE -AKDAANNTN -------------------
hbh : ...................
srg '

accpet3
accpet"
accpetl
accpea -
n3dcarn :
n3dcon
n3dca :
eb_2 :
gib3 :
gin : ...................
gibE :
gibs : ...................
gibl :
gaH .
isp? :
daocss :

 

 

 

 

32

Figure 1.1 continued.

ispc
issc
ids
eggdoa
hbhE
hbh

srg
accpet3
accpetfl
accpetl
accpea
n3dcarn
n3dcon
n3dca
e&_2
gib3
gibu
giba
gibS
gibl
gaH
isp?
daocss

ispc
issc
ids
eggdoa
hbha
hbh

srg
accpet3
accpetH
accpetl
accpea
n3dcarn
n3dcon
n3dca
ea_B
gib3
gin
giba
gibS
gibl
gaH
isp?
daocss

------------------------------------ EHaDaIVAGYYLS
------------------------------------ DNP-HV NGYYKA‘
---------------------¢ ------------ - ----------- FCG
--------------------------------------- TGAAKHYSSSA
--------------------------------------- RGAATLYSSSA
------------------------------------ LPLEaKAKLv-VEG

 

---------------------------------------- TRIVLYYSN
------------------------------------ ASSFTG FS--TK
--4 --------------------------------- ASSFTG FS--TK
------------------------------------ ASSFVG FS--SK
------------------------------------ ASSFLG-FA--TK
----------e ------------------------- TNSFFG FA--$
------------------------------------ A-SLVSHLSSISKC
ISGIDFEAGSYPGEAPLPPSSIGYVLPPSSLANGEGSSflFDADflTTSNAI
------------------------------------ VPTHRRGFTGLESE

PEKKEVESFC LNPNFKPDHPLIGSKTPTHEVNVNPDEKKHPG RE- FA

 

 
  
 

PGRK VESFC LNPDFGEDHPNIAAGTPHHEVNLMPDEERHPR RP FC
SAFEILGEKY IDVLELLY- PLPSG ------- anDuPHKPEnLREgv- -G
KHYEsEEHRY RDVLEHSAN-LDGK ------- DRErijKPSREREuI- -G
KHYEEEEHRY RDVLEHSCN-LDGE ------- DKKTMPINPPRYREQI G
EGLSNEEFLY KDTLAHGCHPLDGD ------- LVNSNPEKPAKYREQV-A
FVVSEDGKLD ADLFFHTVGPVELR ------- KPHLFPKLPLPERprL-l
EAVEEEVTDLDUE- STFFLRHLPVS ------- NISEVPWLDEEYREEH -R
EAvaEEVTnLDuE-STFFLRHLva ------- NISEVPELDDEYREVM-
EcvaEEvrpnnuE- -STFFLKHLPIS ------- NISEVP>LDEEYREVH -R
EchgEIDDanE- STFFLRHLPVS----r--SISEIPDLDQDERKgn-K
SHLEGEvvan REIVTYFSYPTNSR ------- neruPnKpsc IK r-
SHLEGEAvaD REIVTYFSTPI RXR ------- DYSRIPEKP cErxvr-g
SHLEGEAvaD REIVTYFSYPI KAR ------- DYSRuPpKPuE RA9T-
DLHTCNKAAN RDTLACYHAP ----------- DPPKLGDLPAVCGEIH- -n
PwKEILsra- s- NDNSGSRTVODY ------- FSDTL aEFEaFGKv v
PMKEILSFa- gs NDNSGSRTVQDY ------- FSDTL aEFEaFaKv Y
PuKEILSFK- SPEEKIHsarvKDF ------- VSKKH IGYEDFGKV -v
PNKETLSFRYSDEDDDKSSKHVGNY ------- ISNLH TDFQEEGRy- -v
PMKEIFSLRCVAAaN---SSAAHDY ------- VLDTL PSFSHHGKE- Y
GPKV§PSLARLST ------------------- ISVNF PNITSTTAISY

AHG1ESISNEERISFYFGNDN'SKDRLLRPFGGPNKU STAGSSFRKALV
STA ITNTGS SD YSHCYSHGTADNLFP --------- SGDFERIuraYFE

133

115
125
125
136
123
115
115
115
115
123
129

128
187
127
133
13“
131
115
EDD
13“

Figure 1.1 continued.

ispc

issc

ids

eggdoa

hbha

hhh

srg

accpet3
accpetH -
accpetl =
accpea =
n3dcarn
n3dcon
n3dca

eB_E :
gib3 :
gin :
giba

gibS

gibl

gaH

isp?

daocss

ispc

issc

ids

eggdoE

hbhE

hbh

srg

accpet3
accpetH
accpetl -
accpea =
n3dcarn '
n3dcon
n3dca =
ea_2 :
gib3

gin

gibE

gibS

gibl

9a“

isp?

daocss

33

 

 

34

Figure 1.] continued.

ispc
issc
ids
eggdoa
hbhE
hbh

srg
accpet3
accpetu
accpetl
accpea
n3dcarn
n3dcon
n3dca
eG_E
gib3
gibu
giba
gibS
gibl
gaH
isp?
daocss

ispc
issc
ids
eggdoE
hbhE
hbh

srg
accpet3
accpetH
accpetl
accpea
n3dcarn
n3dcon
n3dca
eb_a
gib3
gin
giba
gibS
gibl
gaH
isp?
daocss

 

 
    

VSLINKNGG

-KHNVV

 
  
 
 
 
  
     
  

-EFAD-IY---
T ----- HK-EYNDG
DYHK--LYAGL
DYHK--LYARL
DYHK--LYAGL
DYflK--LYHGL
YRRKHAKDLEIARH

 
 
 
 
 
  
   
 

---EFTOK
---EFTOK
---EFTGK
---EFTOK
---EHTQK
---RTKAT
GNSYTSHTT

35

Table 1.2 Putative iron binding regions containing strictly conserved
histidine residues from the groups of the a-ketoglutarate—dependent
dioxygenase superfamily. Most of the conserved histidines are paired with
a conserved aspartic acid, asparagine, or glutamine residue. The regions of
the deA cluster are very similar to those of the gbb cluster, similar to that
of the a-keto acid cluster, and unlike the 4-hppd regions.

 

 

 

Subgroup ___Rutatiic_lmn_Binding_R§giona
His-1 His-2
a-ketoacid xHxDxxxl Nxxxxxxxl-i (8-15x) Rx2 b
gbb lxxHTD NleLHZ
deA WHxDxx4 lleDNRxxxH (13-14)RxT
deA/gbbC xHxD NxxxxH
4-hppd lxDI—lxxxN GxGlQl—l lex2 x3 1

 

“ Letters represent standard IUPAC amino acids, letters represent clusters of
similar amino acids (l-LIVM, 2-STA, 3-LF, 4-F Y), and x represents a non-conserved
residue with the group of enzymes.

b The asparagine residue in this motif is conserved in all but one of the proteins in
this group.

° This row represents the strictly conserved residues in an alignment of the
sequence of both the deA group and the gbb group (Figure 1.8).

36

the group that displays this motif have only 15% identity in amino acid
sequence.

The multilpe sequence alignment was used for phylogenetic analysis
using neighbor-joining (NJ) method (Figure 1.2). Representative sequences
were used for the analysis in Figure 1.2. Most of the sequences used are
plant enzymes, and, with the exception of ACC oxidase from Pseudomonas
syringe AK-l, cluster discreetly from the fungal and bacterialantibiotic
synthesis dioxygenases (bootstrap value of 93). Though many of the
sequences clearly cluster reliably with the other sequences with the same
substrate utilization profile (the ACC oxidases, the giberellin oxidases), a
number of the sequences show intermediate characteristics (ids,srg,isp7).
Interestingly, there is no phylogentic demarcation between those
dioxygenases that utilize a-ketoglutarate and those that do not (i.e. - the a-
ketoglutarate-dependent flavone hydroxylases (n3 dcam) and the
cephalosporin synthases (daocs)) are interspersed with enzymes, like
isopenicillin N synthase (inps) and the ACC oxidases, that do not require 0.-
ketoglutarate. This indicated that thwere is very little predictive vslue
regarding substrate utilization with intermediate sequences.

The most phylogentically interesting enzyme of this cluster is the

ACC oxidase from Pseudomonas syringe AK-l , It does not cluster with the

37

92'
99

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

169
29
160 I
196 L——-———-
25
7B
58
56
l 3' 3'
61
166 ]
fig j
109
93 109
89

 

 

 

 

 

me

 

p—i—

 

 

accpet4
accpet3

L— accpetl

accpea
egngZ
h6h2
h6h
ids
8P9

ea 2
n3dcarn
n3dca
gib4
gib3
gib2
gibS
gibi
954
38p?
ispc
issc
daocss.

Figure 1.2 Neighbor-joining tree of the multiple sequence aligned a-keto
acid cluster of the a-ketoglutarate-dependent dioxygenase superfamily.
This tree shows the analysis of representative sequences of this cluster. The
numbers noted are the bootstrap values for this tree, a number representing
the number of times per hundred that this tree is recreated during a random
reanalysis process. The closer the bootstrap is to 100, the higher the
statistical significance.

38

other ACC oxidases, nor with the prokaryotic enzymes. This enymes is
atypically large and has several regions of very high homology to the ACC
oxidase sequences from tomato. This could indicate that this enzyme is the
result of genetic exchange with its plant host or may have more than one
biochemical function.

Of the enzymes of this group, detailed biochemical data was available
only for isopenicillin N synthase. Studies of the metallocenter of this
enzyme revealed that two histidines and an aspartic acid residue are four of
the six ligands with the ferrous iron. These residues are among the strictly
conserved residues predicted in this work. All of these residues appear in
predicted regions of similarity. it is very likely that the conserved residues
that serve as iron ligands in isopenicillin N synthase also have the same

function in other members of this group.

The hydroxyphenylpyruvate dioxygenase group, which contains a
bacterial melanin synthase (MelA), bacterial and eucaryotic 4-
hydroxyphenylpyruvate dioxygenases (4-HPPD), and several other
membrane associated enzymes of unknown function. The 4-
hydroxyphenylpyruvate dioxygenase group enzymes are membrane

associated enzymes that are likely to function as the bacterial and eucaryotic

39

4-hydroxyphenyl pyruvate dioxygenases. Again, these sequences exhibit
two C-terminal invariant histidines, but only one resides in a region that is
significantly similar to those found in the other groups. These enzymes have
been referred to as a second functional subgroup of this superfamily
(Bradley, 1986), as they catalyze a reaction where the a-keto group as well
as the aromatic ring of the substrate is oxygenated using the same
mechanism as the two substrate dioxygenases (Pascal, 1985).

Database searches for sequence that are similar to human 4-
hydroxyphenylpyruvate dioxygenase noted distant but significant similarity
(27.9% identity, 51.7% similarity) with a reported melanin synthesis
enzyme of Shewanella colwelliana (F uqua, 1991), and two other
prokaryotic sequences. All statistically significant (as determined by
BLASTP (Altshul, 1990)) were examined for conserved resides that were
potential iron ligands. These sequences were then aligned (Figure 1.3) and
found to be highly similar, containing many conserved residues even though
the group contains eukaryotic and prokaryotic sequences. The alignment
noted several potential iron ligands: three conserved histidines, three
conserved tyrosines, two aspartic acids, and two glutamic acids.

Phylogenetic analysis using the NJ method indicates that the prokaryotic

40

and eukaryotic sequences are significantly related, but form discreet
subclusters within the group (bootstrap of 100, Figure 1.4).

Human 4-hppd and the Pseudomonas MelA show strong c-terminal
similarity (52% identity over 71 amino acids). This c-terminal region is the
same region containing the conserved histidines and regions of similarity
implicated in iron binding in the a-keto acid dependent group. These
regions do have some similarity, leaving open the possibility of the presence

of specific binding motifs for these enzymes.

The gamma-butyrobetaine hydroxylase group, has only one well
characterized seqeunce (gbb), and contains several new sequences of
unknown function. Gamma—butryobetaine hydroxylase is an on-
ketoglutarate-dependent enzyme, and has been partially biochemically
characterized (Blanchard, 1983, Englard, 1985, Ng, 1991). Multiple
sequence alignment of these sequences indicated the presence of several
conserved residues, including four aspartic acid residues, three arginines,
two histidines, and an asparagine (Figure 1.5). This group also displays a c-
terminal histidine-X-aspartic acid motif and a glutamate-4X-histidine motif

within regions of high similarity. Binding studies using gbb indicate that

4]

Figure 1.3 Multiple sequence alignment of the 4-hydroxyphenyl-
pyruvate dioxygenase cluster of the a-ketoglutarate-dependent
dioxygenase superfamily. This cluster displays a considerable number of
conserved residues.

thpdb
lig
mela
trcl
trcE
thpd
thpdh
thpdm
fpro
thpdr
"hPPdP
thpdce
f-ag
thpdsa

thpdb
lig
mela
trcl
trcE
"hppd
thpdh
Mhppdm
fpro
thpdr
I”Hump
thpdce
f-ag
Mhppdsa

' thpdb
lig
mela
trcl
trca
thpd
thpdh
thpdm
fpro
thpdr
“hppdp
thpdce
f-ag
thpdsa

u n u an n

42

h HGdg dba V d

   

ATPUGHF

A

 

 

EAmASTHcPcACAm FKVK AKAAF HAIAHch----Afla- DAP

SYFAAEHGPSVCGHAFRVEESQKAYKRALEL6A0----EIHIHTG

    

7VKHGDGVfiDIAF-VLDCEH

 

VEHGDGVHDIAFvaDCDE
vchDEv:Dv|FrvsDLn&

 

AHAYAIF

u.

HGAWSEAE

ELKDE

   

43

Figure 1.3 continued.

y 9 pg
thpdb : GEGSSIYD--
lig : —HE
mela : --
trcl :
trca :
thpd :
thpdh
thpdm
fpro
thpdr
"hppdp
thpdce
f-ag
thpdsa

thpdb

119

male

trcl

trcE

thpd
thpdh .
thpdm :
fpro =
thpdr
Mhppdp
thpdce
f-ag
thpdsa

 

 

 

 

 

thpdb

lig

mela

trcl -
trca :
“hppd :
thpdh :
thpdm :
fpro :
thpdr :
"hppdp '
thpdce
f-ag

: m‘IKnPINEPA KRL,
uhppdsa : ATEYSALHSKVVAEGTLKVKFPINEPALAKKF--SOIDEYLEFHI$

 

u u u u n n u to n u n u

44

Figure 1.3 continued.

6 GbGHbAl 3 lb

thpdb
lig
mela
trcl
trca
thpd
thpdh
thpdm
fpro
thpdr
"hppdp

thpdsa

thpdb
lig
mela
trcl
trca
thpd
thpdh
thpdm
fpro
thpdr
qhppdp
thpdce
f-ag
thpdsa

tbF E IaR n GFG GNF LF
thpdb
lig
mela
trcl
trcE
thpd
thpdh
uhppdm :
fpro :
thpdr :
“hppdp
thpdce
f-ag
thpdsa

E 90

 

 

45

me 4hPPdm
99 r—{Fpr‘o
160 9—Q ——-4hppdr‘
106 4hppd
—fifid 4hppdh
' *-———- 4hppdp
4hppdca
F-ag
trcl
tPC2
4hppdsa
mela

q?[ 4hppdb

139

 

 

 

 

 

166

 

 

1‘39

 

 

 

 

 

 

 

r—w

166'

 

 

 

 

 

 

 

Figure 1.4 Neighbor-joining tree of the multiple sequence aligned 4-
hydroxyphenylpyruvate (4-hppd) dioxygenase cluster of the OL-
ketoglutarate-dependent dioxygenase superfamily. The numbers noted
are the bootstrap values for this tree, a number representing the number of
times per hundred that this tree is recreated during a random reanalysis
process. The closer the bootstrap is to 100, the higher the statistical
significance. This tree shows highly significant clustering of eucaryotic and
procaryotic 4-hppd-like sequences.

46

Figure 1.5 Multiple sequence alignment of the gamma-butryobetaine
hydroxylase cluster of the a-ketoglutarate-dependent dioxygenase
superfamily. This cluster contains a single well characterized enzyme, gbb,
and several other recent sequences that probably have a similar function.
This cluster has a number of strictly conserved residues, including four
aspartic acid residues, three arginines, two histidines, and an asparagine.

gbb

gbb?
yhcl
gbb3

gbb

gbba
yhcl
gbb3

gbb

gbba
yhcl
gbb3

gbb

gbb?
yhcl
gbb3

gbb

gbb?
yhcl
gbb3

gbb

gbba
yhcl
gbb3

gbb

gbbB
yhcl
gbb3

gbb

gbba
yhcl
gbb3

   
    

47

r 1 a a
4NKIADY TFPLI PL ---------- ASA .
HLSNLLI NIRNA KL .......... Agv
"LRSNLC GSRIL RLTTTPRTYTSAATA

 

f _L5LRD

 

 
 

 
 

 

TnLEFDv VEan--ARKLDID ANC K E ESGVL
_---: ------ AR DIaIs *GKS K KDGGHHD BLG EiDY
(CDTTSISKIKHS QVIID ATNS a v IchaaK KIfiN

E DDGflA A HPE RAH
- ASE KIR

yFLvADv VP'EDIavEJAST Em-

L
AYDA_éSL§ EAARPH HRw ---aGIs ------- EPv ;GAEfl
----- NPs aEARRRR KVY PEDTucKAEIEGKEKKs SEEFfl
KGSSFVSPATRKaEsRYRPa NKRIL LKDNvKDLsts EFIDP
---REGKVEKNVSNDNEIYE NSKSL KDVPR ----------- as

. KVRDVG HGV TE- PGfi --IPb AK
QAVCID VLKGA GGVRGfi A
FISGT SSSSEG

 

   

as sKNLvKT IIVDGTEGTSEAI---
a b m P
G-VLFDyRsK D DSTAY----T F- EHTDL TRELa
SLK D SNHAY-—--A N66 FHTDf SESHP
gcTFDvNAsaA stAH----Y MK ’ FL v

-FLVFENSITNDEPAYEDTATGSDE GPHTDGTYF

fVD iae LR e e S L

 
  

GLaFL C VNDAT§ N T ----- FVDGF ALRI A A,
EEEQ ----- FVDGFHY aLRv

_ EGEDPNTRPNNYFVDAFTA RR VRESDF A

GIa'vEm‘ TP KTLeEDIv ----- LVDFsTc KLRNES D

    

---------- RN DRHS YRCT P--:
TTds EEIEEGYIWHEIS GEE” CRHKTI;R .........
GIVE N IYENGDKRYY s LIEHHDI INEDNTLLGNYEALIKCI
ENTK SHHTLEGSPPGSSIHSV VSLIKPV VIEREsF -----------

103
101
125

55

136
IHH
170
11%

lbfl
192
815
151

211
233
855
205

859
872
301
895

863
306
395
E73

Figure 1.5 continued.

gbb

gbbB
yhcl
gbb3

gbb

gbb?
yhcl
gbb3

gbb

gbb?
yhcl
gbb3

 

LFKS
ETIK IEHTE.

r

f

DP .G-D H
G RN PEKA
L--- SSNGGNK

TfiasT----Bom

  
  

TLT
LK

 

48

  

3%
D

  

QEUCFDNLRVLHARP

DEELNENRLLHTR'

RD

86

aERLa PENCC IFNNRRILHANS

 

LSLgaNaPSA
-LK EE FPHDK -----

SQTSL -----

305
33“
365
305

363
H81
H55
365

49

one of the positively charged residues is involved in a-ketoglutarate binding

in this strain (N g, 1991).

The 2,4-dichlorophenoxyacetic acid/a-ketoglutarate-dependent
dioxygenase (deA) group, which contains deA, close relatives (>90%
identity at the protein sequence level), and distant relatives (>27% identity
at the protein sequence level). Only one of the distantly related sequences
has been characterized to any extent, and is know as taurine dioxygenase
(TauD, noted in this work as ecox1)(van der Ploeg,1996). Multiple
sequence alignment of this group indicated that there are a number of
conserved residues, including three histidines, three threonines, two
arginines, two aspartic acid residues, and an asparaginine residue (Figure
1.6). There is a c-terminal pair of histidine motifs: [tryptophan-histidine-X-
aspartic acid] and [tryptophan-aspartic acid-asparaginie-3x-histidine]. The
forat motif somewhat resembles the first c-terminal histidine region of the
a-keto acid group. NJ analysis of these sequences indicated that the more
distantly related sequences have diverged significantly from deA and from
each other, as indicated by a lack of reliable branches (Figure 1.7).
Chemical inactivation studies of deA indicated that histidines are very

likely to be the essential iron lignads (Fukumori, 1993a).

50

Figure 1.6 Multiple sequence alignment of the 2,4-D/0L-ketoglutarate-
dependent dioxygenase cluster of the a-ketoglutarate-dependent
dioxygenase superfamily. This cluster contains two highly similar deA
sequences, deA and deArasc, and several other recent sequences that
probably have a similar mechanism of action. Ecoxl encodes taurine
dioxygenase and id known as TauD. The other enzymes of this subgroup are
not biochemically characterized. The members of this subgroup share a
number of strictly conserved residues, including three histidines, three
threonines, two arginines, two aspartic acids, and an asparagine residue.

rascA
tfdA

ecoxl
mtoxl
scoxl
utoxE

rascA
tfdA

ecoxl
mtoxl
scoxl
mtoxE

rascA
tfdA

ecoxl
mtoxl
scoxl
mtoxa

rascA
tfdA

ecoxl
mtoxl
scoxl
mtoxa

rascA
tfdA

ecoxl
mtoxl
scoxl
mtoxa

rascA
tfdA

ecoxl
mtoxl
scoxl
mtoxE

51

5 93 b g 1 L“ b b

INSEY H LFVGavDN ALaGA sP EvR ENE ‘
VVANP H LFAAGVED DLREA GS EVR ERE
Ls TPLG thgalg AD TRP SDNGPEG YHA
IT KK .,° .,RLGGD DP AyN RAA
LK KK IEINjiaLT—D SD AKD ALE AgfivavFRN
mT KVKfiEGWGAQVTGVD- -PKN DnITTn RDI YTNKLVVLKD

     
      

  

 
     

GPLDOE EEGGEIKvNaRPSRlKYA L D';N
aPLsa EGGFIKVNGRPSRFKYA aLADI;N
a-AITPQr - , E HI -------- PPv PHAEGvDEII
GHQLDEA a--E .VGLL TPIG -------- HP--AAIALADDEP
aNFAD GPDYVTE GEH :EHI -------- HGTSGHPGNNPELH
---VHPSPREFEKLG-II VPYY ------ EPnIHBEaHPBIFv
O———-a
wH D 5
vDGKvAEAD REVVGNFANdLmHSDs EEPAAR SAEEAIVL
LDGKvAGRD EEVVGNFANGLstDs anAARYs_<‘
--------- LDTHNDNPPDNDNNHTDV, IETPPAGA ,
--------- I PINSEFGKANRwHIDV, AANYPAASILBAvSL
FRRPDAEEF RVFDDSTSSGGwHTDV ELaPP~YTFFXVVEG
SATEEGGGVPK HPEEFAfisnvLPLAI

 

 

 
 
  

PPSGIITEFC AR‘gvDALPRDLGSE R E YALNERFL GD
PSTG TLwTSGIA Y ALSVPF aL R E DF KEPPEYEY
PSYGGITLwAg TAAAYAELPEPL CLTEN w L TN YDYVTT -
PDGGGDTLFA TIEAFDRLSKPLGDFIST HVI SS --------
PGHDRGTYFI gLAvaQSLPAAKBDPARETVIT DP R- HIKEEP

1 1 H
PPSG,> R‘,YDDLPEDFEKE EEIR E YALHERFI GD
Ed

 

 

 

IDYSEAORNAAPP Mm --------- PLVRTHAGEGRKF g-

TEYSEEG-NAnPP'sw .........
v
RKTEEEHaRwREA VA NP pg--LfipVVRTHvadxaA NEG

:gEGRPVA-{GRL
AS EGLPVA--EGR AWELLE ATaR-
TT £VDVSEK--ESEA SFLFAHITKA -

VRSFVGLDSH--ESR FEVLGRRITNB-

&KIVEKRa--SES NFLYNLSlESSH
IEDKDGNPVDPE aELmAATGaLDP

 

132
133
llS
11%
135
11?

l7?
1??
150
155
153
lb}

212
213
202
19%
205
eon

850
250
EHO
23?
EM"
EH?

Figure 1.6 continued.

rascA
tfdA

ecoxl
mtoxl
scoxl
mtoxE

Ag
f

52

 

53

 

 

 

 

as . ecoxl
E? mtoxl
scoxl

' thR

rascR
mtox2

 

 

 

J-

1‘30

 

 

Figure 1.7 Neighbor-joining tree of the multiple sequence aligned 2,4-
dichlorophenoxyacetic acid/a-ketoglutarate-dependent (deA)
dioxygenase cluster of the a-ketoglutarate-dependent dioxygenase
superfamily. The numbers noted are the bootstrap values for this tree, a
number representing the number of times per hundred that this tree is
recreated during a random reanalysis process. The closer the bootstrap is to
100, the higher the statistical significance. The very closely related (>90%
identity) deA and RascA cluster reliably, with the other sequences being
nearly equally evolutionary distant from one another.

54

Figure 1.8 Multiple sequence alignment of the 2,4-D/0L-ketoglutarate-
dependent dioxygenase cluster and the gamma-butyrobetaine
hydroxylase cluster of the a-ketoglutarate-dependent dioxygenase
superfamily. These clusters show the most similarity within the
superfamily, sharing two regions of high similarity containing histidine and
aspartic acid or asparagine residues.

gbb
gbb?
yhc1
gbb3
rascA
tfdA
ecoxl
mtoxl
scoxl
ntoxE

gbb
gbba
yhcl
gbb3
rascA
tfdA
ecoxl
mtoxl
scoxl
mtoxa

gbb
gbbB
yhcl
gbb3
rascA
tfdA
ecoxl
mtoxl
scoxl
mtoxE

gbb
gbba
yhcl
gbb3
rascA
tfdA
ecoxl
mtoxl
scoxl
mtoxa

55

   
   

PEaTuGKAEIEGK LKKFs HEEFHKNEQVVH FLa7
NKRI DNVKDLLSVSYN- EFIDPKDDSKLFO TLvN
NSKS

--aGIs ------- IPIYD- HfiAvnaDFDTiLIHLL_

    
  
  
    
 
 

 

vva EDIDLREALGSTEVR IERL
ERLs SGADLTRPLspNGFE

DLIT . DGgRLGGDLDPAALN

GGEK . NGIaLT- DLs AAKD 1

-- T TGLD- -PKNED ITTD »

I .
S TPSSSSEGLTIGKICERI
DGVEGTSEAT---EKLCGSL
RGQPLDGT b
RGGPLSG

    

ELHI -------- HPVYPHAEGVDE
TPIG -------- HP--AAIALADD

RNGNFAD KLHI -------- HQTSGHPGNNPE
GIVPYY ------ EPHYHHEDHPEI
H
H D

     
 
 

FDggs ADADSNfiY----TAF-IEPLHTD PTRELDPG
FE ;

     

--------- TPINSEFGKANRAHTD :FAANYP;A:
RPDAEEFARVFDDSTSSGGAHTDPéYELar

 

1H2
15H
158
1H5
172
178
155
159
155
155

56

Figure 1.8 continued.

gbb
gbb?
yhcl
gbb3
rascA
tfdA
ecoxl
mtoxl
scoxl
mtoxE

gbb
gbba
yhcl
gbb3
rascA
tfdA
ecoxl
mtoxl
scoxl
mtoxB

gbb
gbb?
yhcl
gbb3
rascA
tfdA
ecoxl
mtoxl
scoxl
mtoxa

gbb
gbbE
yhcl
gbb3
rascA
tfdA
ecoxl
mtoxl
scoxl
mtoxa

SRFILGD

DYVTTK--PLT

 

H

Mr H
YRRFIaNTREPRFCFTRRLEAGQLNCFDNRRvFHARDAFDPASG-
HKTFTEYCYaPRNNLgFRLELGDTVLwflNaRLLHTRDGFRNAPEK
LNLFESHINDFNNaFaLaLPENcCVIFNNRRIPHANSL-—-TSSN
YEKFSKICHNPDNSIEISLRPGSVIHIDNFRIEHSETSFaGY---
fiR-RYDVTAR
TaR EFVYRH----RENVG§--LVHwDNRCV:HRER-RYDISAR

  

 

 

  

 

ATaP-KFVYLH----squcDLVLvanNRcngR

A _ .

ITKP-EFaVRw----R@aPND--IAImDNRVTaHYLNADYLPa-R

ITHP-ENTIBw----NMAPGD--VAIUDNRATGHR‘IDDYDDa-H

VESSHDLaLRA-—--KwEPHs--vv1wDNRR aHsgVIDwEEPIH

GaLDPEYGSPFIHTGHYQVGD--IILMDNRV nHR KHGSAAGTL
q I ,

DRHFaGCYEDBpEL : 250

ARTLTGCYFDngQ : 275

aauL’ : 303

-RaHCGCY> _d : aLe

EELRi . ' ' : aaa

KELREATT as?

RINHEAII L” , : 3??

LLNHQVIL_ h . : avu

SHAFEIEPGAERPQ : eaa

'TTYBLIHEDGLKT : agq

 

132

EN?
80?
BBB

225
283
231
BEG

57

Interestingly, the c-terminal motifs of this cluster and the gamma-
butyrobetaine hydroxylase group are quite similar. The amino acid
sequences of deA and be have display low identity but a high (53%)
similarity. When optimally aligned, only a limited version of the two c-
terminal histidine motifs emerge as conserved between the two groups
(Figure 1.8). These motifs - [histidine-X-aspartic acid] and [asparagine-4X-
hisidine] - occur in the region of greatest similarity between the two groups.
These results indicate that deA and be may not be evolutionarily related,

but may share an active site with a similar three dimensional configuration.

The clavamate synthase group, with several very closely related
members. This group appears to form a unique group in this superfamily,
and bears little resemblance to any other sequence in the sequence
databases. Some initial biochemical work indicates that cysteine residues
are essential ligands in the metallocenter of this enzyme. These sequences
are so similar, performing alignment and phylogenetic analysis do not

generate any additional information.

The remaining sequences. Several of the other sequences (lysyl

hydroxylase, prolyl hydroxylase, aspartyl hydroxylase) did not share

58

sufficient similarity to any of the other sequences to be included in the
groups defined above. These sequences are also thought to use histidine

residues to bind iron, and several short regions of similarity that contain

histidine residues have been noted (Myllyla,1992). Thses sequences may be

distantly related, but this relationship is not revealed using the techniques

used in this study.

59

SUMMARY

It was possible to generate a phylogeny of the a-ketoglutarate-dependent
dioxygenase superfamily that identified distinct clusters of divergent
enzymes that retain patterns of conserved residues. This analysis provides
some insight into the evolution of these enzymes, illustrating that the
superfamily is likely to be polyphyletic, though there are several clusters of
enzymes within the superfamily that are probably related or share domains
or regions which are related. The clustering of the a-ketoglutarate—
dependent dioxygenases with the ascorbate-dependent dioxygenases is
intriguing. Further, the presence of regions of similarity containing invariant
histidines among severalk of the groups could indicate convergent evolution
of the metallocenters from disparate ancestral enzymes or the divergent
evolution of ancient motifs capable of binding metals. Several of the
enzymes of this superfamily do not exhibit similarity to nay of the other
sequences, though do contain short regions of similarity. These enzymes
may be unrelated to any of the outline subgroups outlined in this work, or
may simply be to divergent to be identified with these methods. Since the
initial analysis of this superfamily, new sequences have become available,

including several bacterial genomes, providing many new but as yet

60

unstudied members of the outlined subgroups of the a-ketoglutarate-
dependent superfamily. It is hoped that this kind of analysis will aid the

ongoing biochemical analysis of these enzymes.

61

REFERENCES

Altschul, S. F., Gish, W., Miller, W., Myers, E. W., and D. J. Lipman. 1990.
Basic local alignment search tool. J. Mol. Biol. 215: 403-410.

Balague, C., Watson, C. F ., Turner, A. J ., Rouge, P., Picton, S., Pech, J.
C.,and D. Grierson. 1993. Isolation of a ripening and wound-induced
cDNA from Cucumis melo L. encoding a protein with homology to
the ethylene-forming enzyme. Eur. J. Biochem. 212: 27-34.

Borovok, I., Landman, 0., Kriesberg-Zakarin, R., Aharonowitz, Y., and G.
Cohen. 1996. Ferrous active site of isopenicillin N synthase: genetic
and sequence analysis of the endogenous lignads. Biochemistry 35:
1981-1987.

Blanchard, J. S., and S. Englard. 1983. Gamma-butyrobetaine hydroxylase:
primary and secondary tritium effects. Biochemistry 22: 5922-5929.

Bradley, F. C., Lindsteadt, S., Lipscomb, J. D., Que, L. Jr., Roe, A. L., and
M. Rundgren. 1986. 4-hydroxyphenylpyruvate hydroxylase is an
iron-tyrosinate protein. J. Biol. Chem. 261: 1 1693-11696.

Britsch, L., Ruhnau-Brich, B., and G. Forkman. 1992. Molecular cloning,
sequence analysis, and in vitro expression of flavanone 3 beta-

hydroxylase from Petunia hybridaJ. Biol. Chem. 267: 53 80-53 87.

62

Britsch, L., Dedio, J ., Saedler, H., and G. F orkman. 1993. Molecular
characterization of flavanone 3 beta-hydroxylases. Consensus
sequence, comparison with related enzymes and the role of conserved
histidine residues. Eur. J. Biochem. 217: 745-754.

Callard, D., Axelos, M., and L. Mazzolini. 1994. Submitted to the EMBL
Data Library.

Carr, L. G., Skatrud, P. L., Scheetz, M. E. II, Queener, S. W., and T. D.
Ingolia. 1986. Cloning and expression of the isopenicillin N synthase
fi'om Penicillium chrysogenum. Gene 48: 257-259.

Chiang, H. H., Hwang, 1., and H. M. Goodman. 1995. Isolation of the
Arabidopsis GA4 locus. Plant Cell 7: 195-201.

Coque, J. J. R., Martin, J. F ., Calzada, J. G., and P. Liras. 1991. The
cephamycin biosynthetic genes pchB, encoding a large multi-
domain peptide synthetase, and pch of Nocardia lactamdurans are
clustered together in an organization different from the same genes in
Acremonium chrysogenum and Streptomyces clavuligerus. Mol.
Microbiol. 5: 1125-1133.

Coque, J. J. R., Martin, J. F., and P. Liras. 1993-Characterization and
expression in Streptomyces lividans of cefD and cefE from Nocardia

lactamdurans: the organization of the cephamycin gene cluster

63

differes from that in Streptomyces clavuligerus. Mol. Gen. Genet.
236: 453-458.

Deikman, J. and R. L. Fischer. 1988. Interaction of a DNA binding factor
with the 5'-flanking region of an ethylene-responsive fi'uit ripening
gene from tomato. EMBO J. 7: 3315-3320.

Echelard, Y., Dymetryszyn, J ., Drolet, M., A. Sasarman. 1988. Nucleotide
sequence of the hemB gene of Escherichia coli K12. Mol. Gen.
Genet. 214: 503-508.

Endo, F., Awata, H., Tanque, A., Ishiguro, M., Eda, Y., Titani, K., and I.
267 : 2423 5-24240.

Endo, F., Awata, H., Katoh, H., and I. Matsuda. 1995. Genomics. 25: 164-
169.

Englard, S., Blanchard, J. S., and C. F. Midlefort. 1985. Gamma-
butyrobetaine hydroxylase: stereochemical course of the
hydroxylation reaction. Biochemistry 24: 1110-1116.

Fukuda, H., Ogawa, T., Ishihara, K., Fujii, T., Nagahama, Omata, T., Inoue,
Y., Tanase, S., and Y. Morino. 1992. Biochem. Biophys. Res.
Commun. 188: 826-832.

Fukumori, F., and RP. Hausinger. 1993a. Alcaligenes eutrophus JMP134

64

“2,4-dichlorophenoxyacetic acid monooxygenase” is an a-
ketoglutarate-dependent dioxygenase. J. Bacteriol. 175:2083-2086.

Fukumori, F., and R. P. Hausinger. 1993b. Purification and characterization
of 2,4-dichlorophenoxyacetate/ cat-ketoglutarate dioxygenase. J. Biol.
Chem. 26822431 1-24317.

Fuqua, W. C., Coyne, V. E., Stein,D. C., Lin, C. M., and R. M. Weiner.
1991. Characterization of melA: a Gene encoding melanin
biosynthesis from the marine bacterium Shewanella colwelliana.
Gene 109: 131-136.

Gershwin, M. E., Coppel, R. L., Bearer, E., Peterson, M. G., Sturgess, A.,
and I. R. McKay. 1987. Molecular cloning of the liver-specific rat F
antigen. J. Immunol. 139: 3828-3833.

Gomez-Lim, M. A., Valdez-Lopez, V. M., Cruz-Hemandez, A., and J. L.
Saucedo-Arias. 1993. Isolation and characterization of a gene
involved in ethylene biosynthesis from Arabadopsis thaliana. Gene
134: 217-221.

Hodgson, J. E., F osberry, A., Rawlinson, N. S., Ross, H. N. M., Neal, R. J .,
Amell, J. C., Earl, A. J ., and E. J. Lawlor. (GenBank, unpublished).

Holdsworth, M. J ., Bird, G. R., Ray, J ., Schuch, W., and D. Grierson. 1987a.

65

Structure and expression of an ethylene -related gene from tomato.
Nucleic Acids Res. 15: 731-739.

Holdsworth, M. J ., Schuch, W., and D. Grierson. 1987b. Nucleotide
sequence of an ethylene-related gene fiom tomato. Nucleic Acids
Res. 15: 10600-10600.

Holloway, B. W., 1965. Variations in restriction and modification of
bacteriophage following the increase in growth temperature of
Pseudomonas aeruginosa. Virology 25: 634-642.

Hopkinson, 1., Smith, S. A., Donne, A., Gregory, H., Franklin, T .J., Grant,
M. E., J. Rosamond. (unpublished). The complete cDNA sequence of
rat prolyl 4-hydroxylase alpha subunit.

Hummel, R., Norgaard, P., Andreasen, P. H., Neve, S., Skjodt, K.,

J omehave, D., and K. Kristiansen. 1992. Tetrahymena gene encodes a
protein that is homologous with the liver-specific F-antigen and
associated with membranes of the Golgi apparatus and transport
vesicles. J. Mol. Biol. 228: 850-861.

J ia, S., Van Dusen, W. J ., Diehl, R. E., Kohl, N. E., Dixon, R. A. F .,
Elliston, K. 0. Stem, A. M., and P. A. Friedman. 1992. cDNA cloning
and expression of bovine aspartyl (asparaginyl) b-hydroxylase. J.

Biol. Chem. 267(20): 14322-14327.

66

Johnson, M., Andrerws, S., Brinkman, R., Cooper, J ., Ding, H., Dover, J .,
Du, Z., Favello, A., Fulton, L., Gattung, S., Geisel, C., Kirsten, J .,
Kucaba, T., Hiller, L., Jier, M., Johnson, L., Langston, Y.,

Latreille, P., Louis, E.J., Macri, C., Mardis, E., Menezes, S., Mouser,
L., Nhan, M., Rifldn, L., Riles. L., St. Peter, H., Trevaskis, E.,
Vaughan, K., Vignati, D., Wilcox, L., Wohldman, P., Waterston, R.,
Wilson, R., and M. Vaudin. 1994. Science. 265: 2077-2082.

Kim, W. T., Silverstone, A., Yip, W. K., Dong, J. G., and S/ F. Yang.
Induction of 1-aminocyclopropane-3-carboxylate synthase mRNA by
auxin in mung bean hypocotyls and cultured apple shoots. Plant
Physiol. 98: 1530-1534.

Kovacevic, S., Weigel, B. J ., Tobin, M. B., Ingolia, T. D., and J. R. Miller.
1989. Cloning, characterization, and expression in Escherichia coli of
the Streptomyces clavuligerus gene encoding deacetoxycephalosporin
C synthetase. J. Bacteriol. 171: 754-760.

Kovacevic, S., and J. R. Miller. 1991. Cloning and sequencing of the beta-
lactam hydroxylase gene (cefF) from Streptomyces clavuligerus:
Gene duplication may have led to separate hydroxylase and

expandase activities in the Actinomycetes. J. Bacteriol. 173: 398-400.

67

Lange, T., Hedden, P., and J. E. Graebe. 1994. Expression cloning of a
gibberellin 20-oxidase, a multifunctional enzyme involved in
gibberellin biosynthesis. Proc. Natl. Acad. Sci. USA. 91: 8552-8556.

Leskiw, B. K., Aharonowitz, Y., Mevarech, M., Wolfe, S., Vining, L. C.,
Westlake, D. W., S. E. Jensen. 1988. Cloning and nucleotide
sequence determination of the isopenicillin N synthetase gene from
Streptomyces clavuligerus. Gene 62: 187-196.

Lindsteadt, S., and M. Rundgren. 1982. Blue color, metal content, and
substrate binding in 4-hydroxyphenylpyruvate dioxygenase fiom
Pseudomonas sp. P.J .874. J. Biol. Chem. 257: 11922-11931.

MacDiarmid, C. W., and R. C. Gardner. 1992. Direct submission of

sequence to EMBL/GENBANK/DDBJ Data Banks.

Majamma, R. K., Gunzler, V., Hanauske-Abel, H. M., Myllyla, R., and K. I.

Kivirikko. 1986. Partial identity of the 2-oxoglutarate and ascorbate
binding sites of prolyl 4-hydroxylase. J. Biol. Chem. 261:7819-7823.

Marsh, E. N., Chang, M. D. T.,and Townsend,C.A. 1992. Two isozymes of
clavaminate synthase central to clavulanic acid formation: cloning
and sequencing of both genes from Streptomyces clavuligerus.

Biochemistry 31: 12648-12657.

68

Matsuda, J ., Okabe, S., Hashimoto, T., and Y. Yamada. 1991. Molecular
cloning of Hyoscyamine 6-beta Hydroxylase, a 2-oxoglutarate-
dependent dioxygenase, from cultured roots of Hyoscyamus niger. J.
Biol. Chem. 266: 9460-9464.

McGarvey, D.J., Yu, H., and R. E. Christoffersen. 1990. Nucleotide
sequence of a ripening-related cDNA from avocado. Plant Mol.
Physiol. 15: 165-167.

McGarvey, D. J. and R. E. Christofferson. 1992. Characterization and
kinetic parameters of ethylene-forming enzyme from avocado fruit. J.
Mol. Biol. 267: 5964-7.

Meldgaard, M. 1992. Expression of chalcone synthase, dihydroflavanol
reductase and flavanone-3-hydroxylase in mutants of barley deficient
in anthrocyanidin biosynthesis. Theor. Appl. Genet. 83: 6 95-706.

Menssen, A., Hoehmann, S., Martin, W., Schnable, P. S., Peterson P. S.,
Saedler, H., and A. Gierl. 1990. The En/Spm transposible element of
Zea mays contains a splce at the termini generating a novel intron
from a dSpm element in the A2 gene. EMBO J. 9: 3051-3056.

Milligan, S. B., and C. S. Gasser. 1995. Nature and regulation of pistil-

expressed genes in tomato. Plant Mol. Biol. 28(4): 691-71 1.

69

Myllyla, R., Majammaa, R. K., Gunzler, V., Hanauske-Abel, H. M., and K.

Kivirikko. Ascorbate is consumed stoichiometrically in the uncoupled
reaction catalyzed by prolyl 4-hydroxylase and lysyl hydroxylase. J.

Biol. Chem. 259:5403-5405.

Myllyla, R., Pihlajaniemi, T., Pajunen, L., Turpeenniemi-Hujamen, T., and

K. I. Kivirikko. 1991. Molecular cloning of chick lysyl hydroxylase.

J. Biol. Chem. 266(5): 2805-2810.

Myllyla, R., Gunzler, V., Kivirikko, K., and D. D. Kaska. 1992.

Modification of vertabrate prolyl 4-hydroxylases and lysyl
hydroxylase by diethyl pyrocarbonate. 286: 923-927.
Nadeau, J. A., Zhang, X., Nair, H.,and S. D. O'Neill. 1992. Submitted
directly to EMBL/GENBANK/DDBJ Data Banks.
Nakanishi, H., Okumura, N., Umehara, Y., Nishizawa, N. K., Chino, M. and
S. Mori. 1994. Expression of a gene specific for iron deficiency
(Ids3) in the roots of Hordeum vulgare. Plant Cell Physiol. 34:
401-410.
Ng, S., Hanauske-Abel, H. M., and S. Englard. 1991. Cosubstrate binding
site of Pseudomonas sp. AK-l gamma-butyrobetaine hydroxylase. J.

Biol. Chem. 266: 1526-1533.

70

Okumura, N., Nishizawa, N.K., Umehara, Y., Ohata, T., Nakanishi, H.,
Yamaguchi, T., Chino, M., and S. Mori. 1994. A dioxygenase gene
(IdsZ) expressed under iron deficiency conditions in the roots of
Hordeum vulgare.Plant Mol. Biol. 25: 705-719.

Pascal, R. A., Oliver, M. A., and Y. C. Chen. 1985. Alternate substrates and
inhibitors of bacterial 4-hydroxyphenylpyruvate dioxygenase.
Biochemistry 24: 3158-3165.

Peck, S. C., Olson, D. C., and H. Kende. 1992. Submitted directly to
EMBL/GENBANK/DDBJ Data Banks.

Phillips, A. L., Ward, D. A., Uknes, S., Appleford, N. E. J ., Lange, T.,
Huttly, A. K., Gaskin, P., Graebe, J. E., and P. Hedden. (unpublished).
Isolation and expression of three gibberellin 20-oxidase cDNA clones
from Arabidopsis.

Pihlajaniemi, T., Helaakoski, T., Tasanen, K., Myllyla, R., Huhtala, M. L.,
Koivu, J ., and K. I. Kivirikko. 1987. Molecular cloning of the beta-
subunit of human prolyl 4-hydroxylase. EMBO J. 6 (3): 643-649.

Prescott, A. G. 1993. A dilemma of dioxygenases (or where biochemistry
and molecular biology fail to meet). J. Exp. Bot. 44:849-861.

Pua, E. C., Sim, G. E., and M L. Chye. 1992. Isolation and sequence

7]

analysis of a cDNA clone encoding ethylene-forming enzyme in
Brassicajuncea (L.) Czern and Coss. Plant Mol. Biol. 19: 541-544.

Randall, C. R., Zang, Y., True, A. E., Que, L. Jr., Chamock, J. M., Garner,
D., Fujishima, Y., Schofield, C. J ., and J. E. Baldwin. 1993. X-ray
absorption studies of the ferrous active site of isopenicillin N
synthase and related model compounds. Biochemistry 32: 6664-6673.

Roach, P. L., Clifton, I. J ., Fulop, V., Harlos, K., Barton, G. L., Hajdu, J .,
Andersson, I., Schofield, C. J ., and J. E. Baldwin. 1995. Crystal
structure of isopenicillin N synthase is the forst from a new structural
family of enzymes. Nature 375: 700-704.

Ruetschi, U., Odelhoeg, B., Lindsteadt, S., Barros-Soederling, J ., Persson,
B., and H. J oemvall. 1992. Characterization of 4-hydroxyphenyl-
pyruvate dioxygenase. Primary structure of the Pseudomonas enzyme.
Eur. J. Biochem. 205: 459-466.

Ruetschi, U., Dellsen, A., Sahlin, P., Stenman, G., Rymo, L., and S.
Lindsteadt. 1993 a. Human 4-hydroxyphenylpyruvate dioxygenase:
primary structure and chromosomal localization of the gene. Eur. J.
Biochem. 213: 1081-1089.

Ruetschi, U., Nordin, I., Odelhoeg, B., Joemvall, H., and S. Lindsteadt.

72

1993. Gamma-butyrobetaine hydroxylase Structural characterization
of the Pseudomonas enzyme. Eur. J. Biochem. 213: 1075-1080.

Ross, G. S., Knighton, M. L., and M. Lay-Yee. 1992. The ethylene-related
cDNA from ripening apples. Plant Mol. Biol. 19: 231-238.

Samson, S. N., Belagaje, R., Blankenship, D. T., Chapman, J. L., Perry, D.,
Skatrud, P. L., VanFrank, R. M., Abraham, E. P., Baldwin, J. E.,
Queener, S. W., and T. D. Ingolia. 1985. Isolation, sequence
determination and expression in Escherichia coli of the isopenicillin
N synthetase gene from Cephalosporium acremonium. Nature 318:
191-194.

Samson, S. M., Dotzlaf, J. E., Slisz, M. L., Becker, G. W., Van Frank, R.
M., Veal, L. E., Yeh, W. K., Miller, J. R., Queener, S. W., and T. D.
Ingolia. 1987. BIO/TECHNOLOGY 5: 1207-1214.

Sato, S., Suzuki, H., Widyastuti, U., Hotta, Y., and S. Tabata. 1994.
Identification and characterization of genes induced during sexual
differentiation in Schizosaccharomyces pombe. Curr. Genet. 26: 31-
37.

Schofield, J. P., Vijayakumar, R. K., And D. B. G. Oliveira. 1991.

Sequences of the mouse F protein alleles and identification of a T-cell

epitope. Eur. J. Immunol. 21: 1235-1240.

73

Shiffrnan, D., Mevarech, M., Jensen, S.E., Cohen, G., and Y. Aharanowitz.
1988. Cloning and comparative sequence analysis of the gene coding
for isopenicillin N synthase in Streptomyces. Mol. Gen. Genet. 214:
562-569.

Shiffinan, D., Cohen, G., Aharonowitz, Y., Von Dohren, H., Kleinkauf, H.,
and M. Mevarech. 1990. Nucleotide sequence of isopenicillin N
synthase (pch) of the gram negative F lavobacterium sp. SC12,154.
Nucleic Acids Res. 18: 660-660.

Sparvoli, F., Martin, C., Sceinza, A., Gavazzi, G., and C. Tonelli. 1994.
Cloning and molecular analysis of structural genes involved in
flavanoid and stilbene biosynthesis in grape (Vitis vinifera L.). Plant
Mol, Biol. 24: 743-755.

Spanau, P., Reinhardt, D., and T. Bollert. 1991. Analysis and cloning of the
ethylene-forming enzyme from tomato by functional expression of its
mRNA in Xenopus laevis oocytes. ENIBO J. 10: 2007-2013.

Tan, D. S. H., and T. Sim. 1996. Functional analysis of conserved histidine
residues in Cephalosporium acremonium isopenicillin N synthase by
site-directed mutagenesis. J. Biol. Chem. 271: 889-894.

Tang, X., Wang, H., Brandt, A. S., and W. R. Woodson. 1993. Organization

74

and structure of the 1-aminocyclopropane-3-carboxylate dioxygenase
gene family from Petunia hybrida. Plant Mol. Biol. 23: 1151-1164.

Taio-Suan, S., and D. S. H. Tan 1994. Histidine-272 if isopenicillin N
synthase of Cephalosporium acremonium, which is possibly involved
in iron binding, is essential for catalytic activity. FEMS Microbiol.
Lett. 120: 241-248.

Trentmann, S. M., and H. Kende. 1994. Submitted to the EMBL Data
Library.

Toguri, T. 1994. Submitted to the EMBL Data Library.

Van der Ploeg, J. R., Weiss, M. A., Saller, A., Nashimoto, H., Saito, N.,
Kertesz, M. A., and T. Liesinger. 1996. Identification of sulfite
starvation-regulated genes in Escherichia coli: a gene cluster
involved in the utilization of taurine as a sulfur source. J. Bacteriol.
178: 5438-46.

Wang, H., and W. R. Woodson. 1991. Reversible inhibition of ethylene
action and interruption of petal senescence in carnation flowers by
nobomadiene. Plant Physiol. 96: 1000-1001 .

Weidler, H., and W. R. Woodson. 1995. Submitted to the EMBL Data

Library.

 

75

Weigel, B. J ., Burgett, S. G., Chen, V. J ., Skatrud, P. L., Frolik, C. A.,
Queener, S. W., and T. D. Ingolia. 1988. Cloning and expression in
Escherichia coli of isopenicillin N synthase from Streptomyces
lipmanii and Aspergillus nidulans. J. Bacteriol. 170: 381 7-3 826.

Wintermeyer, E., Flugel, M., Ott, M., Steinert, M., Rdest, U., Mann,K. H.,
and J. Hacker. 1994. Sequence determination and mutational analysis
of the lly locus of Legionella pneurnophila. Infect. Immun. 62(3):
1109-1117.

Wu, K., Li, L., Gage, D. A., and J. A. D. Zeevaart. (unpublished). Cloning
and photoperiod-regulated expression of gibberellin 20-oxidase from
the long-day plant spinach.

Wyckoff, E. E., Pishko, E. J ., Kirkland, T. N., and G. T. Cole. 1995. Cloning
and expression of a gene encoding a T-cell reactive protein from
Coccidioides immitis: homology to 4-hydroxyphenylpyruvate

dioxygenase and the mammalian F antigen.Gene 161: 107-111.

CHAPTER 2.

Use of 4-Nitrophenoxyacetic Acid for the Detection and
Quantification of 2,4-Dichlorophenoxyacetic Acid (2,4-D)Ia-
ketoglutarate Dioxygenase Activity in Natural and Engineered

Microorganisms

ABSTRACT

deA is the initial enzyme in the canonical 2,4-D degradation
pathway, one of the best studied of the chloro-substituted aromatic
degradation pathways in bacteria. Purified 2,4-D/0t-ketoglutarate
dioxygenase (deA) was shown to use 4-nitrophenoxyacetic acid (4-NPAA)
(Km = 0.89 i- 0.04 mM, kw, = 540 i 10 mini), producing intensely yellow 4-
nitrophenol. The generation of this intensely yellow chromophore from 4-
NPAA was used to develop a rapid, continuous, colorimetric assay for the
detection of deA and analogous activities in 2,4-D degrading bacterial
cells and extracts. The 4-NPAA assay was found to be suitable for large
scale colony screening and direct, quantitative activity measurements, and
these methods were shown to offer significant advantages over previous
approaches. A diverse collection of environmental 2,4-D degrading strains
was screened, revealing significant diversity among 2,4-D degrading

isolates.

76

 

77

INTRODUCTION

The canonical 2,4-dicholrophenoxyacetic acid (2,4-D) degradation
pathway is that of Alcaligenes eutrophus JMP134 (pJP4), which uses six
enzymes to convert 2,4-D degradation to TCA cycle intermediates (Don,
1985; Perkins, 1990; Streber, 1987). The first enzyme in the pathway, 2,4-
D/a-ketoglutarate dioxygenase (deA), converts 2,4-D, a-ketoglutarate, and
oxygen to 2,4-dichlorophenol, glyoxylate, carbon dioxide, and succinate
(Fukumori, 1993a). Analysis of deA-like activities in 2,4-D degrading
microorganisms requires the availability of a functional assay for the
enzyme. Several methods to assay deA-like activities have been described,
but all have limitations. l-IPLC (Fukumori, 1993a) and GC (Perkins, 1988)
methods can be used to monitor 2,4-D disappearance or 2,4-dichlorophenol
production, but these methods are time consuming and require sophisticated
equipment. Oxygen electrodes can be used for continuous assay of the
activity (Streber, 1987), but this approach is confounded by the presence of
other oxygenases and requires specialized equipment. Radioactive methods
to measure the release of 14C02 from labeled substrate (Fukumori, 1993b;
Fulthorpe, 1996) are known, but these approaches require special handling.
A discontinuous spectrophotometric assay for substituted phenols that uses
4-aminoantipyrene to form a highly colored complex has been used for the
in vitro detection of substituted phenols (Fukumori, 1993b; King, 1991),
and was used in a petri plate assay to detect cells with overexpressed or

deregulated deA systems (King, 1991). Unfortunately, the required high

78

pH is lethal to the cells, and reagent does not work well for regulated
pathways in the plate assay (King, 1991). Here, we evaluated 4-
nitrophenoxyacetic acid (4-NPAA, Tokyo Kasai, Tokyo, Japan) as a

sensitive, continuous spectrophotometric assay for deA.

79

RESULTS AND DISCUSSION

Using purified Alcaligenes eutrophus JMP134 deA and the kinetics
methods as previously reported (Fukumori, 1993b), 4-NPAA was shown to
be a substrate (K, = 0.89 i 0.04 mM, Itcat 540 i 10 min" , kw/Km = 610 mM'
I min'l). When compared to published values for the decomposition of 2,4-
D by this enzyme (K, = 17.5 i 1.0 uM, km, = 529 min'1 (Fukumori, 1993b)),
it is clear that the affinity for 4-NPAA is 50-fold less, the turnover number
is unchanged, and the catalytic efficiency is approximately 2.0% of that for
2,4-D. One product of 4-NPAA decomposition was demonstrated to be 4-
nitrophenol (4-NP) on the basis of UV-visible spectroscopy and HPLC
analysis using a Hewlett Packard Series 1050 HPLC unit with a Hibar
Lichrosorb RP-18 column and a 60% methanol and 40% 20 mM KH2PO4
buffer at pH 3.

We sought to use the intensely yellow (18.4 mIVTl cm") 4-nitrophenol
chromophore as a method to detect whole cell deA-like activities in a set
of control and engineered strains (Table 2.1) and in a collection of 2,4-D
degrading strains (Table 2.2) using a plate assay. Colonies were carefully
wetted with freshly prepared assay solution (10 mM 4-NPAA, 50 M
Fe(NI-I4)2(SO4)2, and 50 uM ascorbate in 20 mM Tris-HCl, pH 7.4), allowed

to dry, and rewetted using an atomizer. Plates were incubated at 30°C with

intermittent observation for 5 minutes to 6 hours. Inclusion of a-
ketoglutarate in the assay solution did not improve the effectiveness of the
assay with most strains. Colonies of A. eutrophus, Burkholderia cepacia

DB01 (pROlOl), and numerous environmental isolates capable of

80

Saba vmfi 758.. E 21
S 33898 Sam“ 83: 05 E 828.»— :oo mo ”Sauna 0288.3 ofi wEm: 5526qu A73. view—“53w 330508323583 9
.063 8mm. no £8 20:? 3265 O-V. N .3 <<m2é 28 m. fl Sow mZé who 53:on Bang
in: mm a 6.5 8:583 be... .3»: on 3 9.2V 28 23%: .3 s
2: s 88 sea Ase gasses. Ea fleas assuage or: 21 2: as, u: a 85% 62 A82 3.8.5 53 b .62
358$ muse 05.530 02m SEES e\eN.. o 5;» .ANE 38.8: D. v. N 28 m. _ 55> wouaofioamsm 83 AD 82on £3 BEE—E e
Rafi museum 2 m 98 .32 nudism 2 I .32 don a Q .Ntfl .Safinomm a me
.Amv cogcflwov D-v.N com macaw 555. 05 co Soto o: 55, .005 325.. Sud mm 5:53 2883 .8882.”
2833 05 80¢ commands on 8 :owficoto 82.80 05 E 83 nfimvmmmommém E 89.35 08% 05 wfifificoo :mmESE
53> £83233 2: mo 86 Hefihoem. 05 0:: euros: fiSoEoE 2833 05 8 86250 nowficoflo 5 Ev Ewan mo Eofiwfim
358... be 5.3 6: 08958822.. e o a 82:3 2835 as be... be as Baas seem mess 38:25 babes
es. 329:8 e es. 8% see Eocene s masses .3 E3 co assess ca. .3 3 e as, 6 28% a 5% Bee: .

 

 

e _ 2 a M: - ebaeaxabb use \ 62 as; was pea :mmzze ma :5 :8 ensuresam

- - Basses: pews \ 62 283 we: pea 282% NS :6 :8 assesses

a S e. 3 - .mee fies \ 82 m new .205 :53 :m Eu :8 ensetsseem

- - NOE m econ S _ m0 :8 uiotefinm

23 H 2 + 3% .3 \ N: m 85% 8mm 388 eessesefism

- - NOE I 28: 5mm 3.33..» atesofitam

- .. La: \ N02 D 28s HwNva/c. engages» Snowmen;

8:; u. go + Sees \ NE a E3 42%,: senses assess?

083.: semi 25050 Ego—om
:omEQEoo «Jaw/7v \ 332 5380 booaocomom «ESQ—m Baum. “mom

 

 

.mzamma fimztv 05 3 34593.. :2: was 2:36 19.32:»: was .2230 fiN wish.

81

 

 

 

 

- - - H OZ =92 2 .FEH
- - QZ H DZ 92 wNQmH
- 30? DZ. H DZ Q2 we .2 .Nm .NfiamH
x83 32... 92 H 32 . Qz ow .29?
- - - m 2 .9 aaosogfim 3-3 .3m
.+ + .. m 3 dm 3§EsM=EQm $3M
- + .. m 2 dm manofiewzimm vamH
- 32. - m 2 d. auzssomsfiw $52m
.+ - - m ~ g Emwckwofiigfiugohmz :2
+ + + H.,a 3 dm stqufigzm Umé
+ + + H a dm utmEofixzm cm .w .H ....m .v .NQHH
.Vm.HN
..mN ..NN . _N
. .ON .3 .2 mm:
+ + + H H ENS: atysofiém. .2 .3 ..m_ .chH
+ + + m o Eugzmfixfixuxmxxonoé aém
+ + + H m azxmshzm umnmmagux 3. .mm wan“:
x83 Bo? + H a dm SbEefiSm memH
+ + + H _ .9. c.2§§.£ % 52m
+ + + Q _ . 2:333» mmzmmacfiw 3” ES:
083mb cows—n 9.on
o noumNBflnE :8qu
56.858 Ens/3. 3% 985.30% coufimvtam coumomuaog 863m £83m

 

 

.Gé.N «Snug «a5 moan—cam .355: no «em 8.83m. a .3 Q73. 3 2.3.82.3 jmzuv N.N 039—.

82

.8888 888: 8:0 888G .:

888888860 888:: 888G m

.8883 :8 .8 888: 8:8: 8 888:: 3838: 888 8 8B 888:8 gnu/3.: :

883.380 830% m88> 83:: 3888: 883 .<<mz-v 88:8 8: 3:3 833

8883 858: vm 8335 888: 0: 88:3 3 .88: 8:88 :88 88883 D8 8388: 88:3 9826

.8888 Wm :38» 883:8 883388 88:3 A+v .8888 8.58: w:o8 E335 88> 88o 85

..<3.C. .8 88:38 8588:: 8.: 803380 8308 888m .380: 88388: 88m 888m 83 wfim: 8:88 :8
3336:: 83 83 883850 .888: :8 :m 888:8 <<mZ-v 8.: 388: 883 O-V.N 3383 8:3 888m 0

.88: 3N :88 888: 88883 0: 88:3

3 .88: 3 8.335 38083 :28 88:3 @523 .88; m 8535 8883 :28 88:3 A+v .<<mz-v 2:: md 38

O-V.N 2:: md .8333: 88:8 88888 85:8 35:: 5 @773 8 <<mZé 8 888:8 28883 3.85 8

388883 8: 88:3 DZ 8:83:83 82:8: 88:3

At .GV 88:: a 8 Sub 888:8 88.88? 8 mew: 88:38.: 8:888 32 :38 8883—83 :888m o
.33 88H m: H 38 .39 8883..“ mm m 39. 8Q m: G .838 888.8% a

.983 85V 888 «3.92.3 83 8 880:8: 8 8

at 8:88: mmd 8 3888 83 :8888 ._. 8 88 388:: 8388 8m .va 3383: 88A 8 :0883833 83338

8 8m 802380 835 R88:: 88w DR: 38 .Sub .38 86: :88: 388 8883833 8 3888.83 b.8888

88> 388.83 883 £8888 8853::an 29:88 8888—8: 8887 8288 Z 38 80858.: 883338

8 88:88 88888 8 x83 £8 E 3388 83:3: :2 :88 535 A3 80858.: MZ<mZmO 8 88888
8888 <72: m3: 838: E 880 >88 E 88883 3888:: 89$ 8082.883: 88on 8885 a

38.888 Nd 88,—.

83

degrading 2,4-D were found to turn yellow and produced a yellow halo of
diffusing 4-NP around the colony (Fig. 2.1). The halo was particularly
useful for identification of activity with pigmented colonies. Importantly,
however, not all 2,4-D degraders presented a yellow halo and selected
engineered strains known to express fldA remained as colorless as the
controls lacking fdA expression (Table 2.1). Similar results were obtained
in batch culture or microtiter plate assays, where cultures were spiked with

1.5 mM 4-NPAA when the cultures were in late log to early stationary phase

and incubated for 5 minutes to six hours at 30°C. Samples of the cultures, or

entire microtiter plates, were centrifuged to pellet the cells, supernatants
were transferred to cuvettes or fresh microtiter plates where 1/10 volumes of
0.1 N NaOH(f1nal pH 1 1) were added, and the absorbances at 401 nm were
measured. As with the plate assay, not all of the cells known to express tfdA
or degrade 2,4-D produced a yellow color.

The lack of color production in selected whole cells is likely due to
the inability of these cultures to transport 4-NPAA through the cell
envelope. For example, when E. coli CBlS33 or E. coli CB1811 cells
(negative in whole cell assays, yet known to express zfdA under the tar:
promoter) were sonicated, centrifuged for 30 minutes at 100,000 x g, and
assayed spectrophotometrically at 401 nm, the production of 4-NP was
observed (Table 2.1). Lysates of Alcaligenes eutrophus JMP134 that were
treated similarly also converted 4-NPAA to 4-NP (Figure 2.2), and
converted this substrate in a quantitative manner (Figure 2.3). Lysate

reaction mixtures include all of the known co-factors of deA - a final

concentration of 1 mM a-ketoglutarate, 50 M ascorbate, 50 M

84

 

Eigure 2.1. The 4-NPAA petri plate assay. Panels on the left are color
photos of assay plates after treatment and incubation. The right panes denote
the strains depicted in the photos. The top pair of panels shows
environmental strains that have different levels of growth and different rates
of 4-NP production after an incubation time of four hours. The strains shown
in the bottom pair of panels are engineered strains (positive control
Burkholderia cepacia DB01 containing pROlOl) and environmental isolates
(numbers indicate TF D strain designations). The negative controls, E. coli
C310] and Burkholderia cepacia DB01, are in the bottom row. Incubation
time was six hours.

 

85

1.4

 

 

 

 

 

-____ ”M7
1.2 - —RASC ;
1 ..
0.8 0
E
0.6 «»
0.4 ~»
0.2 ‘-
0 Ti J; T
0 1 2 3 4 5 6 7 a 9 10

Time (min)

Figure 2.2 4-NPAA cell lysate assay using Alcaligenes eutrophus
JMP134 and Burkholderia cepacia RASC. Continuous assay of 4-
nitrophenol produced by deA was measured spectrophotometrically by
absorbance at 401 nm. deA fiom Alcalz'genes and deArasc from
Burkholderia are 91% identical at the amino acid level.

86

 

07 ‘ * ‘ 1 A 4 1
100ul
055‘ 80 on
r
“21 60 ul .

 

A401

 

 

 

 

 

 

  

 

 

 

 

Time (min)
0.4
8. 0.3 —
£2 CLZEf
"’ 0.1 -
0 i 4 r

C) CD CD CD 'CD CD

0“ <r GD 00 CD

volume of lysate ‘-

 

 

 

Efigure 2.3. 4-NPAA cell lysate assay using Alcaligenes eutrophus
JMP134 indicating 4-NPAA assay is quantitative. Top figure: continuous
assay of 4-nitrophenol produced by deA was measured
spectrophotometrically by absorbance at 401 nm. Bottom figure shows the

linearity of the increase of 4-NP production over time versus the amount of
enzyme used in the assay.

 

87

Fe(NH4)2(SO4)2 - and 1.5 mM 4-NPAA in 20 mM Tris-HCl, pH 7.4.
Reaction mixtures varied in protein content, and the resulting 4-NP
production was normalized to cell lysate protein content using Lowry assay
determination (Lowry, 1951). Inclusion of a 0.1 volume 0.1 N NaOH
quench step in the assay increased its sensitivity by converting all of the
released 4-NP to its deprotonated form (pKa = 7.15); however, the assay can
not be used in a continuous manner at high pH values (Fukumori, 1993b). F
Among the six environmental isolates that were negative in whole cell
assays, only one, Nitrobacter winogradskii strain M1 (F ulthorpe, 1995), was
positive in the lysate assay (Table 2.2). Of note, much of the activity in this

 

IF-

strain was found in the particulate fraction. The lack of activity in whole
cells and presence of activity in lysates is taken as evidence of a transport
deficiency. Attempts to permeabilize E. coli cells using chloroform or
toluene vapor, or by treatment with Triton X-100 at various concentrations
in conjunction with this assay, were ineffective or completely eliminated
deA activity.

4-NPAA can also be used effectively with many isolates to assay for
deA-like activity in batch cultures. Cultures of 2,4-D degrading isolates
can be tested for the induction of fdA by including 4-NPAA in the media,
using 2,4-D as the grth substrate (Appendix B). Since induction of the
2,4-D pathway varies widely among isolates, testing a sample of a culture
for induction by adding 100 mM 4-NPAA is an excellent way to ensure the
presence of pathway enzymes in cell lysates. Spiking a sample of the large
culture is advisable, since the presence of 4-NPAA or 4-NP in the media has
widely varying effects on the grth of different species. This technique

also allows the effective use of supplementary carbon sources to increase

88

the cell mass of strains which do not grow to sufficient cell density for cell
lysate preparation. Casamino acids were the best general carbon
supplementation, with sugars and yeast extracts having an inhibitory effect
on the induction of the 2,4-D pathway of several strains including
Alcaligenes eutrophus JMP134. 4-NPAA was not a gratuitous inducer of the
2,4—D pathway in any of the strains tested, and none of the strains in this
study could utilize 4-NP as a carbon source (Appendix B).

The 4-NPAA whole cell and cell lysate assay results for the
environmental isolates demonstrate that caution must be used in making
interpretations based on this substrate. Five 2,4-D degrading isolates (B6-5,
B6-7 and TFD], TFDl 1, and TFD28 (F ulthorpe, 1995; Tonso, 1995)) were
negative for 4-NPAA conversion both as whole cells and in extracts (Table
2). These results are consistent with the presence of a 2,4-D degrading
enzyme that fails to recognize 4-NPAA or fails to convert 4-NPAA under
conditions used in this assay. Lysates of these strains also failed to convert
2,4-D to 2,4-dichlorophenol. Several other strains (EML146, and TFD12,
TF D32, TFD35, TFD44, and TFD45 (F ulthorpe, 1995; Tonso, 1995)) were
able to degrade 4-NPAA as whole cells, but exhibited no activity in extracts.
This outcome could suggest the presence of a highly labile enzyme or one
that metabolizes 2,4-D by an alternate mechanism not requiring 0L-
ketoglutarate. Nevertheless, for most environmental isolates, 4-NPAA can
serve as a useful substrate to allow the detection and continuous
spectrophotometric assay of deA activity in whole cell enriched fractions

or cell lysates.

89

SUMMARY

Purified 2,4-D/0t-ketoglutarate dioxygenase (deA) was shown
to use 4-NPAA at about 0.2% the rate of 2,4-D, producing intensely yellow
4-nitrophenol. This reagent was used to develop a rapid, continuous,
colorimetric assay for the detection of deA and analogous activities in 2,4-
D degrading bacterial cells and extracts. The assay was developed for whole
cells in liquid, cell lysates, and petri plates, can be used in a qualitative or
quantitative manner, and may be used in a continuous manner unlike many
of the other assays. This study indicates that there is a wide diversity among
environmental isolates in 4-NPAA conversion. All strains that hybridize to a
fdA probe convert 4-NPAA to 4-NP readily, though the assay can also
detect deA-like activity in strains that do not hybridize to tfdA. None of the
strains tested can utilize 4-NPAA in the absence of 2,4-D, nor can they
utilize 4-NP as a carbon source in the presence or absence of 2,4-D. Growth
in a media that contains both 2,4-D and 4-NPAA has widely varying
deleterious effects on the growth of various environmental strains. The

assay also provides indication that there is a wide diversity of ability to take

up 2,4-D and 4-NPAA among 2,4-D degrading environmental isolates.

 

90

REFERENCES

Bachmann, B. J. 1972. Pedigrees of some mutant strains of Escherichia coli
K12. Bacteriol. Rev. 36:525-557.

Don, R.H., A.J. Weightman, H.J. Knackmuss, and K.N. Timmis. 1985.
Transposon mutagenesis and cloning analysis of the pathways for
degradation of 2,4-dichlorophenoxyacetic acid and 3-chlorobenzoate
in Alcaligenes eutrophus JMPl34(pJP4). J. Bacteriol. 161:85-90.

Frey, J ., M. Bagdasarian, and K. N. Timmis. 1984. Escherichia coli cloning

 

vectors that auto-amplify in the stationary phase of bacterial growth, -.
p.3 77-3 83. In S. Mitsuhashi and V. Kréméry (ed.), Transferable

Antibiotic Resistance: Plasmids and Gene Manipulation. Avericum
Praha and Springer, Heidelberg.

Fukumori, F., and RP. Hausinger. 1993a. Alcaligenes eutrophus JMP134
“2,4-dichlorophenoxyacetic acid monooxygenase” is an on-
ketoglutarate-dependent dioxygenase. J. Bacteriol. 175:2083-2086.

Fukumori, F ., and R. P. Hausinger. 1993b. Purification and characterization
of 2,4-dichlorophenoxyacetate/ a-ketoglutarate dioxygenase. J. Biol.
Chem. 268:2431 1-24317.

Fulthorpe, R.R., C. McGowan, O.V. Maltseva, W.E. Holben, and J .M.
Tiedje. 1995. 2,4-dichlorophenoxyacetic acid-degrading bacteria
contain mosaics of catabolic genes. Appl. Environ. Microbiol. 61:
3274-3281.

F ulthorpe, R.R., A. N. Rhodes, and J. M. Tiedje. 1996. Pristine soils

mineralize 3-chlorobenzoate and 2,4-dichlorophenoxyacetate via

91

different microbial populations. Appl. Environ. Microbiol. 62:1159-
1166.

Harker, A. R., R. H. Olsen, and R. J. Seidler. 1989. Phenoxyacetic acid
degradation by the 2,4-dichlorophenoxyacetic acid (TFD) pathway of
plasmid pJP4: mapping and characterization of the TFD regulatory
gene, rde. 171:314-320.

King, R. J ., K. A. Short, and R. J. Seidler. 1991. Assay for detection and r
enumeration of genetically engineered microorganisms which is
based on the activity of a deregulated 2,4-dichlorophenoxyacetate
monooxygenase. Appl. Environ. Microbiol. 57: 1790-1792.

Lowry, O. H., N. J. Rosebrough, A. L. Farr, and R. J. Randall. 1951. '-

 

Protein measurement with the Folin phenol reagent. J. Biol. Chem.
193: 265-275.

Overbye Michel, L. M. Sandkvist, and M. Bagdasarian. 1995. Specificity of
the protein secretory apparatus: secretion of the heat-labile
enterotoxin B subunit pentamers by different species of Gram'
bacteria. 152141-45.

Perkins, E.J., M.P. Gordon, O. Caceres, and RF. Lurquin. 1990.
Organization and sequence analysis of the 2,4-dichlorophenol
hydroxylase and dichlorocatechol oxidative operons of plasmid pJP4.
J. Bacteriol. 172:2351-2359.

Perkins, E. J ., and P. F. Lurquin. 1988. Duplication of a 2,4-
dichlorophenoxyacetic acid monooxygenase gene in Alcaligenes
eutrophus JMP 1 34(pJP4). J. Bacteriol. 170:5669-5 672.

Streber, W.R., K.N. Timmis, and M.H. Zenk. 1987. Analysis, cloning, and

high level expression of 2,4-dichlorophenoxyacetic acid

92

monooxygenase gene tfdA of Alcaligenes eutrophus M134. J.
Bacteriol. 169:2950—2955.

Tonso, N.L., V.G. Matheson, and WE. Holben. 1995. Polyphasic
characterization of a suite of bacterial isolates capable of degrading

2,4-D. Microbial Ecol. 30:3-24.

Chapter 3:

2,4-D Degrading Strain Characterization and the Conjugal
Capture of a Plasmid from Nitrobacter winogradskii M1 That
Rescues 2,4-Dichlorophenoxyacetic Acid Degradation in a deA'

Alcaligenes eutrophus Strain.

INTRODUCTION

As part of the effort to explore the evolution of catabolic pathways by
closely examining 2,4—D degradation, a phylogenetically diverse collection
of Gram-negative isolates that could degrade 2,4-D as a sole source of
carbon and energy was examined. This work focused on ascertaining
whether the collection contained alternate degradation pathways or enzymes
to the canonical 2,4-D degradation pathway of pJP4, examining the
diversity of genes among the pathways found, and perhaps identifying non-
homologous isofunctional enzymes. Since chlorocatechol cleavage
pathways are common in soil bacteria (F ulthorpe, 1996) and phenol
hydroxylases are also fairly common, the search for a non-homologous
isofunctional enzyme became restricted to the initial enzymatic reaction of
2,4-D degradation. deA, the initial enzyme of the canonical pathway from
Alcaligenes eutrophus JMP134 (pJP4), was reclassified as 2,4-D/0t-
ketoglutarate dioxygenase, rather than a monooxygenase as originally
reported (Fukumori, 1993). This enzyme converts 2,4-D and oc-ketoglutarate

to 2,4—dichlorophenol, glyoxylate, carbon dioxide, and succinate (Fukumori,

93

 

94

1993). In nature, (fdA is thought to be comparatively rare (F ulthorpe, 1996),
and has been found on plasmids independent from the other genes of the
2,4-D pathway (Top, 1996).

Extensive strain characterization at the genomic and biochemical
level - including FAME and BIOLOG analysis (Tonso, 1995), REP pattern,
hybridization to probes to the rfdA, lde, and 2:de genes (Fulthorpe, 1995),
and tentative identification based on partial l6S rDNA analysis (McGowan,
1994) - was performed by a number of workers (F ulthorpe, 1996; Suwa,
1996). Some of the 2,4-D degrading isolates failed to hybridize with a zfdA
specific probe (Table 2.1 and 2.2) and failed to amplify during PCR using
tfdA specific primers, indicating that there may be significant diversity
among the genes that have not been examined. Four a-proteobacterial
strains showed the least similarity to the canonical pathway and were
studied in hopes of finding at least one strain that uses an evolutionarily
distinct deA-like enzyme. Preliminary work that suggests that among the
four strains selected, there is at least one strain, Nitrobacter winogradskii
M1, that carries a plasmid borne deA-like enzyme that is functionally
similar but has distinct biochemical properties from the well studied deA
from Alcaligenes eutrophus JMP134. Further work will be required to

determine if this enzyme is evolutionarily unrelated to deA.

95

RESULTS AND DISCUSSION:

General comments. The strains and plasmids used in this chapter are
found in Table 3.1. Detailed methods are provided in Appendix C.

Phylogenetic identification of the a-proteobacterial strains. Most
of the strains in the diverse collection of 2,4-D degrading strains compiled
for study by the NSF Center for Microbial Ecologyare members of the B-
proteobacterial subdivision (McGowan, 1995) and many contain genes that
are highly homologous to zfdA (F ulthorpe, 1995; Tonso, 1995). None of the
four isolates that were classified as a-proteobacteria hybridized with the
(fdA probe, making them candidates for possible alternate deA-like
enzymes. Three of the strains (TFD44, EMLl46, K1443) were identified as
Sphingomonas species. The genus Sphingomonas is a recent reclassification
division of the a—proteobacteria that may still be polyphyletic (Takeuchi,
1994), a concept supported by the variable characteristics of the three
strains in this study (Table 3.2 and 3.3). The initial identification of the
fourth strain, M1, indicated it was Rhodopseudomonas palustris, but was
identified with the qualification that it was atypically non-pigmented and
therefore may not be correct (McGowan, 1995). A subsequent examination
of the partial 16S rDNA sequence of strain Ml using multiple sequence
alignment and neighbor-joining (NJ) phylogenetic analysis resulted in the
identification of M1 as Nitrobacter winogradskii (Figure 3.1). M1 clustered
with a bootstrap of 84 in a clade containing N. winogradskii, N.
hamburgensis, Bradyrhizobium japonicum, Blastobacter denitrzficans, and
R. palustris. Within that clade, Nitrobacter winogradskii and strain M1

clustered with a bootstrap value of 94, indicating a clustering with relative

96

 

 

32 $9.333 m: 3.” 560 .25: Sam :8 .m

32 .swfianwam mg a? $30 .88 Sam :8 .m

«5 .saefimm m: 88 82mm :8 .m

89a 5 .coaou . m3 _ 088 «$3133 .28: Emma :8 .m

32 .ssfimsm am + m: 93% meg 282% 550 :8 fiskfifi

£2 .28 m: be: cause 03.95 .% séssmsfim

32 .889 N: has .0323: 3.93. .& manosomfimm

a2 .Q N: nab .0332: $3M .% auzosomsfim

82 .Eaanwam mg 25: . «3.2.x mesa m

53 ..m .xmkozom mu one: omommm 0.83%:me .&

82 .2553 «2 32a Améwué 22a 3 issues: senescfi

32 .835“: m2 rat .A+§.8 REE 8E 58% .m

82 .925 a: be: 0%: SSH 3038 .m

32 £83 N2 Cs» up: 8 5% 8mm 38% .m

mwfl Gav—3M mu econ Ema Sesame SSEofiSm

use E: m: hat 2% Ba 83% E .. E senses .w

.52 dog x2 + 5 p35 88 Hwang,: senses .V

59 dog N: be: E3 42%,: senses .V

32 d8. E + 5 n5 .-<ab 82:3 :53 sighs ”3&8?
vacuumed .332 8.380 «03050 €963ngme Embm

 

.mmEzuom ax__-<bb _o>o: 33:32: 2 new: mEEmmE can «525 .9.» 03¢...

97

t...‘ .IIIpl‘":II-.1L

._E\w1 mm 8 08:98:88 28 4:31.03: 3280;888:020 48%: on an Eon 08638: 48%: 2: “a com:

98 593893.55 ea: Eomfimgm .8288 5803830538 05 E 395088 08 582830598 8602 a
.moumtouofimno E85 .8 380508030 Boa mosmtflogmno 2:583
588.835 «888 < 5505883. 5 525588.85 EEQE 3 326:0: .55 B5: 0.8 528583 a

 

 

82 88838 :5 28828888858 .8: : SS5 :8 m
82 58838 m: 58858888882: .88 : 235 :8 .m
52 .885 88:8 + 5 58:28.55: 8: :5 :8 .m
52 .885 8588 + mg ”88:88:58 £8 :5 :8 .m
52 .885 88> + m: 58:88:58 we: :5 :8 .m
.52 .885 88> + m: 8:83:58 £8 :5 :8 .m
52 8888885 o... + m: not 85:8 .88 SE: :8 .m
82 .5888: m: .85 58553888 .88 SE: :8 .m
82 .8880 85 + m: 858828 .88 Sam :8 .m
82 .8882 8o + m: 88 88228 .88 Sam :8 cm
vacuumed 88:32 5380 8093280 Em>2o~<EEmmE 885m

 

 

88588 .3 288

98

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

3.3 + 3.8 + 3-3 + - 9-3 + - .593”
- - - - - - <<mZé
Q5 + 93 + 9.5 + - 9-8 + - O-V.N
3 + - - - 3 + - «£80-:
98 + - - - - - 5.80-:
A3 + - - - - - Eon ofioomzxoqonm
- - - - + - 38058000802048
@ + - - - 3 32m - oumoNconosioé
- - - - - - ewmoNconoeoEo-N

+ - - + + + 28 ow 8883
9c + - - 32m + + $3.: 40.8855
:5 + - - .85 + + .88.: .8885
:5 + 3 >85 3 >85 85 3 + + .x: .8882
3 + 8.. 8-8+ 85 8.8.. + 88888
o5 + - - + - + 8886
C5 + + - + g + + Baafifiwopoxé
A3 + 325 - - .3 + 8+ 888.:
3808:. :2 93M 333: 3:35 5mm «venom 80980

 

 

 

 

 

 

 

acts—Eran:— 0=o=ou .Su can: 558.55

3.3.58. can 538.55 “Ewing“. gin _a_._30§_080.a-n .58 :0 .5555: 3.505 non—.30 N6 2an

 

99

 

26:3 33%. 2 £33 .3509 as $58 330% 2 £23 63020 385 0228
60000 0.8 mammbm 05 we :0 .chtométsq 3 \S 02008 x2908 003:0 :O 02:28 30:0»

Ab EB £30200 833 C3 £03200 amass 3v £08200 .830 8 “$530 0 00000003180023

0:008 00 womb 0008020 :0 @200 05 m0 0200 E 0w§20 m 80: $850080 E 88004 0
.0 com 00 00:09:05 .80: mm a 000mm 3 550% mo 000030 0:0 .mA o: fl

:0 005 :05 $2 bannmmswcummw At m4 2 2 00 005 8 8:86 00 0020000 003 53000 D
00.88 00980 080: 0% mo 28 ~ :33 0080820900 am“ 030:

$0; 0:0 303 690535 0:3 30:58 33 0003 0:008 05 .0800 830050 00015 a

 

+ + + + + + _ 0200 0580000 _

 

 

 

 

 

 

- - - - - - _ n2.:

 

 

680.80 .3 2...?

 

 

100

Figure 3.1. Phylogenetic tree representing the neighbor-joining analysis
of the 168 rDNA sequence of strain M1 and close relatives. This analysis
supports the identification of M1 as Nitrobacter winogradskii (bootstrap =
94). This analysis was completed using MEGA (Kamar, 1993), with the
sequence identification and alignment as noted in Figure 3.3. Numbers at
the nodes of the tree represent bootstrap values using Tamura—Nei distance
estimation, complete removal of identical sites, and 500 replications.
Sequence abbreviations are as follows: ml- strain M1; nitrob w - N.
winogradskii; nitrob h - N. hamburgensis; blasb denitrif - Blastobacter
denitrificans; rps pal - Rhodopseudomonas palustris; brhiz jap -
Bradyrhizobiumjaponicum; rps acid - R. acidophila; rps sp2 and rps spS -
Rhodopseudomonas sp. SP2 and SP3; methb extorq - Methylobacterium
extorquens; azor caulinod - Azorhizobium caulinodans; rps viridis - R.
viridis; rps marina- R. marina; rps sp3and rps sp4- Rhodopseudomonas sp.
SP3 and SP4; rbacter cap- Rhodobacter capsulatus; rbacterl- Rhodobacter
euryhalinus; rps blastica- R. blastica; agrb tum- Agrobacterium
tumefaciens; ps dim- Pseudomonas diminuta; rspir rubrum -
Rhodospirillum rubrum; rdphila globif - Rhodophila globiformis; and e coli
- Escherichia coli.

101

44

84

94

nitrob

 

46

 

 

 

 

35

 

 
  
 

ml

w

nitrob h

blasb denitrif
rps pal

- brhiz jap

 

10

 

 

 

07

 

 

rps acid

rps ap2

99L rps spS

 

 

20

 

 

32

rps viridis

 

 

methb extorq

azor caulinod

rps marina

loor rps sp3

 

 

63

 

31

 

47

 

L rps sp4

rbacter cap

 

 

 

 

 

52

92

 

99

 

 

 

agrb tum

 

 

 

 

 

 

[ rbacter l
100 rap blastica

pS dimun

rspir rubrum

rdphila globif

e coli

 

Scale: each — is approximately equal to the distance of 0.002205

 

102

confidence that clearly distinguishes between the closely related sequences.
The NJ tree was supported by subsequent testing using the same alignment
and maximum parsimony (MP) analysis (Felsenstein, 1981). MP analysis
produced 10 most parsimonius trees, all of which supported the clustering
of M1 and N. winogradskii.

Strain characterizations. Whole cell assays. The a-proteobacterial
2,4-D-degrading strains were tested for growth on various carbon sources:
acetate, a-ketoglutarate, citrate, glucose, maltose, sucrose, glycerol,
benzoate, 2-chlorobenzoate, 3-chlorobenzoate, 4-chlorobenzoate,
phenoxyacetic acid (PAA), 2-chlorophenoxyacetic acid (2-CPAA), 4-
chlorophenoxyacetic acid (4-CPAA), 2,4-D, 4-NPAA, 2,4-DCP, 4-NP, and
casamino acids (Table 3.2). All of the strains were able to utilize casamino
acids as a growth substrate and grew well in diluted complex media (1/ 10
LB). Most were capable of utilizing TCA cycle intermediates as carbon
sources. Growth on the other carbon sources varied. Among the on-
proteobacteria, strain M1 and strain K1443 had the most similarity in carbon
source utilization, particularly noting their inability to degrade all
phenoxyacetic acid compounds other than 2,4-D. The other two strains of
this group, TFD44 and EMLl46, have a distinct pattern from one another as
well as from MI and K1443. TFD44 was unique in its ability to utilize all
chlorophenoxy acids tested, suggesting that the deA-like enzyme in these
strains may be distinct from one another and from deA.

Cell lysate assays. Assays using cell lysates of the oc-proteobacterial
2,4-D degrading strains indicate that all of the strains possess 2,4-
dichlorophenol hydroxylase (data not shown, Table 3.3; T.Sassanella and H.

Takemi, personal observation) and 3,5-dichlorocatechol dioxygenase

103

 

 

 

 

 

 

 

 

 

 

 

 

 

00 > 00 > 030533 00 > 0000003 .00 0000005
00 > 00 > 00> 00 > 0.000003 00 ©3500 005-003.
02 oz oz 02 “0% a 002500
00 > 00 > 00 > 00 > m000.09: E 093000 85-90:.
00003 .00> x003 .00> x003 .00> x003 .00> mmwb 8 302033
00050000 .00> £00 .00> 0 Z #00008 .00> 000003 00 53:00 0x=-<U.C.
oz 02 oz oz Name a 002500
00050000 .00> £00 .00> 02 80030 .00> A8096 0200000000 {1052-0
*02 00 > 32m .00> 00 > 900 2055 020002000 «3052-0
32m 00 > 32m 00 > O-V.N 00 53000
008300.02 00:050M503m. 00:050M030W 00:050M239w. 05000
:2 mg in 00 H 435 350:. .

 

 

 

 

 

 

.302 682203 00000000 b08380 0003 003000 05 .00 0080 00.0 00000w>x0m0 6000000000306

-m.m 00.“ 00000 05. .0 000000“? 00 00010000 000 000 :00de 8023 0:3 020000.800 E 0000 0003 30600
EEC 00030800? ~000fi€030€é§ 00m 0%000< .33 cabana—Ha 3 00808000 00>» 0000530033 050.50
wit-“$00 O-V.N EESoSoSoEé .58 .«0 0030030000000 @3300 000E: .000 guafictnhm g

104

activities (F ulthorpe, 1995). Additionally, all strains that hybridize to the
tfdA probe (as well as some that do not hybridize) possess activity when
using 4-nitrophenoxyacetic acid (4-NPAA) in a chromogenic assay for
deA (Sassanella, 1997; summarized in Table 2.2). Strains that do not
convert 4-NPAA in cell lysates also lost their deA-like activity, but
retained their 2,4-dichlorophenol hydroxylase and 3,5-dichlorocatechol
dioxygenase activities (F ulthorpe, 1995; Sassanella, 1997). These studies
indicate that all the strains in the collection follow an ortho-cleavage
pathway similar to that found on the catabolic plasmid pJP4, though the
hybridization studies indicate a significant diversity among individual genes
in different pathways (i.e - isofunctional genes of pathways from different
isolates vary in their level of identity to the probe, forming a mosaic pattern
of genes with differing relatedness to the canonical pathway). Further, some
of the genes were not detectable by hybridization in a number of strains,
though their enzymatic activity was present (Table 2.2 and 3.3). These
results indicate that there may be non-homologous isofunctional genes
involved in the degradation pathway of some of the collection strains.

The lysates of these strains were also tested in a series of reactions to
determine if the a-proteobacterial deA-like enzymes used the same co-
factors as deA. All lysates exhibited a requirement for additional iron, but
neither M1 nor K1443 required a-ketoglutarate, and M1 did not seem to
require ascorbate (Figure 3.2).

Enzymatic activity partitioning. The alternate substrate 4-NPAA was
used in place of 2,4-D in many of the subsequent reactions because 4-NP is
not utilized by these strains and accumulates (Sassanella, 1997). Though the
deA-like activity in fractionated lysates of TFD44 are similar to JMP134 in

105

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

0.2500
IALL -
I No '
0.2000 Ascorbate
I No Fe+2
D No AKG
0.1500 ~~—~ _~,
3
I
8
‘a
3 0.1000 ~-~ .
fl
.3
'5
.8
<
0.0500 .0— h a
0.0000 .
O O X In
(D m -‘ I
9 v o
r s 5 §
00500 3
Strain

Figure 3.2. Effect of known deA co-factors on the conversion of 4-
NPAA to 4-NP by the crude lysates of 2,4-D degrading bacteria. The
production of 4-NP by the cytosolic fraction (DB01, DB01/101, BHSOl) or
pellet fraction (Pl , K1443) of crude lysates was measured
spectrophotometrically after a 20 min incubation. Co-factor dependence was
determined by including all co-factors in the initial reaction (f.c. — 1 mM
Tris (pH 6.8), 1.0 mM 4-NPAA, 50 0M Fe”, 1 mM ascorbate, 1 mM a-
ketoglutarate), then removing a single co-factor (on-ketoglutarate, ascorbate,
and iron) in subsequent reactions. The results were normalized to the
absorption of the reaction mix containing none of the co-factors.

106

being nearly completely cytosolic, the lysates of strains M1 and K1443
indicate that the deA-like activity in these lysates is at least partially
membrane associated (Figure 3.3). deA is a cytosolic enzyme, with
approximately 5% of the lysate activity remaining in the pellet fraction in
the standard assay. K1443 has approximately 23% of its deA-like activity
present in the pellet fraction and M1 had approximately 67% (results
identical to P1, Figure 3.3).

Genetic testing of the a-proteobacterial strains. Attempts were
made by several researchers to clone the deA-like enzyme in all of these
strains, but the Sphingomonas strains (EMLl46, TFD44, K1443) proved
intractable to attempts at genetic manipulation. All three of the
Sphingomonas strains conjugated poorly with Escherichia coli and
displayed a strong resistance to transposon mutagenesis. All attempts at
electroporating these strains under various conditions failed. Plasmid DNA
from these strains was easy to obtain, but performed poorly in standard
digestion and ligation reactions. Nitrobacter winogradskii M1 also proved
to be difficult, resisting mutagenesis by 8 different transposons using four
delivery systems (S-17/pSUP203 [TnSOl], pSUP101[Tn1],
pSUP2021[Tn5], pSUP2017 [Tn7] (Simon, 1983); SM10(7\.pir)/pUT[mini-
Tn5 lach] and (hpir)/pUT[mini-Tn5 lacZZ] (deLorenzo, 1990);

H3 101/ColE1 [Tn3] and ColE1[Tn7] (Bagdasarian, 1997),
DHSa/pPRLlO62a[Km'] (Wolk, 1991; Cohen, in press)). M1 also
conjugated poorly, with conjugal matings producing no transconjugants
when using Escherichia coli strains (HBlOl, DH50L, XL-l Blue),
Pseudomonas putida KT2442, or Pseudamonas aeruginasa PBZO76 as

donors or recipients. M1 would conjugate at very low frequencies with

l07

 

0% 20% 101/L __g§g%g 7 80% 100%
El% in lysate
LI% in pellet

 

 

Figure 3.3. C r t ‘ " " of deA-like activity in several 2,4-D
degrading strains. The strains shown, from top to bottom: K1443, M1,
JMP134, AE228, and BH501. The bars represent 100% of the deA-like
activity found in each of the lysates, are not representative of total units of
activity, and are divided into the activity found in the cytosolic and pellet
fractions of crude lysates. In the case of DBOl, the level of 4-NP production
is 0.05% that of DB01 carrying pJP4, yet the compartmentalization of that
activity is the same. deA is a cytosolic enzyme, typically showing about
5% of the cellular conversion in the pellet fraction. M1 and K1443 are
notable among the strains tested, having 23% and 67% of their activity in
the pellet fraction. Strain P1 produced results identical to that of M1. This
may indicate membrane association of their deA-like activity.

 

108

Alcaligenes eutrophus strains (3.12x10'9). M1 was also resistant to
electroporation under the normal range of conditions and after cell
preparation as Pseudomonas or Escherichia species. Since M1 was the most
amenable to manipulation of the a-proteobacteria strains, further work was
focused on this strain.

Conjugal capture of M1 plasmids. CHEF gel analysis of strain M1
indicated the presence of at least two native plasmids. Since direct genetic
manipulation of M1 was difficult, a plasmid capture conjugal mating
experiment was preformed similar to that done for tfdA in soil microcosms
(Top, 1996). Alcaligenes eutrophus BHSOl (A. eutrophus JMP228,
containing a kanamycin resistance marked pJP4 plasmid that does not
express deA- pBHSOldy) was used as a recipient in a broth mating of
liquid log phase cultures for 18 hours on non-selective, rich media. This
mating procedure was successful in producing colonies capable of growing
on minimal salts noble agar with lOOuM kanamycin and 100 mM 2,4-D as
primary carbon source. Six colonies were isolated, and verified as A.
eutrophus by carbon source utilization, colony color, colony morphology,
and grth rate. These strains are given the Alcaligenes eutrophus strain
names P1 through P6. A parallel experiment using M1 as a donor and
JlVIP222 as the recipient produced no colonies capable of growing on 2,4-D.
This may indicate that an incomplete pathway is being transferred on the
M1 plasmid since the plasmid alone does not allow for grth in
Alcaligenes. In the complemented strains, the frequency of complementation
was more indicative of conjugation rather than transposition.

Growth and genetic characteristics of M1 and P strains.

Nitrobacter species are typically slow growing bacteria that are facultative

109

lithoautotrophs. Mixotrophic grth is possible, and in the presence of both
nitrate and organic substances, a biphasic growth pattern is observed (Holt,
1994). Strain Ml grew slowly on 2,4-D after a considerable lag phase in a
minimal medium that used ammonium as the primary nitrogen source. This
lag phase was repeated whenever the strain was allowed to enter stationary
phase, or whenever the media type was changed.

Chromosomal plugs were prepared fi'om A. eutrophus strains Pl
through P6 and their parental strains, M1 and BHSOl. CHEF gel analysis of
these strains indicated that an approximately 60 KB plasmid from strain M1
was conjugally transferred to Alcaligenes eutrophus BHSOl (Figure 3.4).
This plasmid was present in all of the P strains. The size of pBHSOldy was
unaffected, suggesting that the deA-like activity was associated with the
transfer of the M1 plasmid rather than from any form of recombination.

The strains were confirmed to be A. eutrophus by carbon utilization,
and it was noted that grth on 2,4-D was substantially slower than
observed for A. eutrophus JMP134. P1, however, does not suffer the long
lag phase of strain M1 grown on 2,4-D. Enzymatic assays using 4-NPAA
indicate membrane association of the deA-like activity and a lack of 0L-
ketoglutarate-dependence, similar to that found in strain M1 and unlike that
of JMP134 (Figure 3.2).

Attempts to isolate and subclone the genes encoding the deA-like
activity. Several attempts to isolate the 60 kb plasmid from pJP4 failed,
perhaps indicating that the plasmid lacks its own transfer genes and is
merely mobilized by pJP4. The 60 kb plasmid was able to replicate in
Pseudomonas putida 2442 and A. eutrophus JMP222, but was not found to

replicate in Escherichia coli afier a series of CHEF gel analyses. Cloning of

110

Chromosome

pJP4/pBH501
9M 1 .l

le.2

 

Figure 3.4. Schematic diagram of the CHEF gel analysis of the intact
genetic elements found in strains used in the conjugal capture of a
deA-like gene from strain M1. This schematic represents an original
CHEF gel, traced directly from a scanned photograph. Strains included:
501, A. eutrophus BH501 (containing pBH501, tfdA'; Top, 1996); P1, P2,
and P3, A. eutrophus strains (containing pBHSOl and a 60 kb plasmid from
strain M1) that may utilize 2,4-D as a sole carbon source; M1, N.
winogradskyi strain M1 (containing at least two uncharacterized plasmids);
228, A. eutrophus JMP228 (containing no plasmids, (Don, 1985)); 43, A.
eutrophus JMP134 (TF D43) (containing pJP4). Sc represents the
Saccharomyces cereviseae chromosomal molecular weight marker, with
bands representing 2200, 1600, 825, 785, 750, 680, 610, 565, 450, 365,450,
365, 285, and 225 kb of linear DNA. Plasmid DNA was undigested and
migrated more slowly. This analysis clearly shows the presence of pJP4 (or
the tfdA' version, pBHSOl) in A. eutrophus strains BH501, P1, P2, P3, and
JMPl34. A second plasmid of approximately 60 kb was present in M1, P1,
P2, and P3. A third, smaller plasmid was present in strain M1 alone. There
may be additional smaller plasmids present in these strains that would not
be seen well under these conditions.

ill

the tfdA-like gene was attempted by shotgun cloning methods. Plasmid
DNA for the cloning attempts was prepared using Quiagen maxi-prep
columns and the large plasmid procedural modifications or by rapid plasmid
DNA preparation (Holmes, 1981). Mixed plasmid DNA from strain M1 or
strain P1 was used for cloning. Shotgun cloning was attempted using
various partial digestions (BamHI, HinDIII, EcoRI, EcoRV, KpnI, Natl,
SalI, SacI, Sau3A, SphI) and broad host range plasmid vectors (pTJS75a
[Tc'] (Schmidhauser, 1985); pMMB207 [Cm'] (Morales, 1991); pMMB503
[Sm'] (Overbye, 1996)). Competent E. coli XLl-Blue cells were
transformed with the ligation mixtures and plated on selective media.
Transformants were conjugally transferred to A. eutrophus BH501 and
screened for grth on 2,4-D as a sole carbon source. No plasmids that
complemented BH501 were found, perhaps indicating that the genes
responsible for complementation in the intact plasmid were arranged in an
operon or in such a way that the shotgun clones were not of suitable size.
Additionally, deregulated or multiple copies of this gene or genes may be

toxic to the host cells.

112

SUMMARY

This study was designed to look at the deA-like enzymes of a
diverse set of 2,4-D degrading bacteria, concentrating on those which may
be least like the canonical enzyme, in hope of finding an alternate 2,4-D
degradation pathway or evolutionarily distinct enzymes. Afier extensive
genetic and biochemical testing, four a-proteobacterial strains were
selected. Three of these strains proved intractable to genetic manipulation.
In a conjugal capture experiment, strains containing a plasmid that
complemented a tfdA' canonical pathway were isolated from Nitrobacter
winogradskii strain M1. These complemented strains had the same distinct
membrane association and indifference to the known co-factors of deA as
strain M1, indicating that this enzyme may be evolutionarily distinct. CHEF
gel analysis confirmed the presence of a 60KB plasmid in the
complemented strains. All attempts to clone the deA-like activity from the
complemented strain or from M1 by shotgun cloning have failed. This
enzyme may be encoded on a fragment too large for this method, or the
enzyme may be toxic in multiple copies. Further work should provide the
DNA sequence of the M1 tfdA-like gene(s) and allow more complete

biochemical elucidation.

113

REFERENCES

Altshul, S. F., Gish, W., Miller, W., Myers, E. W., and D. J. Lipman. 1990.
Basic Local Alignment Search Tool. J. Mol. Biol. 215: 403-410.

Amy, P.S., Schulke, J. W., Frazierand, L. M., and R.J. Seidler. 1985.
Characterization of aquatic bacteria and cloning of genes specifying
partial degradation of 2,4-dichlorophenoxyacetic acid. Appl. Environ.
Microbiol. 49: l 237- 1245.

Bachmann, B. J. 1972. Pedigrees of some mutant strains of Escherichia coli
K12. Bacteriol. Rev. 36:525-557.

Bagdasarian, M. 1997. Strain collection. Mailing address: S-112 Plant
Biology Building, Michigan State University, E. Lansing, MI 48224.

Cohen, M. F., Meeks, J .C., Cai, Y. A., and C. P. Wolk. In press. Transposon
mutagenesis of heterocyst-forming filamentous cyanobacteria.

DeLorenzo, V., Herrero, M., J akubzik, U., and K. Timmis. Mini-Tn5
transposon derivatives for insertion mutagenesis, promoter probing,
and chromosomal insertion of cloned DNA in Gram-negative
Eubacteria. 1990. J. Bacteriol. 172: 6568-6572.

Devereaux, J. 1989. The GC G Sequence Analysis Software Package
Versions 6.0 - 8.1, Genetics Computer Group, Inc., University
Research Park, 575 Science Dr., Suite B, Madison, WI, 53711, USA.

Don, R.H., A.J. Weightman, H.J. Knackmuss, and K.N. Timmis.
Transposon mutagenesis and cloning analysis of the pathways for

degradation of 2,4-dichlorophenoxyacetic acid and 3-chlorobenzoate

114

in Alcaligenes eutrophus JMP134 (pJP4). 1985. J. Bacteriol. 161:85-
90.

F elsenstein, J. 1981. Evolutionary trees from DNA sequences: a maximum
likelihood approach. J. Mol. Evol. 31(6): 368-76.

Figurski, D. H., and D. R. Helinski. 1979. Replication of an origin-
containing derivative of plasmid RK2 dependent on a plasmid
function provided in trans. Proc. Natl. Acad. Sci. USA 76: 1648-
1652.

Fukumori, F. and R. P. Hausinger. 1993. Purification and characterization of
2,4-dichlorophenoxyacetate/a-ketoglutarate dioxygenase. J. Biol.
Chem. 268:2431 1-24317.

Fulthorpe, R.R., C. McGowan, O.V. Maltseva, W.E. Holben, and J .M.
Tiedje. 1995. 2,4-dichlorophenoxyacetic acid-degrading bacteria
contain mosaics of catabolic genes. Appl. Environ. Microbiol. 61:
3274-3281.

Fulthorpe, R.R., A. N. Rhodes, and J. M. Tiedje. 1996. Pristine soils
mineralize 3-chlorobenzoate and 2,4-dichlorophenoxyacetate via
different microbial populations. Appl. Environ. Microbiol. 62:1 159-
1166.

Herrero, M,. deLorenzo, V., and K. N. Timmis. 1990. Transposon vectors
containing non-antibiotic resistance selection markers for cloning and
stable chromosomal insertion of foreign genes in Gram-negative
bacteria. 1990. J. Bacteriol. 172: 6557-6567.

Holmes, D. S., and M. Quigley. 1981. A rapid boiling method for the
preparation of bacterial plasmids.Anal. Biochem. 114: 193-197.

Holt, J. G., Krieg, N. R., Sneath, P. H. A., Staley, J. T., and S. T. Williams,

115

eds. ngey's Manual of Determinative Bacteriology. Nint_h Edition,
Williams and Williams, Philadelphia, 1994.

Ka, J .O., Holben, W.E., and J .M. Tiedje. 1994. Genetic and phenotypic
diversity of 2,4-dichlorophenoxyacetic acid (2,4-D)-degrading
bacteria isolated from 2,4-D-treated field soils. Appl. Environ.
Microbiol. 60:1 106-1 1 15.

Kamar, S., Tamura, K., and M. Nei. 1993. MEGA: Molecular Evolutionary
Genetics Analysis, version 1.0, The Pennsylvania State University,
University Park, PA, 16802, USA.

Matheson, V. G., Forney, L. J ., Suwa, Y., Nakatsu, C. H., Sexstone, A. J .,
and W. E. Holben. 1996. Evidence for acquisition in nature of a
chromosomal 2,4-dichlorophenoxyacetic acid/a-ketoglutarate
dioxygenase gene by different Burkholderia spp. Appl. Environ.
Microbiol. 62(7): 2457-2463.

McGowan, C. 1995. Interspecies gene transfer in the evolution of 2,4-
dichlorophenoxyacetate degrading bacteria. Ph.D. thesis, Department
of Microbiology, Michigan State University.

Morales, V. M., Bakman, R., and M. Bagdasarian. 1991. A series of wide-
host-range low-copy-number vectors that allow direct screening for
recombinants. Gene 97: 39-47.

Overbye Michel, L. M. Sandkvist, and M. Bagdasarian. 1995. Specificity of
the protein secretory apparatus: secretion of the heat-labile
enterotoxin B subunit pentamers by different species of Gram'
bacteria. Gene 152: 41-45.

Peck, S. C., Reinhardt, D., Olson, D. C., Boller, T., and H. Kende. 1992.

116

Localization of the ethylene-forming enzyme from tomatoes, 1-
aminocyclopropane-l-carboxylate oxidase, in transgenic yeast. J.
Plant Physiol. 140(6): 681-686.

Sassanella, T., F. F ukumori, M. M. Bagdasarian, and R. P. Hausinger. Use
of 4-Nitrophenoxyacetic Acid for the Detection and Quantification of
2,4-D/0t-Ketoglutarate Dioxygenase Activity in 2,4-D Degrading
Microorganisms. Appl. Environ. Microbiol. 63(3): 1 189-1 191.

Schmidhauser, T. J. and D. R. Helinski. 1985. Regions of broad-host-range
plasmid RK2 involved in replication and stable maintenance in nine

species of Gram-negative bacteria. J. Bacteriol. 164: 446-455.
Simon, R., Priefer, U., and A. Piihler. A broad host range mobilization

system for in vivo genetic engineering: transposon mutagenesis in
Gram-negative bacteria. 1983. BioTechnology. 1: 784-790.

Suwa, Y., Wright, A. D., Fukumori, F., Nummy, K. A., Hausinger, R. P.,
Holben, W. E., Forney, L. J. 1996. Characterization of a
chromosomally encoded 2,4-dichlorophenoxyacetic acid/01-
ketoglutarate dioxygenase from Burkholderia sp. Strain RASC. Appl.
Environ. Microbiol. 67(2): 2464-2469.

Takeuchi, M., Sawada, H., Oyaizu, H., and A. Yokota. 1994. Phylogenetic
evidence for Sphingomonas and Rhizomonas as nonphotosynthetic
members of the 0t—4 subclass of the Proteobacteria. Int. J. Syst.
Bacteriol. 44(2): 308-14.

Tonso, N.L., V.G. Matheson, and WE. Holben. 1995. Polyphasic
characterization of a suite of bacterial isolates capable of degrading

2,4-D. Microbial Ecol. 30: 3-24.

117

Top, E. M., Holben, W. E., and L. J. F omey. 1995. Characterization of
diverse 2,4-D degradative plasmids isolated from soil by
complementation. Appl. Environ. Microbiol. 61 : 1691-1698.

Top, E. M., O. M. Maltseva, and L. J. F omey. 1996. Capture of a catabolic
plasmid that encodes only 2,4-dichlorophenoxyacetic acid/0L-
ketoglutarate dioxygenase (deA) by genetic complementation. Appl.
Environ. Microbiol. 62(7): 2470-2476.

Wolk , C. P., Cai, Y., and J. M. Panoff. 1991. Use of a transposon with
luciferase as a reporter to identify environmentally responsive genes
in a cyanobacterium. Proc. Natl. Acad. Sci. USA 88: 5355-

5359.

 

Chapter 4:

Variable 2,4-D Permeability in Bacteria

ABSTRACT

During the investigation of the evolution and assembly of 2,4-
dichlorophenoxyacetic acid (2,4-D) catabolic pathways, a colorimetric assay
was developed for the detection and quantification of 2,4-
dichlorophenoxyacetic acid (2,4-D)/0t-ketoglutarate dioxygenase (deA)
activity in extracts and whole cells of environmental and genetically
engineered bacteria (Chapter 2). During the construction of many of the
engineered strains used in these experiments, results were observed that
suggested that cell permeability to 4-NPAA (and thereby 2,4-D) varied by
species, and that this permeability was affected in some strains by the
introduction of the low copy number pJP4 plasmid, but not affected by the
higher copy number expression vector containing an overexpressible tfdA
gene. In addition, radiolabeled 2,4-D uptake assays indicate substantially
different levels of permeability to 2,4-D among 2,4-D degrading isolates
and engineered strains. Permeation is very likely to be a significant factor in
successful 2,4-D degradation. The canonical 2,4-D degradation plasmid
pJP4 influences 2,4-D uptake in some cells and this activity can be
eliminated by transposon mutagenesis. These findings could have

significant biochemical and evolutionary implications.

118

119

INTRODUCTION

The selective transport of solutes across cellular membranes is a
significant part of cellular metabolism. Gram-negative bacteria have a
complex cell surface, consisting of an outer membrane comprised of
phospholipids, proteins and lipopolysaccharides, a cell wall comprised of
peptidoglycan, and an inner, or cytoplasmic, membrane composed of
phospholipids and proteins (Nikaido, 1985). Solutes have to pass through
considerable obstacles before they may be metabolized in the cytoplasm.
The outer membrane provides limited permeability to small solutes via
proteinaceous channels that may be non-specific (porins)(Sukhan, 1995), or
substrate specific (N ikaido, 1985). The cell wall is thought to be completely
permeable to most solutes while providing rigidity to the cell membrane
complex. The cytoplasmic membrane, however, is impermeable to most
solutes unless a transport system is provided. Gram-positive bacteria have
only a single membrane, and its permeability characteristics are like those of
the Gram-negative cytoplasmic membrane.

It has been noted that Gram-negative bacteria have more resistance to
lipophilic and amphiphilic compounds (dyes, detergents, antibiotics) than
Gram-positive bacteria (Nikaido, 1996). This intrinsic resistance of these
bacteria was once entirely attributed to the outer membrane, as narrow porin
channels slow the penetration of even small hydrophobic solutes and the

low fluidity of the lipopolysaccharide decreases the diffusion of lipophilic
solutes (Nikaido, 1985, Plésiat, 1992). Even with these characteristics,

however, solute equilibrium across the outer membrane is achieved rapidly.

Periplasmic concentrations of many antibiotics can reach 50% of the

120

external concentration in 10 to 30 seconds in Pseudomonas aeruginosa,
known as a highly impermeable strain, and in a much shorter in E. coli
(Nikaido, 1994). Solutes like 2,4-D should have no difficulty in rapidly
penetrating to the periplasmic space of most Gram-negative bacteria, where
transport across the inner membrane is typically facilitated in some manner.
The general impermeability of the cytoplasmic membrane is likely to have
greater significance in the case of xenobiotic compounds due the lack of
specific transporters. Many of these compounds may have to rely on
fortuitous transport by permeases specialized for the transport of structurally
or chemically related substrates.

Bacterial permeases can be broadly grouped according to their
mechanism of energy coupling: those driven by electrochemical gradients
(synport, antiport, uniport), and those driven by substrate-level
phosphorylation (including the ABC import permeases) (Ames, 1990).
Electrochemical gradient transporters are typically osmotic shock resistant,
are a single, highly hydrophobic transmembrane protein, transport substrates
by synport or antiport mechanisms driven by an ion or proton gradient, have
a lower affinity for the substrate, and generate low to moderate
concentration gradients. Substrate-level phosphorylation permeases are
typically osmotic shock sensitive, are complex in structure, have a higher
affinity for their substrates, and can generate very large concentration
gradients (up to 105-fold).

Among the import permeases, ABC transporters are members of the
second group, forming an evolutionarily related superfamily of proteins that
transport a wide variety of substances (sugars, amino acids, peptides, ions,

and vitamins). They are present in Bacteria, Eucarya (Ames, 1986), and

 

121

Archaea (J ovell, 1996). In Gram-negative bacteria, a periplasmic substrate-
binding protein will bind the substrate once it diffiises to the periplasm. The
binding protein then presents the substrate in a concentrated form to a
cytoplasmic membrane bound complex. This concentration is effective due
to the high binding affinity of the periplasmic protein and the high
concentration of the protein in the periplasm, which can be in the mM
range. The binding protein interacts with the hydrophobic membrane "
spanning subunits of the complex, releasing the substrate and allowing it to i
be transported across the membrane in an ATP driven mechanism. Each
transport requires the hydrolysis of one or two ATP molecules.

Due to the chemical structure of 2,4-D, it is likely that 2,4-D is L

 

transported into some cells fortuitously by one or more native permeases,
like the aromatic amino acid permeases for phenylalanine (pheP),
tryptophan (mtr and tnaB), or tyrosine (ter) (reviewed in Sarsero, 1991).
Additionally there are permeases, like aroP the general aromatic amino acid
permease (Sarsero, 1991), that have a greater substrate range and may
transport 2,4-D across the cytoplasmic membrane. The permeases that are
most likely to be involved in 2,4-D transport are the substrate-level
phosphorylation driven ABC importers, due to the high substrate affinity
required to transport 2,4-D in the environment and the degree to which 2,4-
D is removed from culture media and contaminated soils.

In nature, conjugative plasmids are known to carry many types of
permease genes, some provide resistance to anti-microbial agents by
increasing efflux (arsenic, cadmium, chromate, tetracycline, ethidium) or by
increasing influx or a detoxifying reaction (mercury). Others assist bacteria

in co-factor assimilation (iron), or by extending potential carbon sources

122

(sucrose and citrate in E. coli) (reviewed in Tisa, 1990). Though many
catabolic plasmids for synthetic chemicals have been studied, few have
examined the permeation of the xenobiotic compounds into bacterial cells
that degrade them. The bioavailability and uptake of these compounds is
likely to be a significant factor in the process of biodegradation, and has
direct biochemical and evolutionary implications. In the case of 2,4,5-T
degradation, there has been some indication that permeation is affected by a F
subcloned region of DNA containing some of the degradation pathway

genes (Haughland, 1991). This work attempts to examine permeation factors

 

that may affect the utilization of 2,4-D. E

123

RESULTS AND DISCUSSION

Identification of Putative Permeation Effects.

After examining the response of various host/vector constructs to the
4-NPAA assay, the results demostrated that native cell permeability to 2,4-D
varied widely by species, from naturally permeable to impermeable, and that
permeability was affected in some strains by the addition of the 90 kb
plasmid pJ P4 (containing the entire known 2,4-D degradation pathway and
several other genes), but was not affected by a high copy number expression
vector containing an overexpressed tfdA gene (Table 2.1 and 4.1).
Escherichia coli CB101 carrying a hilly expressed ydA gene (pMMBSl 1)
produced no 4-NP from 4-NPAA in whole cells, though when the cells were
sonicated, the lysates rapidly converted 4-NPAA to 4-NP. The same strain
containing pJP4 reacted similarly, though tfdA was expressed at a much

lower level. Burkholderia cepacia DB01 did not convert 4-NPAA, but

Table 4.1. Response of various engineered host/plasmid combinations
using 4-NPAA as an alternate substrate to assay for deA activity. (+)
indicates production of 4-nitrophenol from 4-NPAA under conditions noted
in Appendix B.

 

Strain no plasmid deA+ deA+ pJP4
wholeneflLlysatetholecelLIXsatcs

 

Escherichia coli CB101 - -

+ - -
Pseudomonas putida PB2442 - - + + +
Alcaligenes eutrophus JMP228 - + + + +
Burkholderia cepacia DB01 - + + + +

 

 

124

produced 4-NP from 4-NPAA when either pJP4 or pMIVIBSIl were present,
indicating that 2,4-D and 4-NPAA were transported into the cell. However,
in Pseudomonas putida PB2442, the introduction of plasmid pJP4 allowed
the uptake and conversion of 4-NPAA, whereas the introduction of
pMMBSll alone did not. These results support the variability of permeation
of 4-NPAA (and likely 2,4-D) in different environmental strains, and
suggest that there may be a factor on the plasmid pJP4 that influences the
permeation of substituted phenoxy compounds.

In order to clarify that the permeation effects noted for 4-NPAA were
also applicable to 2,4-D, a 14C-2,4-D uptake assay was developed (Figure
4.2). E. coli DH50L was shown to be impermeable to 2,4-D and 4-NPAA
using the 4-NPAA assay and the radiolabeled uptake assay (Figure 4.2). The
addition of the test plasmids (pJP4 and pMMBSl 1) had no effect on the
permeability of this strain to 2,4-D. Burkholderia cepacia DB01 and
Alcaligenes eutrophus JIVIP228 were shown to retain l4C-2,4-D at a low
level (presumably cell adsorption or cellular transport equilibrium). When
carrying pMMBSll (expression vector with q‘dA alone), both strains readily
transported and converted 4-NPAA (Table 2.1), and when carrying pJP4,
both showed significant uptake and conversion of 14C-2,4-D over a 5 minute
period (Figure 4.2). However, in the case of Pseudomonas putida PB2442,
l4C-2,4-D was retained very poorly in the absence of a plasmid or when
carrying pMMBSll (in a similar manner as for 4-NPAA; Fig. 4.1), yet
transported 2,4-D into the cell at a much higher rate in the presence of pJ P4.
These findings indicate that the 4-NPAA results are a good reflection of the
transport of 2,4-D, that there is substantial variability among bacterial

 

125

3.5 -

 

 

 

 

 

 

 

2.5--

 

 

 

 

 

 

 

 

  
 
 

0.2564

010“ 0.0729 0.071
0 - .
JMP134 K1443 EML146 PppJP4 PptfdA Pp m8 Ppm11

Strains

Figure 4.1. Variability in the permeation or utilization of 2,4-D of
environmental strains and putative permease mutants is indicated by 4-
NPAA whole cell assay. Among the native 2,4-D degrading isolates shown
(A. eutrophus JMP134, Sphingomonas sp. K1443 and EMLl46), there is a
substantial difference in the amount of 4-NPAA that is converted to 4-NP
over 24 hours in liquid cultures. Preconditioned log-phase cultures grown on
M2 medium were used to inoculate secondary cultures of M2 media
supplemented with 1 mM 4-NPAA (M24 media). Pp denotes engineered
Pseudomonas putida PB2442 strains: pJP4 denotes the presence of that
plasmid, tfdA denotes the presence of the expressible #dA construct
pMMBSll (Table 2.1), and m8 and M11 are TnS insertion mutants of pJP4.
P. putida strains were preconditioned in M2 media supplemented with 0.1%
casamino acids and treated as noted. All cultures were well shaken at 30°C
for 24 hours. One ml cultures samples were chilled to 4°C, pelleted, and the
absorbance of the supernatant measured at 401 nm.

126

40000

 

+ D801
35000 -<>- DBO1/101
—)le— OHS-alpha
30000 + JMP222
-I-— JMP134
25000 e. K‘19
I l
E 20000 ‘
0
15000
10000
5000
0
1 2 3 4 5
Time (min)

Figure 4.2. Determination of 2,4-D permeation in various environmental
and engineered strains using a 2,4-D uptake assay. Environmental strains
Burkholderia cepacia DB01, Alcaligenes autrophus JMP228 and JMP134,
and Alcaligenes sp. K-l9, and the engineered strains E. coli DH50t and B.
cepacia DB01/101 were tested for permeation of 2,4-D. These strains were
grown in the presence of 2,4-D, washed in minimal medium without 2,4-D,
and incubated in a large volume of the same medium. The culture was then
spiked with 1 mM 2,4-D with a 14C-2,4-D tracer, the suspension was mixed
thoroughly, and samples drawn at one minute intervals. The samples were
rapidly filtered through a 0.45 um Millipore nitrocellulose filter, and washed
three times with minimal salts medium containing 1 mM 2,4-D. The filters
were immediately paced into a toluene based scintillation fluid. Uptake of
the radiolabeled 2,4-D was measured by scintillation counter, and adjusted
for background and absorbance of 2,4-D to the nitrocellulose filters.

127

isolates in the ability to take up 2,4-D, and that the plasmid pJP4 allows P.
putida PB2442 to take up 2,4-D at a higher rate than the native strain.

The Mutagenesis of 3P4, and the Subseguent Screening for

Permease Mutants.

To test the idea that p] P4 may harbor an unknown gene that affects
the permeability of bacterial cells like P. putida to 2,4-D and 4-NPAA,
random transposon mutagenesis of pJP4 was performed in E. coli, and the
mutant plasmids were mated into P. putida PB2442. The transconjugants
were then screened using the 4-NPAA colony assay (Appendix B), and 22
colonies that failed to convert 4-NP were chosen. These strains were
purified, then tested for 4-NPAA conversion again, using the whole cell
liquid culture assay (Appendix B). The same strains were tested as cell
lysates, and two strains, denoted m8 and m22, were found to convert 4-

NPAA in lysates, but not in whole cell assay.

Hybridization Experiments.

Southern hybridization experiments using the pJP4 mutant plasmids m8 and
m22 and an internal fragment of Tn5 as a probe indicated that the

transposon insertions did not occur within the known 2,4-D genes, but

rather in a 33 kb fragment containing the replication and maintenance genes.

Production of 4-NP using the whole cell 4-NPAA assay to show uptake and
conversion by deA in these strains is shown (Fig. 4.1). The mutant strains
have about 28% of the transport rate of the wild type pJP4.

m

128

2,4-D Incorporation Assay.
The deleterious effect of the pJP4:Tn5 transposon insertions m8 and m22

on permeation of the host strain was tested more rigorously using a l4C-2,4-
D incorporation assay. Alcaligenes eutrophus JMP134 (wild-type pJP4)
incorporated radiolabeled 2,4-D, whereas A. eutrophus JMP228 (no
plasmid) and A. eutrophus carrying the mutant pJP4 plasmids m8 and m22
did not appreciably incorporate radiolabeled 2,4-D within 150 minutes
(Figure 4.3). The increase in incorporation of carbon fi'om 2,4-D by JMP134
was significant, and could not be attributed to simple increase in cell mass

(or cell absorbance) (Figure 4.3). All of the strains except JMP228 had a

 

detectable level of deA activity when tested in lysates. These findings were
further supported by a simple grth study where the strains were
inoculated into minimal 2,4-D medium and incubated for 5 days. JMP134
grew to maximum density in less than 48 hours and m22 produced a small
increase in cell mass - perhaps indicating that the transposon insertion hasn't
completely eliminated the function of the permeation factor - whereas none

of the other strains grew appreciably.

Partial Subcloning of the Transposon Insertion Site.

Several methods of cloning the putative permeation factor have been
attempted unsuccessfully due to what is thought to be toxic effects of
multiple copies of the region. Asymmetric subclones of the transposon
insertion site have been isolated by partially digesting plasmid DNA from
m8 and m22 using BamHl. This enzyme cuts in the center of Tn5, yet
leaves the neoR gene intact, allowing the tracking of the fiagment by

screening for kanamycin resistance. A subclone from m22 was isolated

129

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

7000 f
{+JMP228
6°00 1"; +JMP134 “-
+m8
5000‘ "m /
a 4000. —Linear(JMP134) ‘ ,
0.
‘3 3000 4—-—-—~-~
2000
1000-
0 . . . . . - . . . .'
0 15 30 45 60 75 90 105 120 138 150
Tlme(mln)
0.55
E 0.5 +JMP222
g +JMP134
m
.“3 0.45 * "'8
: -D—m11
O
E 0.4
0
£0.35

 

.0
w

0 1 5 30 45 60 75 90 105 120 1 35 1 50
Time (mln)

Figure 4.3. 2,4-D incorporation assay. The permeation of 2,4-D in
Alcaligenes eutrophus strains was tested by measuring incorporation of 14C
from ring-labeled l4C-2,4-D. JMP222 contains no plasmids, JMP134 carries
pJP4 (the canonical 2,4-D pathway), and m8 and m1 1 are Tn5 mutants of
pJP4 that exhibited decreased ability to convert 4-NPAA to 4-NP in whole
cells, yet expressed tfdA in cell lysates. The incorporation of carbon from
2,4-D degradation into biomass was measured as DPM from washed culture
samples taken at various times (top panel). The cell density of the cultures
was measured simultaneously (bottom panel).

 

 

 

130

 

 

 

 

 

 

 

s 1|! H E
vector neor pJP4 fi'agment vector
——>< > < >4—
+ . + ’
Tn 5 fragment

Scale: 1" = approx. 1 kb

Figure 4.4. Physical map of plasmid me2. This fi'a'gment is a portion of
the plasmid m8 (pJP4:Tn5)(this work) in the vector pMMB207, which is
only partly shown (Morales, 1991). B represents restriaction sites for BgII, E
represents EcoRI, H represents HinDIII, and N represents NotI; neo'
provides resistance to neomycin and kanamycin. The fragment was isolated
by a 'walking' experiment - using the unique SalI site and the kanamycin
resistance gene of Tn5 to capture one end of the transposon as well as a
portion of the insertion site on pJP4. It is likely that the distal end of this
fragment has rearranged leaving the polylinker sites of the vector intact. This
plasmid is relatively unstable, being maintained only under antibiotic
selection, and is difficult to purify.

131

using pMMB207 as a vector. This plasmid was difficult to purify in large
quantities, having a strong tendency to partition to the hyrophobic phase
during phenol/chloroform extractions. Plasmid DNA was purified using the
Quiagen mini-plasmid preparation columns. This 5.3 kb subcloned fragment
contains the portion of Tn5 that contains the neoR gene and extends out past
the end of the transposon (Fig. 4.4). This fragment cannot be excised using
BamHl , indicating that this fragment may have rearranged at the distal end
to the transposon to generate the stable subclone. Since the entire fragment
cannot be excised for transfer to a more suitable sequencing vector, direct
sequencing of the putative permeation factor using a primer to the IS
sequence of the transposon was attempted. Four automated attempts at three
facilities and manual single strand PCR sequencing attempts failed at the

time this study ended.

Evolutionag and Ecological Implications.

The effect of differential permeability of compounds in nature can
have significant ecological and evolutionary impact. If there is a permeation
step previous to a cytosolic degradation reaction, the understanding of the
kinetics of xenobiotic degradation pathways could be significantly altered.
Additionally, if permeation were the slow step in the degradation reaction,
this could have significant ecological impact on the competitiveness of the
strains in the environment, and, in a practical sense, the predicted utility of
the strains in biodegradative processes. Biochemical elucidation of the
effect of various targeted or fortuitous permease reactions would also assist
in understanding the effect of bioavailability of the substrate in natural or

engineered environments. Since it is clear there are substantial differences

132

in the ability of Gram-negative bacteria to transport 2,4-D into the cell,
several significant issues emerge. If differential permeation of compounds
occurs in environmental bacteria, it is likely that only a limited number of
bacterial strains are suitable for metabolic level 2,4-D degradation. It is
known that there are plasmids that carry catabolic pathways that encode
transport proteins that increase the level of permeation of specific substrates
in their host cells. There have been some indications that this phenomenon
may be more widespread that was thought previously. Further, if the level of
permeability can be altered by the addition of plasmid borne factors, there
may be a limitation on the species that may be affected, perhaps due to poor
membrane insertion, inability for periplasmic facilitators to reach the
periplasmic space, or the lack of required associated membrane proteins.
Non-specific transporters may be more important for the evolution of novel
catabolic pathways, and may be a contributing factor in the recalcitrance of
certain compounds to microbial degradation.

This work suggests that there are strains that do not transport 2,4-D
across their cell membranes in the native state, and therefore the number of
strains that may be able to utilize 2,4-D as a significant source of carbon and
energy may be limited. It is highly likely that there is some kind of
permeation or 2,4-D uptake altering factor present on pJP4, the canonical
2,4-D degradation plasmid, and that factor may be disrupted by site directed
mutagenesis (producing significant loss of 2,4-D uptake) without
interrupting the expression of tfdA. This factor only functions in some of the
strains, and is not needed in others, as there are native transporters that
allow significant uptake of 2,4-D. 4-NPAA and 2,4-D are transported in the

same manner in all the strains that would tested. The mechanism of 2,4-D

 

133

uptake in bacterial strains is unknown, but it is likely that 2,4-D is
transported by common broad-substrate range ABC import permeases, like
aroP. The transport of 2,4-D is probably driven by substrate-level
phosphorylation, due to the apparent high affinity of the permease and the

virtually complete removal of 2,4-D from culture media.

134

REFERENCES

Ames, G. F .-L. 1986. Bcaterieal periplasmic transport systems: structure,
firnction, mechanism, and evolution. Ann. Rev. Biochem. 55: 397-
425.

Ames, G. F.-L. 1988. Structure and mechanism of bacterial periplasmic
transport systems. J. Bioenerg. Biomembr. 20(1): 1-18.

Ames, G. F.-L., and A. K. Joshi. 1990. Energy coupling in bacterial
periplasmic permeases. J. Bacteriol. 172(8): 4133-4137.

Boorsma, A., van der Rest, M. E., Lolkema, J. S., and W. N. Konings. 1996.
Secondary transporters for citrate and the Mg+2-citrate complex in
Bacillus subtilis are homolog proteins. J. Bacteriol. 178(21): 6216-
6222.

DeLorenzo, V., Herrero, M., Jakubzik, U., and K. Timmis. Mini-TnS
transposon derivatives for insertion mutagenesis, promoter probing,
and chromosomal insertion of cloned DNA in Gram-negative
Eubacteria.1990. J. Bacteriol. 172: 6568-6572.

Fath, M. J ., and R. Kolter. 1993. ABC transporters: bacterial exporters.
Microbiol. Rev. 57(4): 995- l 0 1 7.

F ukumori, F., and RP. Hausinger. 1993a. Alcaligenes eutrophus JMP134

“2,4-dichlorophenoxyacetic acid monooxygenase” is an or-
ketoglutarate-dependent dioxygenase. J. Bacteriol. 175:2083-2086.
Fukumori, F., and R. P. Hausinger. 1993b. Purification and characterization
of 2,4-dichlorophenoxyacetate/ a-ketoglutarate dioxygenase. J. Biol.
Chem. 268:2431 1-24317.
Haughland, R. A., Sangodkar, U. M. X., Sferra, P. R., and A. M.

 

135

Chakrabarty. 1991. Cloning and characterization of a chromosomal
DNA region required for growth on 2,4,5-T by Pseudomonas cepacia
AC1100. Gene. 100: 65-73.

Jovell, R. J ., Macario, A. J. L., and E. C. deMacario. 1996. ABC

transporters
in archaea: two genes encoding homologs of the nucleotide-binding
components in the methanogen Methosarcina mazei S-6. Gene. 174:
281-284.

Nikaido, H. and M. Vaara. 1985. Molecular basis of bacterial outer
membrane permeability. Microbiol. Rev. 49: 1-32.

Nikaido, H. 1994. Prevention of drug access to bacterial targets:
permeability barriers and active efflux. Science. 264: 382-388.

Nikaido, H. 1996. Multidrug efflux pumps of Gram-negative bacteria. J.
Bacteriol. 178(20): 5853-5859.

Plésiat, P., and H. Nikaido. 1992. Outer membrances of Gram-negative

bacteria are permeable to steroid probes. Mol. Microbiol. 6: 1323-
1333.

Ramos, A., Poolman, B., Santos, H., Lolkema, J. S. , and W. N. Konings.
1994. Uniport of anionic citrate and proton consumption in
Leuconostoc oenos. J. Bacteriol. 176: 4899-4905.

Sarsero, J. P., Wookey, P. J ., Gollnick, P., Yanofsky, C., and A. J. Pitard.
1991. A new family of integral membrane proteins involved in
transport of aromatic amino acids in Escherichia coli. J. Bacteriol.
173(10): 3231-3234.

Sassanella, T., F. Fukumori, M. M. Bagdasarian, and R. P. Hausinger. Use

 

136

of 4-Nitrophenoxyacetic Acid for the Detection and Quantification of
2,4-D/0t-Ketog1utarate Dioxygenase Activity in 2,4-D Degrading
Microorganisms. Appl. Environ. Microbiol. 63(3):] 189-1 191.

Sukhan, A., and R. E. Hancock. 1995. Insertion mutagenesis of the
Pseudomonas aeruginosa phosphate-specific porin OprP. J.
Bacteriol. 177(17): 4914-4920.

Tisa, L. S. and B. P. Rosen. 1990. Transport systems encoded by bacterial
plasmids. J. Bioenerg. Biomembr. 22(4): 493-507.

Thayer, J. R., and M. L. Wheelis. 1976. Characterization of a benzoate
permease mutant of Pseudomonas putida. Arch. Microbiol. 110: 37-
42.

 

CONCLUSION

This work examines the initial step of the known 2,4-D degradation
pathways in 2,4-D degrading soil bacteria. From the canonical pathway, 2,4-
D/a-ketoglutarate dioxygenase (deA) was the focus of this diverse set of
investigations into the microbial ecology and the evolution of catabolic

pathways. The significant findings of this work include:

 

o The a-ketoglutarate-dependent dioxygenase superfamily, to which
deA belongs, can be divided into several groups that are defined
by the conservation of residues thought or known to be essential
for iron binding and activity. Though these enzymes are thought to
have a very similar mechanism, it is likely that this superfamily is

polyphyletic.

o A continuous, quantitative chromogenic assay for deA and
analogous activities was developed for the screening of
environmental and genetically engineered strains using both intact

- cells and cell lysates. Colony screening assay for petri plates 1

allows rapid testing of large numbers of colonies. This 4-

137

138

nitrophenoxyacetic acid (4-NPAA) assay was used to survey a
diverse collection of 2,4-D degrading soil bacteria and genetically
engineered strains, indicating substantial diversity among Gram-
negative bacteria in the ability to transport and/or metabolize 4-

NPAA and 2,4-D.

o Nitrobacter winogradskyi strain M1 was eXamined in greater
detail, and was found to carry a plasmid that encodes an atypical
deA-like activity that is both membrane associated and requires

different co-factors than deA.

0 Differential response when using the 4-NPAAassay indicated that
there could be differential permeation of 4-NPAA among soil
bacteria. Differential permeation was confirmed for 2,4-D using
l4C-2,4-D uptake and incorporation assays. Transposon
mutagenesis of the canonical 2,4-D degradation plasmid pJP4
indicated that the plasmid encodes an unknown factor that

influences the permeation of 2,4-D in some strains.

APPENDICES

 

APPENDIX A

 

APPENDIX A

Detailed Methods From Chapter 1.

Strains, enzymatic functions, organisms of origin, and accession
numbers are described in Table 1.1. Citations listed here are noted as

References in Chapter 1.

Data file assembly. The known a-ketoglutarate-dependent dioxygenase
sequences were obtained from world-wide protein databases. All sequences
used were amino acid sequences, as more character states were required for
a significant analysis due to the complexity of this highly divergent group.
The superfamily family was initially defined as those enzymes that bind 01-
ketoglutarate, ferrous iron, oxygen, and an additional variable substrate, and
that theoretically have a similar mechanism (though functionally fall into
different categories). This group of enzymes is reviewed by Prescott (1993),
and additional sequences from literature searches were added. Using the
Genetics Computer Group (GCG) suite of programs (Deveraux, 1987), the
sequences were collected and formatted for use. Sequences that did not have
the same set of substrates were compared pairwise using bestfit or gap.
Larger groups were tested for sequence similarity using distances.
Sequences that displayed more than 30% identity of amino acid sequence
were automatically placed into the same initial group within the
superfamily. The program pileup was used to perform multiple sequence
alignments of the initial groups, and neighbor-joining analysis was

preformed to determine the most divergent members of the group. These

139

 

sequences were then used by blastp to search for similar sequences.
Sequences located by blastp were examined for similarity to the search
sequence over a minimum of 100 amino acids and for the conservation of
residues as evidenced by the members of the initial groups. The sequences
of the superfamily were increased as it became apparent that the non-heme
iron(II) oxygenases that require a reductant, usually ascorbate, are
significantly similar to some of the sequences that require a-ketoglutarate.
This 'similarity walking‘ was done for all of the groups, and resulted in the

addition of a number of sequences with unknown function.

Cluster analysis. When the similarity walking was completed, all of the
sequences for each group were aligned and conserved residues noted.
Considering the breadth of source organisms in several of the clusters, the
utilization of such a wide variety of substrates, and the diversity of
enzymatic functions that these enzymes catalyze, the retention of specific

residues among the group sequences is remarkable and quite characteristic.

NJ Analysis procedure for the a-ketoglutarate—dependent dioxygenases.
The multiple sequence alignments for the groups were used to determine the
phylogenetic relationships among members of each of the clusters. The
MEGA software package (Kamar, 1993)was used for analysis using the
Neighbor-Joining (NJ) method. Because the data was strictly protein
sequence, p-distance was used as the measure of genetic distance for
Neighbor-joining. All analyses were done under the following conditions:
NJ analysis, p-distance estimation, 500 replicate bootstrap, and complete

deletion option. A table of genetic distances and standard errors was

140

 

generated for each of the datasets. The trees were then edited to a
presentation format, and saved as text files, typically having the bootstrap

values at each node of the tree.

Intracluster analysis. Several of the sequences of each cluster, as well as
the sequences that did not fall into a cluster, were then aligned and analyzed
using the NJ method. The clusters and individual sequences presented in
this work are statistically distinct using this method. It must be noted,
however, this analysis is based on the entire amino acid sequence, and may
not distinguish the relatedness of specific regions or domains of the
sequences. This method also does not detect conservation of three-
dimensional structure. Sequences like deA and be are remarkably similar
in sequence (53% similarity), and may share similar conserved histidine

motifs.

141

 

APPENDIX B

 

APPENDIX B
Detailed Methods From Chapter 2.

4-AAP assay for deA-like enzymes:

The following assay reagents were prepared immediately before the
assay: 5 mM ascorbate, 10 mM a-ketoglutarate, 5 mM Fe”, 2% 4-
aminoantipyrene (4-AAP). Stock reagents include: 50 mM imidazole buffer
(pH 6.75), 10 mM 2,4-D, 0.5 M EDTA, Stop Buffer (pH 10), 8%
K3[Fe(CN)6]. The assays were started by mixing 480 pl H20, 200 pl
imidizole buffer, 100 1112,4-D, 10 ul ascorbate, 100 pl a-ketoglutarate, and
10 ul EDTA (EDTA in the negative control only). After vortexing, the
following was added: 10 ul Fe+2 and 100 111 cell lysate. The amount of cell
lysate may be adjusted, equilibrating the volumes with H20. Reaction
mixtures were incubated for 30 min at 30°C in heating block. After
incubation, the following was added 10 ul EDTA (to stop the reaction), 10
ul 4-AAP, 100 pl Stop Buffer, and 10 pl K3[Fe(CN)6]. Reactions were
allowed to sit at room temperature for 20 minutes. The absorbance at 501
nm was measured, centrifuging the reaction mix at top speed in a bench top
centrifuge for 1 minute if the reaction mix was cloudy. Cell lysates absorb
well at this wavelength, so a reference reaction with EDTA should be done

for each strain and/or differing volume of cell lysate..

4-NPAA cell free assav for TftLL:

Cells were grown in minimal inducing medium (M2, MC2, or MC +

IPTG), harvested by centrifugation at 10,000 x g for 10 min, and washed

142

 

once in 20 mM Tris-HCl buffer, pH 7.4. The cells were resuspended in a
small volume of buffer (1/20 to 1/50 of the original volume, minimum 1 ml)

to equivalent cell densities, chilled in an ice bath, and sonicated while being
sure to keep the lysate below 30°C at all times. The lysates were centrifuged

for 30 minutes at 100,000 x g. The resulting supernatants were used as the
cytosolic fractions of the lysates. Cell pellets were washed once, and
resuspended in 1 ml of the buffer. The production of 4-NP was monitored
continuously at 401 nm for reaction mixtures consisting of 1 mM or-
ketoglutarate, 50 11M ascorbate, 50 uM Fe(NI-I4)2(SO4)2, and 1 mM 4-NPAA
in 20 mM Tris-HCl, pH 7.4. The extinction coefficient of 4-NP at pH 7.4 is
17.8 mM'1 cm". The volume of lysate used in the reaction varied between
experimental runs, but the assay results were standardized by adjusting for

lysate protein concentration, as determined by the Lowry assay (15).

4—NPAA liquid media/microtiter plate assay for Tftlé;

Natural and engineered isolates were grown in M24 or M24+ IPTG
media, the cells were pelleted by centrifugation, and the amounts of 4-NP in
the supernatants were measured by absorbance at 401 nm. This method was
adapted for use in microtiter plates. Cells were grown in the wells of the

microtiter plates in appropriate inducing medium. 4-NPAA was added and
the cells were incubated at 30°C for 5 minutes to 4 hours. The cells were
pelleted by centrifugation of the microtiter plate, the supernatants were
transferred to a fresh microtiter plate, NaOH was added to a final

concentration of 0.1N, and the amounts of 4-NP in the supernatants were

detected using a microtiter plate reader measuring absorbance at 401 nm.

143

 

The concentrations of 4-NP were calculated using an experimentally

determined absorption coefficient of 18.4 mM‘1 cm'l.

4-NPAA agar plate colony assay for deA:

Petri plates containing colonies of the various isolates were prepared
by inoculating M2 agar plates (for those bacteria that will grow on 2,4-D as
a sole carbon source) or MC2 agar plates (for those that do not). The
cultures were diluted to approximately 200 CFU per plate, or picked onto a

grid of similar density. The plates were incubated at 30°C until small to

medium-sized colonies were observed. The colonies were assayed for deA
activity by carefully wetting them with freshly prepared assay solution (10

mM 4-NPAA, 50 uM Fe(NI-I4)2(SO4)2, and 50 0M ascorbate in 20 mM Tris-
HCl, pH 7.4). An atomizer or thin-layer chromatography sprayer was used
for the wetting, and this process was repeated once the excess surface liquid

had evaporated or been absorbed. Plates were incubated for 5 minutes to 4
hours at 30°C to allow for the production of 4-NP and color development. A

yellow ‘halo’ was observed around positive colonies.

4-NP utilization testing;
All of the strains that were found to convert 4-NPAA to 4-NP were
grown on M2 agar plates, inoculated into M24 liquid media, and grown

overnight. The cultures were inoculated into M2 medium + 0.250 mM 4-

NP, with shaking at 30°C for 48 hours, and 1 ml aliquots of each culture

were removed and prepared for HPLC as noted above, using a 0.250 M 4-

NP standard for reference.

144

 

APPENDIX C

 

APPENDIX C

Detailed Methods From Chapter 3.

Assay for Antibiotic Resistances:

Strains used in this assay were previously grown on M2 (Sassanella,
1997) media, and inoculated directly onto 1/10 Luria-Bertani medium
containing 16% agar and various concentrations of seven antibiotics (Table
3.1). Inoculated plates were then grown for 36 hours at 30°C and examined

for growth.

Auxotrophy Testing;

Strains tested were grown on 1/10 Luria-Bertani medium, then used
to inoculate four plates containing minimal salts (Fulthorpe, 1995) and 16%
noble agar plates that contained three or more of the following amino acid
mixtures: I - alanine, valine, cysteine, threonine, histidine, and methionine;
II - arginine, tyrosine, cysteine, serine, and leucine; III - phenylalanine,
tryptophan, glutamic acid, hydroxyproline, and isoleucine; IV - aspartic
acid, asparginine, glycine, proline, and lysine. If grth occurred on the
plate containing all four amino acid mixtures, the strain was tested with
several types of the same minimal medium containing a single carbon

source, such as lmM glucose or 1 mM citrate.

145

l6S rDNA Phylogenetic Analysis Methods:
Data file assembly. The E. coli, Rhodopseudomonas palustris, and M1

partial 16S sequences were obtained C. McGowan. GenBank 16S sequences
for E. coli. and all Rhodopseudomonas sp. were then stringsearched and
fetched. The sequences of Nitrobacter sp. were fetched because of the
reported closeness in genetic distance. Agrobacter tumefaciens,
Pseudomonas dimunata, and Erythrobacter longus were also obtained to
represent deeper branches within the same reported clade (Holt, 1994; m
Procaryotes). The sequences were then aligned using pileup, and edited
down to the 300 bp size of the bacterial variable region using lineup. This

process was repeated several times, including different taxa.

Phylogenetic Analysis of strain M1 and the established
Rhodopseudomonas taxa. The data files generated above were then opened
in MEGA, and analyzed by the Neighbor-Joining (NJ) method. Twenty-
three 16S rDNA sequences of reportedly related bacteria were included in a
NJ analysis. This analysis used the Jukes-Cantor method of determining
distance, which is more sensitive to varying rates of evolution in taxa, but
allows for confidence levels to be determined on the branch lengths
generated by NJ. The analysis was then repeated using Maximum
Parsimony (MP) analysis. The standard conditions for MP analysis included
the use of sequence gaps and additional state (which takes into
consideration insertions and deletions normally excluded by the NJ
method), the Branch and Bound Algorithm, and a 50% rule for creating a
consensus tree if more than one most parsimonious tree is created. MP

analysis is the best method for determining the general branching order of a

146

 

group of taxa. It does not determine branch lengths, and thus is insensitive
to differing rates of evolution among included taxa (which can lead to some

inaccuracies when using a breadth of bacterial species).

Confirmation of accuracy of the general branching pattern by using full
16S RNA sequences. Full-length 16S rDNA sequences were used to
confirm the general branching pattern found using the bacterial variable
region. NJ analysis clusters the Nitrobacter species with 87% accuracy. The
t-test for confidence was extremely high, with all values over 61 (5 of 8

over 94).

Transposon Mutagenesis Protocols:

All strains were streaked from glycerol stocks onto selective media.
Single colonies were grown in Luria-Bertani broth to an optical density of
0.3, and the strains were combined directly on Luria-Bertani agar plate in a
1:4 ratio of donor to recipient (l :1 :4 ratio for tri-parental matings). After a
16 h. incubation, the cells were scraped into minimal salts media, vortexed,
and plated on appropriate selective media. The pUT (deLorenzo, 1990) and
pSUP (Simon, 1983) mutagenesis systems required no additional helper
plasmid, but when required, pRK2013 (Figurski, 1979) was used for
mobilization. The mating of strain M1 and A. eutrophus strains was
enhanced by liquid mating variant. The log-phase donor and recipient cells
were mixed in a microfuge tube and allowed to incubate for 20 h, then

plated as above.

147

 

Chromosomal Plug Preparation for CHEF Gel Analysis:

Bacterial cells were grown overnight or longer in appropriate media
until good grth was observed. Cells may be pelleted at low speed (7,000x
g) and frozen at -20°C until needed. Pellets were thawed and resuspended to
a chromosomal concentration of ( lOpg/ml) in 10 mM Tris (pH 7.2)+20 mM
NaCl + 100mM EDTA. 1.2 ml of the cell resuspension was warmed to 70C,
well mixed with 1.2 ml of prewarmed 1.6% Biorad Ultrapure agarose in 1x
TBE, and aliquoted (100 pl) into the gel form, which was chilled to 4°C.
Once hardened, the plugs were removed fi'om the gel form using sterile,
non-metal tools, and placed in a sterile 50 ml Corning tube. Ten ml of the
lysing solution (10 mM Tris (pH7.2), 50 mM NaCl, 100 mM EDTA, 0.2%
SDS, 1% N-laurylsarcosine) was added to each tube, and they were gently
shaken for 90 minutes at 30°C. The plugs were washed twice in 5 ml Wash
Solution (20 mM Tris (pH 8.0), 50 mM EDTA), with a 10 minute
incubation shaking at room temperature for each wash, and treated with PK
Buffer (1 mg/ml proteinase K, 100 mM EDTA, 0.2% SDS, 1% N-
laurylsarcosine) for 18 h at 42°C. The plugs were washed once, and treated
with PSMF solution (lmM PSMF in wash solution) for 1 h with gentle
shaking. The plugs were washed twice in Wash Solution, then once in
Storage Solution (0.1X Wash Solution). Plugs were stored at 4C in storage

solution.

CHEF Gel Analysis:
A 0.8% Biorad Ultrapure agarose in 0.5x TBE buffer was cast.
Chromosomal plugs were introduced into the wells of the cell and sealed

with warm agarose. The gel was run in a Bio-Rad CHEF gel electrophoresis

148

 

unit using 0.5xTBA buffer at 14°C, for 22 hours at 200 mV and a 10 second

to 60 second ramp time.

Citations mentioned in this appendix are included in Chapter 3

References.

149

 

APPENDIX D

 

APPENDIX D

Detailed Methods From Chapter 4.

Strains used are described in Table 4.1. Citations included are

References as listed in Chapter 4.

Conjugal Matings and Transposon Mutagenesis:
Mating procedures are outlined in Appendix C. The

S17/pSUP2021[Tn5] mutagenesis system was used for random transposon
mutagenesis of pJP4 (deLorenzo, 1990). E. coli CB1360 (Sm’) carrying
pJP4 was mated with S-l7/pSUP2021 on LB agar overnight. The cells were
resuspended in minimal medium, plated on a selective media containing 100
pg/ml each streptomycin and kanamycin, and incubated 18 h at 37°C.
Colonies on these plates were then replicated onto a fresh selective plate
and onto an LB plate previously inoculated with log phase P. putida
PB2442. The selective plate was grown overnight at 37°C and stored at 4°C
as a master plate. The Alcaligenes mating plate was incubated at 30°C
overnight, then replicated twice onto minimal salts medium containing 20
mM benzoate and 100 g/ml kanamycin. Approximately 10,000 colonies
were then screened with the 4-NPAA petri plate assay (Chapter 2), and 100
colonies that appeared to have lost the ability to produce 4-NP fiom 4-
NPAA were retained. These colonies, taken from the untested replicate,
were assayed again the following day, after transfer to fresh minimal
benzoate with kanamycin plate. Eleven colonies were selected as having

lost the ability to convert 4-NPAA to 4-NP. These colonies were streaked

150

for isolation, then tested for conversion of 4-NPAA in liquid cultures over
24 hours. The colonies were also tested for 4-NPAA conversion in lysates
after grth in M2 medium supplemented with 0.1% casamino acids. Of the
eleven, one strain, m1 1, had lost the ability to convert 4-NPAA in whole
cells and lysates (presumably a tfdA insertion), and two strains, m8 and
m22, converted 4-NPAA in lysates but not in whole cells (indicating loss of
permeation). The P. putida strains were then mated into A. eutrophus

JMPZZZ by a broth culture mating (Appendix C).

4-NPAA Assay in Whole Cells:

Strains used in this assay were previously grown on M2 (Table 4.1)
media. Overnight broth cultures were used to inoculate fresh M2 media
supplemented with 100 mM 4-NPAA (pH 7.0). Cultures were then shaken
at 30°C for 24 hours. One ml of the culture was transferred to a microfuge
tube and the samples were centrifuged at 14,000 rpm in an Eppendorf
microfuge for 1 minute. Supernatants were transferred to a cuvette and 10 pl
of 10N NaOH added to enhance the sensitivity of the assay. Absorbances

were then taken at 401 nm.

Minitial Uptake A_ss;§yg

This assay was based on the benzoate permease assay of Thayer and
Wheelis (1976). Two ml of bacterial cells were grown overnight on 1 mM
2,4-D (or 1 mM casamino acids + 1 mM 2,4-D). Uninduced controls are
grown on a similar concentration of a non-aromatic amino acid or TCA
intermediate. Overnight cultures were added to 40 ml of the same medium

in 250 ml Erlenmeyer flasks and grown to mid-log phase. Cultures were

151

 

chilled to 4°C, pelleted by centrifugation at 7,000 x g for 10 minutes in pre-
weighed centrifuge tubes, and washed twice with 20 ml of 20 mM Tris pH
6.8 -7. The cells were resuspended to a concentration of 0.25 g cell/ml in the
same buffer at 4°C. Washed cells were incubated in a shaking water bath at
30°C for 10-15 minutes prior to the addition of 14C-2,4-D (~2.6 pCi/ml). For
initial uptake, 100 pl of 10 mM 2,4-D (1/10 ”0-2,4-0) was added to 5.0 ml
cells, each concentration done consecutively to allow for accurate sampling.
One ml samples were transferred onto pre-wetted 0.45 pm Millipore
nitrocellulose filters and immediately washed with 5 m1 of washing buffer
(same as above supplemented with 75 mM sodium azide and lmM 2,4-D) at
60, 120, 180, 240, and 300 second intervals. The filters were air dried,
immersed in a toluene based scintillation fluid, and counted in a scintillation

counter set for I4C detection.

2,4-D Incorporation Assay:

Bacterial cells were inoculated from glycerol stocks and streaked on
M2 (Table 2.1), if appropriate, or LB agar plates supplemented with
appropriate antibiotics and incubated until colonies were evident. The cells
were preconditioned by inoculating 10 ml liquid cultures of M2
supplemented with 20 mM citrate and incubation overnight or until the cells
had attained a clearly visible cell density. One ml samples of the
preconditioned cultures were measured for cell density by taking the
spectrophotometric A650. Using these readings, the initial inocula for the
next cultures were equalized. Aliquots of the same media were inoculated to
a total volume of 20 ml. The inoculated culture was well mixed and divided

into two cultures, forming two duplicate starter cultures. To one of the pair,

152

 

2.5 mCi (5 pl) of l4C-2,4-D was added. The cultures were then incubated at
30°C, shaking rapidly. At time 0, one ml of the non-labeled culture was
used to measure the cell density (A650). One ml of the radioactive culture
was filtered rapidly through a Millipore Nybond 0.45 pm filter using a
Millipore Vacuum manifold. The filter was washed 3 times with M2 media.
The filter was allowed to dry somewhat as the cell density of the non-
labeled culture was measured. The filters were removed, placed in
scintillation vials, and 5 ml toluene based scintillation fluid was added (for
counting and to stop any possible l4CO; evolution due to further
metabolism). The sampling procedure was repeated every fifieen minutes
for 150 minutes. The samples were then counted in a scintillation counter
using setting appropriate for 14C detection, and the results adjusted for

background and absorbance of 2,4-D to the membrane.

153

 

 

"11111111111110