19F NMR study on the biodegradation of fluorophenols by various Rhodococcus species

Article abstract of DOI:10.1023/A:1008391906885

Of all NMR observable isotopes ¹⁹F is the one perhaps most convenient for studies on biodegradation of environmental pollutants. THe reasons underlying this potential of ¹⁹F NMR are discussed are illustrated on the basis of a study on the biodegradation of fluorophenols by four Rhodococcus strains. The results indicate marked differences between the biodegradation pathways of fluorophenols among the various Rhodococcus species. This holds not only for the level and nature of the fluorinated biodegradation pathway intermediates that accumulate, but also for the regioselectivity of the initial hyroxylation step. Several of the Rhodococcus species contain a phenol hydroxylase that catalyses the oxidative defluorination of ortho- fluorinated di- and trifluorophenols. Furthermore it is illustrated how the ¹⁹F NMR technique can be used as a tool in the process of identification of an accumulated unknown metabolite, in this case most likely 5- fluoromaleylacetate. Altogether, the ¹⁹F NMR technique proved valid to obtain detailed information on the microbial biodegradation pathways of fluorinated organics, but also provide information on the specificity of enzymes generally considered unstable and, for this reason, not much studied so far.

Full text of DOI:10.1023/A:1008391906885

Biodegradation 9: 475–486, 1998.
© 1998 Kluwer Academic Publishers. Printed in the Netherlands.
475
19F NMR study on the biodegradation of fluorophenols by various
Rhodococcus species
Vladimir S. Bondar1,3, Marelle G. Boersma1, Eugene L. Golovlev2, Jacques Vervoort1, Willem
J.H. Van Berkel1, Zoya I. Finkelstein2, Inna P. Solyanikova2, Ludmila A. Golovleva2 & Ivonne
M.C.M. Rietjens1,∗
1 Department of Biomolecular Sciences, Laboratory of Biochemistry, Agricultural University, Dreijenlaan 3, 6703
HA Wageningen, The Netherlands; 2 Institute of Biochemistry and Physiology of Microorganisms, Russian Acad-
3
emy of Sciences, Pushchino, Russia; Institute of Biophysics, Russian Academy of Sciences, Siberian Branch,
660036 Krasnoyarsk, Russia (∗ author for correspondence)
Received 2 March 1998; accepted 15 July 1998
Key words: biodegradation, fluorophenols, 19F NMR, oxidative defluorination, Rhodococcus species
Abstract
Of all NMR observable isotopes 19F is the one perhaps most convenient for studies on biodegradation of environ-
mental pollutants. The reasons underlying this potential of 19F NMR are discussed and illustrated on the basis of a
study on the biodegradation of fluorophenols by four Rhodococcus strains. The results indicate marked differences
between the biodegradation pathways of fluorophenols among the various Rhodococcus species. This holds not
only for the level and nature of the fluorinated biodegradation pathway intermediates that accumulate, but also
for the regioselectivity of the initial hydroxylation step. Several of the Rhodococcus species contain a phenol
hydroxylase that catalyses the oxidative defluorination of ortho-fluorinated di- and trifluorophenols. Furthermore,
it is illustrated how the 19F NMR technique can be used as a tool in the process of identification of an accumulated
unknown metabolite, in this case most likely 5-fluoromaleylacetate. Altogether, the 19F NMR technique proved
valid to obtain detailed information on the microbial biodegradation pathways of fluorinated organics, but also to
provide information on the specificity of enzymes generally considered unstable and, for this reason, not much
studied so far.
Abbreviation: NMR – Nuclear Magnetic Resonance
Introduction
The objective of the present study is to illustrate the
use of 19F NMR as a technique to study the microbial
degradation of fluorinated environmental pollutants.
Of all NMR observable isotopes 19F is the one per-
haps most convenient for studies on biodegradation
of environmental pollutants (Malet-Martino & Mar-
tino 1989; Rietjens et al. 1993). This originates from
several advantages of the 19F isotope as compared to
other nuclei. First, the intrinsic sensitivity of the 19F
nucleus is high and almost comparable to that of the
1H nucleus. Sensitivity is an important issue, because
xenobiotics and their metabolites are usually present
The number of fluorine containing xenobiotics and
environmental contaminants has strongly increased
during the past decades (Key et al. 1997). A number
of fluorine containing drugs are currently in clini-
cal use, a large number of fluorinated compounds
are intermediates or end products in the synthesis
of industrial and agrochemicals, and many fluorine
containing biodegradation products result from the
chemical and/or microbial degradation of these fluo-
rinated chemicals (Banks & Lowe 1994; Banks 1995;
Walker 1990; Banks & Tatlow 1994; Edwards 1994).
MENNEN/SCHRIKS:DISK/CP: Pips Nr.:185419; Ordernr.:233022-mc (biodkap:bio2fam) v.1.1
biod533.tex; 13/01/1999; 14:06; p.1

476
in relatively low concentrations. Second, for fluorine
the sensitivity is further increased due to the absence
of background signals, because biological systems do
not contain 19F NMR visible fluorinated endogenous
compounds. This implies that all resonances observed
can be unambiguously ascribed to the xenobiotic com-
the corresponding fluorophenols using purified phe-
nol hydroxylase from Trichosporon cutaneum (Peelen
et al. 1995). Fluoromuconates were prepared and
identified as described previously (Boersma et al.,
1998) by incubating the fluorocatechols with catechol
1,2-dioxygenase from Pseudomonas arvilla C-1.
pound and its biodegradation products. Third, the 19
F
nucleus is known to have a broad chemical shift range
of about 500 ppm. This is large compared to the chem-
ical shift range of for example H (15 ppm) and that
Isolation and growth of strains
1
The conversions of a series of ortho-fluorinated di- and
trifluorophenols by four different strains of Rhodococ-
cus were investigated. These strains were originally
isolated while searching for bacteria able to degrade
crude oil, but showed also capable of growing on
phenols.
of 13C (250 ppm). The chemical shift of a 19F nucleus
is highly sensitive to its molecular surroundings re-
sulting in widespread changes in chemical shifts upon
biotransformation of fluorinated organic compounds,
thereby reducing the chances of peak overlap.
In the framework of a project devoted to the
biodegradation of halophenols by gram-positive bac-
teria, the objective of the present study was to use the
19F NMR technique to investigate the biodegradation
pathways of fluorophenols in several Rhodococcus
strains. Special emphasis was on the first step in the
catabolic pathways, namely the conversion of fluo-
rophenols to fluorocatechols. Generally, in Rhodococ-
cus species these studies are limited by the instability
and/or poor activity of the phenol hydroxylase(s) in-
volved (Straube 1987; Janke et al. 1988). However, the
19F NMR technique provides the unique opportunity
to study the activity of the phenol hydroxylasesin vivo.
Thus, the conversions of a series of ortho fluori-
nated di- and trifluorophenols by four different strains
of Rhodococcus were investigated. These strains were
originally isolated in the framework of a project
searching for bacteria able to degrade crude oil, but
showed also capable of growing on phenols. All flu-
orophenol derivatives studied contain a fluorinated
C2 position and a non-fluorinated C6 position. This
provides the possibility to detect unambiguously the
regioselectivity of the initial hydroxylation step catal-
ysed by the phenol hydroxylase.
The strain Rhodococcus opacus 1G was isolated
by selective enrichment from the soil of oil fields in
the Volga river valley near Samara. Chemostat cultiva-
tion with liquid paraffins as carbon source was used
for ◦isolation of the strain. The cells were grown at
28 C on an orbital shaker (250 rpm) in a synthetic
medium containing per liter: NH4NO3, 1g; K2HPO4,
1g; KH2PO4, 1g; MgSO4∗7H2O, 0.2 g; CaCl2, 0.02 g;
FeCl3, 2 drops of a saturated solution; pH 7.2. Phenol,
used as the sole carbon and energy source, was added
several times at 1.0 mM (final concentration) over a
◦
period of 24 h. The strain was maintained at 4 C on
agar slants of the same medium containing phenol (0.2
g/l). Rhodococcus corallinus 135 was isolated from
soil contaminated with diesel oil from an abandoned
military base. The strain Rhodococcus strain 89 was
isolated from sand soil of pine forest in the Volga
river valley near Nignyi Novgorod. For isolation of
both strains a continuous enrichment culture technique
was applied using n-alkanes as a sole source of car-
◦
bon. The strains were grown at 28 C on a synthetic
medium containing per liter: NH4NO3, 1g; K2HPO4,
1g; KH2PO4, 1g; MgSO4∗7H2O, 0.2 g; CaCl2, 0.02 g;
FeCl3, 2 drops of a saturated solution; pH 7.2. Phenol,
used as the substrate, was added several times up to
3.2 mM (total concentration) over a period of 24 h.
The strain was maintained at 4 ◦C on agar slants of the
same medium containing phenol (0.2 g/l).
Materials and methods
Rhodococcus erythropolis 1CP was isolated as de-
scribed previously (Gorlatov et al. 1989). The strain
was grown at 28 C on a synthetic medium contain-
ing per liter: NH4NO3, 1g; K2HPO4, 1g; KH2PO4,
1g; MgSO4∗7H2O, 0.2 g; CaCl2, 0.02 g; FeCl3, 2
drops of a saturated solution; pH 7.2. For growth, 4-
chlorophenol, proven to be the best substrate for Rh.
erythropolis 1CP, was used as the sole carbon and en-
Chemicals
◦
Phenol was purchased from Merck (Darmstadt, Ger-
many). 2,4-Difluoro- and 2,5-difluorophenol were
purchased from Aldrich (Steinheim, Germany). 2,3-
Difluoro-, 2,3,4-trifluoro-, 2,3,5-trifluoro- and 2,4,5-
trifluorophenol were obtained from Fluorochem (Der-
byshire, UK). Fluorocatechols were prepared from
biod533.tex; 13/01/1999; 14:06; p.2

477
ergy source. The substrate was added several times up
to 3.2 mM (total concentration) over a period of 24 h.
The strain was maintained at 4 ◦C on agar slants of the
same medium containing 4-chlorophenol (0.2 g/l).
with phenol as final concentration of phenol hydroxy-
lase and 7.2 × 10−3 U/ml with catechol as final con-
centration of catechol 1,2-dioxygenase). The sample
thus obtained was analysed by 19F NMR, showing sig-
nificant formation of 2,3-difluoromuconate, and then
used for another incubation with purified chloromu-
conate cycloisomerase. Upon addition of a catalytic
amount of chloromuconate cycloisomerase (9.6 ×
10−3 U/ml with 2-chloromuconate as final concen-
tration) the conversion of 2,3-difluoromuconate was
followed in time by recording 19F NMR spectra while
incubating the sample in the NMR spectrometer at
30 ◦C.
Incubations with whole cells
Cells were harvested by centrifugation (6500 g) and
washed three times in medium without phenol. To test
biodegradation of the fluorophenol derivatives, cells
were resuspended in medium without phenol contain-
ing per liter: peptone, 10 g; yeast extract, 5 g; NaCl,
8 g; pH 7.4. The optical density at 600 nm of this
cell suspension was 1.8-2.1. To 80 ml of this cul-
ture the fluorophenol to be tested was added to a
final concentration of 0.7 mM and the cultures were
incubated at 28 ◦C on an orbital shaker. Although
the detection limit of the 19F NMR measurements is
1 µm, the substrate was added in much higher con-
centration in order to provide optimal possibilities for
detection of intermediates. To prevent autoxidation of
the fluorocatechols formed, 2 mM ascorbic acid (final
concentration) was added at the start of the incuba-
tion followed by 1 mM every next hour. To monitor
the rate and characteristics of the conversion of the
fluorophenols, samples were taken at one hour time
intervals for 6 hours. Before freezing the samples into
liquid nitrogen, another 10 mM of ascorbic acid was
added. Samples were stored at −20 ◦C until analysed.
Before 19F NMR analysis, samples were thawed and
centrifuged (5 min 13,000 g at 0 ◦C).
19F NMR measurements
19F NMR measurements were performed on a Bruker
AMX 300 and a Bruker DPX 400 NMR spectrometer
as described previously (Vervoort et al.◦1990: Boersma
et al. 1998). The temperature was 7 C unless indi-
cated otherwise. A dedicated 10 mM 19F NMR probe-
head was used at both instruments. The spectral width
for the 19F NMR measurements was between 20,000
and 50,000 Hz depending on the sample of interest.
The number of datapoints used for data acquisition
was 65536. Pulse angles of 30◦ were used. Between
2,000 and 60,000 scans were recorded, depending
on the concentrations of the fluorine-containing com-
pounds and the signal to noise ratio required. The
detection limit of an overnight run (60,000 scans) is
1 µM. The sample volume was 1.71 ml, containing
1.4 ml incubation mixture, 200 µl 0.8 M potassium
phosphate pH 7.6, 100 µl of 2H2O, used as deuterium
lock and 10 µl of 8.4 mM 4-fluorobenzoate added
as an internal standard. Concentrations of the various
metabolites were calculated by comparison of the inte-
grals of the 19F NMR resonances of the metabolites to
the integral of the 4-fluorobenzoate resonance. Chem-
ical shifts are reported relative to CFCl3. 19F NMR
chemical shift values of the various fluorine containing
compounds were identified using authentic reference
compounds for fluoride anions and all fluorophenols.
The resonances of the different fluorocatechol and
fluoromuconate metabolites have been identified and
reported previously (Peelen et al. 1995; Boersma et
al. 1998). 1H decoupling was achieved with a Waltz16
decoupling sequence.
Incubations with purified enzymes
Phenol hydroxylase from Trichosporon cutaneum was
purified essentially as described by Sejlitz and Neujahr
(1987). Catechol 1,2-dioxygenase from Pseudomonas
arvilla C-1 was purified as described by Nakai et
al. (1988). Chloromuconate cycloisomerase from
Rhodococcus erythropolis 1CP was purified essen-
tially as reported elsewhere (Solyanikova et al. 1995).
Incubation of 2,3-difluorophenol with phenol hydrox-
ylase and catechol 1,2-dioxygenase was performed at
30 ◦C in closed reaction vessels to prevent evaporation
of the phenolic substrate. Incubations contained (final
concentrations) 0.1 M potassium phosphate pH 7.6,
0.7 mM of 2,3-difluorophenol, added as 1% (v/v) of a
70 mM stock solution in dimethylsulphoxide, 10 µM
FAD, 1 mM ascorbic acid and 1 mM NADPH. The
incubations were started by the addition of catalytic
amounts of the purified enzymes (5.3 × 10−3 U/ml
biod533.tex; 13/01/1999; 14:06; p.3

478
19
Figure 1. Representative F NMR spectra showing metabolite pro-
files of a) 2,3-difluorophenol and b) 2,3,4-trifluorophenol, by Rh.
opacus 1G. Spectra presented were those obtained at t = 2 h. Reso-
nances were identified as previously described (Peelen et al. 1995;
Boersma et al. 1998). The resonance marked IS is from the internal
standard 4-fluorobenzoate. The resonances marked with an asteriks
indicate unidentified metabolite signals.
19
Figure 2. Representative F NMR spectra showing metabolite pro-
files of a) 2,3-difluorophenol and b) 2,3,4-trifluorophenol, by Rh.
corallinus 135. Spectra presented were those obtained at t = 2 h.
Resonances were identified as described in the legend of Figure 1.
of the different Rhodococcus species with the various
fluorophenols.
Results
All Rhodococcus species appeared able to convert
the fluorophenols, although at a different rate. Based
on the results presented (Table 1), the overall extent of
fluorophenol conversion decreased for the four species
in the order Rh. corallinus 135 > Rh. opacus 1G > Rh.
erythropolis 1CP > Rh. strain 89.
Fluorophenol biodegradation and fluoride anion
formation as determined by 19F NMR
Figures 1–4 present some representative 19F NMR
spectra showing the metabolite patterns of the ortho-
fluorinated di- and trifluorophenols by different
Rhodococcus strains. First of all, these spectra illus-
trate that the 19F NMR analysis provides a way to
characterise and quantify metabolite profiles. The po-
sitions of the peaks indicate which metabolites are
present, the peak areas reflect the amount of the
various metabolites, and the number of fluorine sig-
nals belonging to one metabolite indicate the number
of different type of fluorine substituents present in
the molecule. From these and additional spectra the
metabolite profiles of the fluorophenols in time by the
different Rhodococcus species could be quantified. Ta-
ble 1 presents the percentage of fluorinated substrate
left, the percentage of fluorine recovered as fluoride
anions, and the percentage of fluorinated metabolic
intermediates observed, all at t = 2 h of incubation
Furthermore, there is no consistent order in the
preference of the Rhococcus species for the different
fluorophenols. For Rh. opacus 1G, the order in the
rate of conversion of the fluorophenols is 2,5-difluoro-
> 2,3,4-trifluoro- > 2,4-difluoro- > 2,3-difluoro- >
2,3,5-trifluoro- = 2,4,5-trifluorophenol (Table 1). Rh.
corallinus 135 converts all fluorophenols extremely
efficient, resulting in full elimination of 5 out of 6
fluorophenols upon 2 to 6 h of incubation (19F NMR
spectra not shown). From the 19F NMR spectra ob-
tained at 2 h (Table 1) the following order for the
rate of degradation of the various fluorophenols by
Rh. corallinus 135 can be derived: 2,5-difluoro- >
2,3,4-trifluoro- > 2,4,5-trifluoro- > 2,3-difluoro- >
2,4-difluoro- > 2,3,5-trifluorophenol. In addition, the
data presented in Table 1 indicate that for Rh. strain
89 the fluorophenols are degraded with rates decreas-
biod533.tex; 13/01/1999; 14:06; p.4

479
19
19
Figure 3. Representative F NMR spectra showing metabolite
Figure 4. Representative F NMR spectra showing metabolite pro-
files of a) 2,3-difluorophenol and b) 2,3,4-trifluorophenol, by Rh.
erythropolis 1CP. Spectra presented were those obtained at t = 4h.
Resonances were identified as described in the legend of Figure 1.
profiles of a) 2,3-difluorophenol and b) 2,3,4-trifluorophenol, by
Rh. strain 89. Spectra presented were those obtained at t = 6 h.
Resonances were identified as described in the legend of Figure 1.
ing in the order 2,3-difluoro- = 2,4-difluoro- > 2,5-
difluoro- > 2,3,4-trifluoro- > 2,3,5-trifluoro- = 2,4,5-
trifluorophenol. For Rh. erythropolis 1CP the rates
decrease in the order 2,5-difluoro- > 2,3,5-trifluoro- =
2,4-difluoro> 2,3-difluoro-> 2,4,5-trifluoro- > 2,3,4-
trifluorophenol. Thus, generally there is a tendency
to convert the difluorophenols with rates higher or
about equal to the rates for conversion of the triflu-
orophenols. This is in line with the expected effect
of fluorine substituents, decreasing the nucleophilic
reactivity of the π-electrons in the fluorophenol, a
characteristic that is of importance for their rate of
conversion by phenol hydroxylase from Trichosporon
cutaneum (Peelen et al. 1995).
verted, the amount of accumulated intermediates starts
to decrease, accompanied by, as expected, a further
increase in fluoride anions.
Thus, the difference between the amount of fluo-
rophenol degraded and the amount of fluoride anions
formed (Table 1) results from the accumulation of
fluorinated intermediates in the biodegradation path-
ways. From the 19F NMR spectra presented in Fig-
ures 1–4 it can be derived that the fluorinated biodegra-
dation metabolites which accumulate are mainly flu-
orocatechols and fluoromuconates. These metabolites
result from ortho-hydroxylation of the fluorophenols
by a phenol hydroxylase, followed by intradiol ring
cleavage of the fluorocatechols to fluoromuconates by
a catechol 1,2-dioxygenase. Other fluorinated inter-
mediate metabolites generally do not accumulate to a
significant extent, except from the conversion of fluo-
rophenols by Rh. erythropolis 1CP (Figure 4). Table 2
presents the relative contribution of the different fluo-
rinated catechols, muconates and unidentified metabo-
lites, to the total amount of fluorine containing inter-
mediates observed. From these data the ratio between
the two possible fluorocatechol metabolites, resulting
from hydroxylation of the fluorophenols at respec-
tively their fluorinated C2 and their non-fluorinated C6
Fluorophenol biodegradation patterns as determined
by 19F NMR
Table 1 and Figures 1–4 also illustrate that the
19F NMR spectra provide insight in the fluorinated
biodegradation pathway intermediates formed upon
the conversion of the different fluorophenols. Dur-
ing the first hours of the incubations and as long
as a residual amount of the parent fluorophenol is
present, the relative amount of fluorinated intermedi-
ates is constant. Only when all fluorophenol is con-
biod533.tex; 13/01/1999; 14:06; p.5

480
Table 1. Relative fluorophenol loss, fluoride anion release and accumulation of fluorinated
intermediates in incubations of Rhodococcus species with different ortho-fluorinated di- and triflu-
orophenols. The incubation time was t = 2 h. The standard deviation was less than 10%. In abiotic
controls no reaction took place
fluorinated
substrate
Rhodococcus
opacus 1G
%
Rhodococcus
Rhodococcus
Rhodococcus
erythropolis 1CP
%
corallinus 135 strain 89
%
%
2,3-difluorophenol
Phenol left
55
26
19
26
38
36
74
9
74
16
10
−
F
release
F in intermediates
17
2,4-difluorophenol
Phenol left
49
42
9
46
53
1
75
11
14
66
32
2
−
F
release
F in intermediates
2,5-difluorophenol
Phenol left
30
57
13
0
90
10
82
3
61
35
4
−
F
release
F in intermediates
15
2,3,4-trifluorophenol
Phenol left
43
31
26
12
35
53
90
3
80
13
7
−
F
release
F in intermediates
7
2,3,5-trifluorophenol
Phenol left
63
23
14
59
38
3
94
4
66
34
0
−
F
release
F in intermediates
2
2,4,5-trifluorophenol
Phenol left
65
28
7
19
79
2
94
5
79
21
0
−
F
release
F in intermediates
1
position, can be derived, as well as the ratio between
the two corresponding muconate metabolites. The
hydroxylationat the fluorinated C2 position of the sub-
strates represents a so called oxidative or oxygenolytic
dehalogenation in which the insertion of an oxygen
atom derived from molecular oxygen is accompanied
by defluorination.
The results obtained and quantified in Table 2
indicate marked differences between the biodegrada-
tion pathways of the fluorophenols by the different
Rhodococcus species. This holds not only for the
extent of accumulation of fluorinated biodegradation
pathway intermediates, but also for the nature of these
intermediates.
metabolites observed, the major part appear to be flu-
orocatechols (Figure 1, Table 2). This points at a rate
limiting role for the catechol 1,2-dioxygenase of Rh.
opacus 1G in the biodegradation of the fluorophenols.
The ratio of C2: C6 hydroxylation of the various flu-
orophenols suggests that the phenol hydroxylase of
Rh. opacus 1G favours formation of the defluorinated
catechol metabolite, i.e. prefers to catalyse oxidative
dehalogenation at the fluorinated C2 of the fluorophe-
nol instead of hydroxylation at the non-fluorinated C6
position.
For Rh. corallinus 135 there is no consistency in
the nature and level of the intermediates that tend
to accumulate. For 2,3-difluorophenol, catechols as
well as muconates accumulate, for 2,3,4-trifluoro- and
2,3,5-trifluorophenol mainly fluorocatechols accumu-
late, for 2,5-difluorophenolaccumulation of especially
For Rh. opacus 1G, incubations with fluorophenols
result in accumulation of especially the correspond-
ing fluorocatechols. Of all fluorinated intermediate
biod533.tex; 13/01/1999; 14:06; p.6

481
Table 2. Relative composition of the metabolite profiles of the various Rhodococcus species incu-
19
bated with ortho-fluorinated di- and trifluorophenols as detected by F NMR. The total amount of
fluorinated products other than the parent phenol and fluoride anions was taken as 100%. Values
presented are the mean of the data obtained at t = 1, 2, 4 and 6 h (n = 4), except for the conversion
of 2,3-difluoro-, 2,5-difluoro- and 2,3,4-trifluorophenol by Rh. corallinus 135, and of 2,3-difluoro-,
2,4-difluoro-, 2,5-difluoro- and 2,3,4-trifluorophenol by R. opacus 1G, for which the mean of t = 1 and
2 h is presented (n = 2) because the amount of phenol was fully degraded at 4 and 6 h. For incubations
where the amount of intermediates was < 2% of the total amount of fluorinated metabolites (Table 1),
the metabolite profile is not specified
Fluorinated
substrate
Rhodococcus
opacus 1G
%
Rhodococcus
corallinus 135
%
Rhodococcus
strain 89
%
Rhodococcus
erythropolis 1CP
%
metabolite
2,3-difluorophenol
3-fluorocatechol
81 ± 2
16 ± 1
1 ± 1
0
50 ± 3
23 ± 5
24 ± 2
1 ± 1
5 ± 3
87 ± 6
0
11 ± 3
3,4-difluorocatechol
2-fluoromuconate
2,3-difluoromuconate
unknown
0
58 ± 4
5 ± 3
3 ± 2
0
∗
2 ± 1
2 ± 1
31 ± 2
2,4-difluorophenol
4-fluorocatechol
47 ± 30
–
–
–
–
–
5 ± 3
94 ± 3
0
–
–
–
–
–
3,5-difluorocatechol
3-fluoromuconate
2,4-difluoromuconate
Unknown
16 ± 8
0
0
1 ± 1
0
∗
37 ± 37
2,5-difluorophenol
4-fluorocatechol
64 ± 9
9 ± 9
12 ± 12
0
0
1 ± 1
2 ± 1
0
3,6-difluorocatechol
3-fluoromuconate
2,5-difluoromuconate
Unknown
0
0
0
47 ± 5
0
74 ± 17
5 ± 5
53 ± 5
17 ± 6
∗
∗
36 ± 9
0
80 ± 7
2,3,4-trifluorophenol
3,4-difluorocatechol
3,4,5-trifluorocatechol
2,3-difluoromuconate
2,3,4-trifluoromuconate
Unknown
81 ± 1
79 ± 1
21 ± 2
1 ± 1
0
0
7 ± 4
3 ± 2
0
19 ± 1
100 ± 0
0
0
0
0
0
0
39 ± 8
∗
0
51 ± 2
2,3,5-trifluorophenol
3,5-difluorocatechol
3,4,6-trifluorocatechol
2,4-difluoromuconate
2,3,5-trifluoromuconate
Unknown
98 ± 1
78 ± 5
2 ± 2
1 ± 1
5 ± 5
14 ± 9
–
–
–
–
–
–
–
–
–
–
0
0
0
2 ± 1
2,4,5-trifluorophenol
4,5-difluorocatechol
3,4,6-trifluorocatechol
3,4-difluoromuconate
2,3,5-trifluoromuconate
Unknown
95 ± 2
–
–
–
–
–
–
–
–
–
–
–
–
–
–
–
0
0
0
5 ± 2
∗ For 2,3-difluorophenol and 2,3,4-trifluorophenol these were a major metabolite at −189.9 ppm and
a minor one at −196.6 ppm (see Figure 4), for 2,4-difluorophenol and 2,5-difluorophenol these were
a major metabolite at −90.3 ppm, and minor metabolites with resonances at −100.3 and −113.2 ppm
19
(for 2,4-difluorophenol) and −126.1 and −113.2 ppm (for 2,5-difluorophenol)( F NMR spectra not
shown).
biod533.tex; 13/01/1999; 14:06; p.7

482
the fluoromuconates is observed, whereas for 2,4-
difluoro- and 2,4,5-trifluorophenol no significant ac-
cumulation of fluorinated intermediates is observed
(Figure 2, Table 1 and 2). This variation in the na-
ture and level of the accumulated metabolites does not
result from different rates of conversion of the par-
ent fluorophenol. This follows for example from the
observation that 2,3,4-trifluorophenol is converted rel-
atively rapidly, leaving no parent fluorophenol upon 4
h of incubation (19F NMR data not shown), whereas
even at t = 6 h the fluorinated catechols still make up
73% of all intermediates observed (19F NMR data not
shown). Furthermore, the nature of the accumulated
metabolites is not influenced primarily by the type
of fluorocatechol formed, since from 2,4-difluoro-
and 2,3,5-trifluorophenol both 3,5-difluorocatechol is
formed, whereas only for 2,3,5-trifluorophenol pref-
erential accumulation of this 3,5-difluorocatechol is
observed. Most likely the phenomenon is dependent
on the as yet not characterised effect of the fluorine
substituent pattern on both the activity of phenol hy-
droxylase as well as on the activity of the catechol
1,2-dioxygenase from Rh. corallinus 135. For Rh.
corallinus 135 also the preference of its phenol hy-
droxylase for C2 as compared to C6 hydroxylation
varies with the fluorine substituent pattern of the flu-
orophenol, although in most cases, as for the phenol
hydroxylase from Rh. opacus 1G, oxidative dehalo-
genation at C2 is favoured over hydroxylation at C6
(Table 2). There is some tendency that when oxida-
tive defluorination in the first step is favoured over
hydroxylation at C6, such as in the case of 2,3,4-
trifluoro-, 2,3,5-trifluorophenol and, to a lesser extent,
2,3-difluorophenol, accumulation of the catechols is
observed. When hydroxylation at C6 of the fluorophe-
nol is preferred, such as for 2,5-difluorophenol, accu-
mulation of the muconates and not of the catechols is
observed. Together these data suggest that the relative
activities of phenol hydroxylase and catechol 1,2-
dioxygenase of Rh. corallinus 135 are affected by the
nature of the fluorine substituent pattern in their sub-
strates. The data also suggest that the in situ activity of
phenol hydroxylase and catechol 1,2-dioxygenase of
Rh. corallinus 135 are in the same order of magnitude,
resulting in the possibility for an influence of the fluo-
rine substituent pattern on the nature of the metabolites
that accumulate. From the fact that fluorocatechols and
fluoromuconates are the only intermediates that accu-
mulate, it is concluded that the rate limiting step in the
biodegradation pathway of the fluorophenols by Rh.
corallinus 135 is related to the enzyme catalysing the
conversion of the fluoromuconates.
For Rh. strain 89, in all cases where significant
metabolite accumulation was observed, except for
2,5-difluorophenol, preferential accumulation of the
fluorocatechols is observed, pointing at a rate limiting
role in the biodegradation pathway for the catechol
1,2-dioxygenase. The regioselectivity of the phenol
hydroxylase from Rh. strain 89 is such that it catalyses
preferential hydroxylation at C6 instead of catalysing
oxidative defluorination at C2 (Figure 3, Table 2).
The metabolite profiles observed for Rh. erythro-
polis 1CP, differ from those of the other Rhodococcus
species, especially because accumulation of as yet not
identified fluorinated intermediates is observed to a
relatively high extent (Figure 4, Table 2). Incubations
with 2,3-difluorophenol and 2,3,4-trifluorophenol give
rise to a major unidentified metabolite with a reso-
nance peak at −189.9 ppm (Figure 4). In incubations
with 2,4-difluorophenol and 2,5-difluorophenol accu-
mulation of a major unidentified metabolite with its
resonance at −90.3 ppm is observed (Table 2, 19F
NMR spectra not shown). Furthermore, in the incu-
bations with Rh. erythropolis 1CP, fluoromuconates
accumulate to a higher extent than the corresponding
fluorocatechols. All together the metabolite profiles
obtained for Rh. erythropolis 1CP indicate that the
rate limiting step in the fluorophenol degradation oc-
curs after the steps catalysed by phenol hydroxylase,
and catechol 1,2-dioxygenase. The regioselectivity of
the reaction catalysed by phenol hydroxylase from Rh.
erythropolis 1CP is dependent on the substituent pat-
tern in the fluorophenol. In analogy with the phenol
hydroxylase from Rh. corallinus 135 and that of Rh.
opacus 1G, the enzyme from Rh. erythropolis 1CP is
able to catalyse oxidative defluorination, for example
of 2,3-difluorophenol (Table 2).
Use of 19F NMR in elucidation of the nature of
unidentified metabolites
The 19F NMR spectra of especially Rh. erythropo-
lis 1CP incubated with either 2,3-difluorophenol or
2,3,4-trifluorophenol showed accumulation of an as
yet unidentified major intermediate metabolite with
its resonance at −189.9 ppm (Figure 4). Close in-
spection of the 19F NMR spectra of incubations of
the other Rhodococcus species of the present study
reveals the presence of a very small amount of this
metabolite also in incubations of Rh. corallinus 135
(Figure 2) and Rh. strain 89 (Figure 3) with 2,3-
biod533.tex; 13/01/1999; 14:06; p.8

483
Figure 6. Schematic presentation of the possible conversion of
2,3-difluoromuconate by chloromuconate cycloisomerase, based
on the conversion of 2-chloromuconate and 3-chloromuconate by
muconate cycloisomerase and chloromuconate cycloisomerase.
19
Figure 5. Time-dependent F NMR spectra of the incubation of
−189.9 ppm, also observed in the incubations with
whole cells (Figure 4), is readily formed upon the
addition of chloromuconate cycloisomerase to 2,3-
difluoromuconate. The fact that the metabolite shows
only one peak in the 19F NMR spectrum indicates that
this compound contains only one (type of) fluorine
substituent.
2,3-difluoromuconate with chloromuconate cycloisomerase from
Rh. erythropolis 1CP. Spectra were obtained at a) t = 0, b) t = 30
min, c) t = 1.5h and d) t = 6 h. Metabolites were assigned on the
basis of arguments presented in the text. The resonance marked IS is
from the internal standard 4-fluorobenzoate. The resonances marked
with an asteriks indicate unidentified signals mostly already present
at t = 0.
Furthermore, the spectra in Figure 5b and 5c show
the formation of another metabolite upon the incu-
bation of 2,3-difluoromuconate with chloromuconate
cycloisomerase. This metabolite contains two fluorine
substituents as reflected by two peaks of identical in-
tensity at −127.2 and −195.3 ppm. This metabolite
seems to be formed initially to a larger extent than
the metabolite at −189.9 ppm, but disappears in time,
giving rise to an increase in the metabolite with its res-
onance at −189.9 ppm and significant fluoride anion
formation. Based on literature data on the conver-
sion of 2-chloro-, 3-chloro- and 3-fluoromuconate by
(chloro)muconate cycloisomerases (Harper & Blak-
ley 1971; Schmidt & Knackmuss 1980; Schreiber
difluoro- and 2,3,4-trifluorophenol. In order to further
illustrate the potential of the 19F NMR technique in
studies on biodegradation, additional 19F NMR exper-
iments were performed to obtain information on the
nature of this major metabolite.
Figure 5a shows the 19F NMR spectrum of
2,3-difluoromuconate prepared by incubation of 2,3-
difluorophenol with purified phenol hydroxylase and
catechol 1,2-dioxygenase. Upon addition to this sam-
ple of chloromuconate cycloisomerase, purified from
Rh. erythropolis 1CP, the 19F NMR spectra presented
in Figure 5b–d were obtained in time. Figure 5b–
d show that the metabolite with its resonance at
biod533.tex; 13/01/1999; 14:06; p.9

484
et al. 1980; Schlömann et al. 1990b; Vollmer et
al. 1994; Vollmer and Schlömann 1995; Prucha et
al. 1996; Blasco et al. 1995; Kaschabek & Reineke
1995), possible pathways for the biodehalogenation
of a 2,3-dihalomuconate by chloromuconate cycloi-
somerase can be put forward. Figure 6 schematically
presents these possible pathways. Cyclisation of 2,3-
difluoromuconate to 2,3-difluoromuconolactone can
not lead to dehalogenation. The other possible prod-
uct resulting from cyclisation of 2,3-difluoromuconate
would be 4,5-difluoromuconolactone. Defluorination
of 4,5-difluoromuconolactone can result in forma-
tion of 5-fluorodienelactone (Figure 6). This reac-
tion would be analogous to the dechlorination of 3-
chloromuconate proceeding by its initial conversion to
4-chloromuconolactone catalysed by chloromuconate
cycloisomerase, followed by chloride elimination re-
sulting in the dienelactone. The 5-fluorodienelactone
formed may be sensitive to chemical hydrolysis lead-
ing to formation of 5-fluoromaleylacetate in either
its enol or keto form (Kaschabek & Reineke 1995)
(Figure 6). Alternatively, hydrolysis of the 4,5-
difluoromuconolactone may occur prior to fluoride
elimination in a pathway also leading to formation of
5-fluoromaleylacetate (Figure 6). This pathway would
be analogous to the one reported for defluorination
of 4-fluoromuconolactone (Harper and Blakley 1971;
Schlömann et al. 1990b). Elimination of the C5-
fluorine substituent from 4,5-difluoromuconolactone,
analogous to the dechlorination of 2-chloromuconate
through 5-chloromuconolactone, is hampered by
the fluorine substituent at C4. Finally, elimina-
tion of a fluoride anion with concomitant decar-
boxylation of 4,5-difluoromuconolactone, to give 5-
fluoroprotoanemonin, can be proposed (Figure 6),
analogous to the conversion of 3-chloromuconate
through 4-chloromuconolactone to protoanemonin
(Blasco et al. 1995). This route, however, seems less
likely, since the second step in this reaction path-
way has been reported to be catalysed by muconate
cycloisomerase, not chloromuconate cycloisomerase
(Blasco et al. 1995).
when the hydrogen and fluorine atoms are bound to an
sp3 type carbon, but between 70–90 Hz when bound
to an sp2 type carbon (Wray 1983). This implies
that the JFH coupling constant of 50.4 Hz, combined
with the various options for the monofluorinated re-
action product formed from 2,3-difluoromuconate by
chloromuconate cycloisomerase (Figure 6), strongly
argue in favour of identification of the metabolite with
the resonance at −189.9 ppm as the keto form of
5-fluoromaleylacetate.
The metabolite in Figure 5b–d with 19F NMR res-
onances at −127.2 and −195.3 ppm both showing a
JFF of 16.6 Hz, could then be ascribed to either one of
the two possible difluorinated intermediates depicted
in Figure 6. This difluorinated intermediate is not ob-
served in the incubations with whole cells (Figures
1–4) and was therefore not identified further at this
stage.
Discussion
In the present study, the biodegradation of ortho-
fluorinated di- and trifluorophenols by different strains
of Rhodococcus was analysed by 19F NMR. The re-
sults obtained illustrate that 19F NMR is an efficient
technique to gain insight in the biodegradation of fluo-
rinated environmental pollutants. Marked differences
between the various Rhodococcus strains were ob-
served especially with respect to: 1) their ability and
rates for biodegradation of the fluorophenols, 2) the
nature of the rate limiting step in the biodegradation
pathways, and, as a result, 3) the nature of the fluo-
rinated metabolites that accumulate, 4) the effect of
the fluorine substituent pattern on the biodegradation
pathway and the enzymes involved and 5) the regiose-
lectivity of the first step in the degradation pathway
catalysed by phenol hydroxylases which, in several
cases, appear to favour the catalysis of oxidative deflu-
orination over hydroxylation at a nonfluorinated ortho
position.
Phenol hydroxylase from Rh. opacus 1G favours
oxidative defluorination of all fluorophenols tested,
whereas the regioselectivity of the phenol hydroxylase
from Rh. strain 89 is such that it catalyses preferen-
tial hydroxylationat C6 instead of catalysing oxidative
defluorination at C2. The activity of the catechol 1,2-
dioxygenase, catalysing the subsequent reaction step,
appears to be an important rate limiting factor in the
biodegradation of the fluorophenols by Rh. opacus
1G and Rh. strain 89, possibly due to the sensitivity
Next, a proton coupled 19F NMR spectrum of the
metabolite with its 19F NMR resonance at -189.9 ppm
was measured to further elucidate its possible nature.
1
The H coupled 19F NMR pattern of the resonance
showed one JFH coupling of 50.4 Hz. This value
can be compared to coupling constants reported in
4
the literature, where JFH values, for coupling over
one double and three single bonds, vary between 1–
2
8 Hz and JFH values are between 45 and 65 Hz
biod533.tex; 13/01/1999; 14:06; p.10

485
of the enzyme to suicide inhibition by fluorocate-
chol metabolites. This impairment of ring cleavage
by the presence of halogen substituents ortho with
respect to the hydroxyl moieties of a ring cleavage
metabolite is similar to the inhibition of protocate-
chuate dioxygenase by ortho-halogenated protocate-
chuates (Walsh and Ballou 1983) and of catechol 2,3-
dioxygenase by ortho-halogenated catechols (Bartels
et al. 1984). The phenomenon was reported to origi-
nate from formation of an abortive complex between
the ortho-halogenated catechol and the dioxygenase
(Walsh et al. 1983). Rh. corallinus 135 appeared
very efficient in fluorophenol biodegradation, being
able to catalyse oxidative defluorination, and con-
taining a catechol 1,2-dioxygenase with an unusual
broad substrate specificity. Especially in incubations
of Rh. erythropolis 1CP, accumulation of fluorinated
metabolites different from the fluorophenols, fluoro-
catechols or fluoromuconates, was observed. From
additional experiments with purified chloromuconate
cycloisomerase from Rh. erythropolis 1CP, evidence
was provided suggesting that one of these metabolites
is the keto form of 5-fluoromaleylacetate (Figure 6).
The experimental data presented for the conversion
of the 2,3-difluoromuconate by chloromuconate cy-
cloisomerase (Figure 5) are in line with literature
data on the conversion of 2-chloro-, 3-chloro- and 3-
fluoromuconate by chloromuconate cycloisomerases
(Harper & Blakley 1971; Schmidt and Knackmuss
1980; Schreiber et al. 1980; Schlömann et al. 1990b;
Solyanikova et al. 1995; Vollmer et al. 1994; Vollmer
and Schlömann 1995; Prucha et al. 1996; Blasco et
al. 1995; Kaschabek & Reineke 1995). The data pre-
sented here provide a first indication on the nature
of the unidentified accumulated metabolites, and are
included to illustrate the use of 19F NMR in stud-
ies on biodegradation. However, the exact nature of
all possible fluoromuconolactone, fluorodienelactone
and fluoromaleylacetate metabolites remains to be
elucidated.
idative decarboxylation step, are similar to the ones
reported here for fluorocatechols, i.e. ring cleavage by
an intradiol dioxygenase followed by defluorination of
the fluoromuconates thus formed in subsequent reac-
tion steps (Goldman et al. 1967; Harper & Blakley
1971; Clarke et al. 1975; Schreiber et al. 1980; Ivoilov
et al. 1987; Engesser et al. 1990; Schlömann et al.
1990a).
Altogether, the present paper illustrates the pos-
sibilities and advantages of the use of 19F NMR
in studies on the biodegradation of fluorinated com-
pounds. Clearly, the main advantages, as compared
to other techniques, such as for example HPLC or
GC-MS, are that the 19F NMR data can be eas-
ily quantified, that no disturbing background signals
are present, that no metabolite extraction or separa-
tion procedures are required and that the amount of
fluoride anion formation, i.e. the extent of dehalogena-
tion, can be quantified. The method is restricted to
compounds containing fluorine substituents. However,
with the expected future increase in the number of
fluorinated chemicals in both industry and agriculture
(Schreiber et al. 1980; Oltmanns et al. 1989; Engesser
et al. 1990), the 19F NMR technique may develop
as an important tool in this field of environmental
microbiology.
Acknowledgements
This work was supported by EC grant ERB IC15-
CT96-0103, by the Wageningen NMR Center, EU
contract CHGE-CT94-0061 and, in part, by the
Russian State Research Programm Novel Methods in
Bioengineering (Environmental Biotechnology).
References
Banks RE & Tatlow JC (1994) Synthesis of organofluorine com-
pounds. In: Banks RE, Smart B & Tatlow JC (Eds) Organoflu-
orine Chemistry: Principles and Commercial Applications (pp
25–57). Plenum Press, New York
Banks RE (1995) Fluorine in Agriculture. Conference Papers, The
University of Manchester, Institute of Science and Technology,
Fluorine Technology Lmt, UK
Banks RE & Lowe KC (1994) Fluorine in Medicine in the 21st Cen-
tury. Conference Papers, The University of Manchester, Institute
of Science and Technology, Fluorine Technology Lmt, UK
Bartels I, Knackmuss HJ & Reineke W (1984) Suicide inactivation
of catechol 2,3-dioxygenase from Pseudomonas putida mt-2 by
3-halocatechols. Appl. Environ. Microbiol. 47, 500–505
Blasco R, Wittich R-M, Mallavarapu M, Timmis KN & Pieper
DH (1995) From xenobiotic to antibiotic, formation of pro-
Previous studies have reported on the biodegra-
dation of fluorobenzoic acids and fluorocatechols by
many different microorganisms (Goldman et al. 1967;
Clarke et al. 1975; Schreiber et al. 1980; Ivoilov
et al. 1987; Engesser et al. 1990; Schlömann et al.
1990a; 1990b), but in only a limited number of stud-
ies, 19F NMR was applied to identify the metabolites
and/or quantify the metabolite profiles (Van Berkel et
al. 1994; Boersma et al. 1998). The metabolic profiles
reported for the degradation of mono-fluorocatechols,
formed from monofluorobenzoates in an initial ox-
biod533.tex; 13/01/1999; 14:06; p.11

486
toanemonin from 4-chlorocatechol by enzymes of the 3-
oxoadipate pathway. J. Biol. Chem. 270: 29229–29235
Rietjens IMCM, Cnubben NHP, de Jager PA, Boersma MG &
Vervoort J (1993) Applications of NMR in biotransformation
studies. In: Weitzner MI (Ed) Developments and ethical consid-
erations in toxicology (pp 94–109)
Boersma MG, Dinarieva TY, Middelhoven WJ, van Berkel WJH,
19
Doran J, Vervoort J & Rietjens IMCM (1998) F nuclear mag-
netic resonance as a tool to investigate the microbial degradation
of fluorophenols to fluorocatechols and fluoromuconates. Appl.
Environ. Microbiol. 64: 1256–1263
Schlömann M, Schmidt E & Knackmuss H-J (1990a) Different
types of dienelactone hydrolase in 4-fluorobenzoate utilizing
bacteria. J. Bacteriol. 172: 5112–5118
Clarke KF, Gallely AG, Livingstone A & Fewson CA (1975)
Metabolism of monofluorobenzoates by Acinetobacter cal-
coaceticus N.C.I.B. 8250. Biochim. Biophys Acta 404: 169–179
Edwards PN (1994) Uses of fluorine in chemotherapy. In: Banks RE,
Smart B & Tatlow JC (Eds) Organofluorine Chemistry: Princi-
ples and Commercial Applications (pp 501–542). Plenum Press,
New York
Engesser KH, Auling G, Busse J & Knackmuss HJ (1990) 3-
Fluorobenzoate enriched bacterial strain FLB 300 degrades
benzoate and all three isomeric monofluorobenzoates. Arch.
Microbiol. 153: 193–199
Schlömann M, Fischer P, Schmidt E & Knackmuss H-J (1990b)
Enzymatic formation, stability, and spontaneous reactions of 4-
fluoromuconolactone, a metabolite of the bacterial degradation
of 4-fluorobenzoate. J. Bacteriol. 172: 5119–5129
Schmidt E & Knackmuss H-J (1980) Chemical structure and
biodegradability of halogenated aromatic compounds. Conver-
sion of chlorinated muconic acids into maleoylacetic acid.
Biochem. J. 192: 339–347
Schreiber A, Dorn MHE, Reineke W & Knackmuss H-J (1980) Crit-
ical reactions in fluorobenzoic acid degradation by Pseudomonas
sp B13. Appl. Environ. Microbiol. 39: 58–67
Goldman P, Milne GWA
&
Pignataro MT (1967) Fluorine
Sejlitz T & Neujahr HY (1987) Phenol hydroxylase from yeast. A
model for phenol binding and an improved purification proce-
dure. Eur. J. Biochem. 170: 343–349
Solyanikova IP, Maltseva OV, Vollmer MD, Golovleva LA & Schlö-
mann M (1995) Characterization of muconate and chloromu-
conate cycloisomerase from Rhodococcus erythropolis 1CP: in-
dications for functionally convergent evolution among bacterial
cycloisomerases. J. Bacteriol. 177: 2821–2826
containing metabolites formed from 2-fluorobenzoic acid by
Pseudomonas species. Arch. Biochem. Biophys. 118: 178–184
Gorlatov S, Maltseva O, Shevchenko V & Golovleva L (1989)
Degradation of chlorophenols by a culture of Rhodococcus
erythropolis. Microbiology (Engl. Transl. Mikrobiologiya) 58:
647–651
Harper DB
& Blakley ER (1971) The metabolism of p-
fluorobenzoic acid by Pseudomonas sp. Can. J. Microbiol. 17:
1015–1023
Straube G (1987) Phenol hydroxylases from Rhodococcus sp. P1. J.
Basic Microbiol. 27: 229–232
Ivoilov VS, Karasevich YN & Surovtseva EG (1987) Conversion
and utilization of 4-fluorobenzoic acid by Corynebacterium fas-
cians. Microbiology (Engl. Transl. Mikrobiologiya) 56: 199–204
Janke D., Al-Mofarji T, Straube G, Schuman P & Prauser H (1988)
Critical steps in degradation of chloroaromatics by rhodococci. I.
Initial enzyme reactions involved in catabolism of aniline, phenol
and benzoate by Rhodococcus sp. An 117 and An 213. J. Basic
Microbiol. 28: 509–518
Kaschabek SR & Reineke W (1995) Maleylacetate reductase of
Pseudomonas sp. strain B13: specificity of substrate conversion
and halide elimination. J. Bacteriol. 177: 320–325
Key BD, Howell RD & Criddle CS (1997) Fluorinated organics in
the biosphere. Environ. Sci. Techn. 31: 2445–2454
Van Berkel WJH, Eppink MHM, Middelhoven WJ, Vervoort J & Ri-
etjens IMCM (1994) Catabolism of 4-hydroxybenzoates in Can-
dida parapsilosis proceeds through initial oxidative decarboxy-
lation by a FAD-dependent 4-hydroxybenzoate 1-hydroxylase.
FEMS Microb Lett 121: 207–215
Vervoort J, de Jager PA, Steenbergen J & Rietjens IMCM (1990)
19
Development of a F-NMR method for studies on the in vivo
and in vitro metabolism of 2-fluoroaniline. Xenobiotica 20: 657–
670
Vollmer MD, Fischer P, Knackmuss H-J & Schlömann M (1994)
Inability of muconate cycloisomerases to cause dehalogena-
tion during conversion of 2-chloro-cis,cis-muconate. J. Bacteriol.
176: 4366–4375
Malet-Martino MC & Martino R (1989) The application of nuclear
magnetic resonance spectroscopy to drug metabolism studies.
Xenobiotica 19: 583–607
Nakai C, Nakazawa T & Nozaki M (1988) Purification and
properties of catechol 1,2-dioxygenase (pyrocatechase) from
Pseudomonas putida mt-2 in comparison with that from
Pseudomonas arvilla C-1. Arch. Biochem. Biophys. 267: 701–
713
Vollmer MD & Schlömann M (1995) Conversion of 2-
chloro-cis,cis-muconate and its metabolites 2-chloro- and 5-
chloromuconolactone by chloromuconate cycloisomerases of
pJP4 and pAC27. J. Bacteriol. 177: 2938–2941
Walker SB (Ed) (1990) Fluorine Compounds as Agrochemicals.
Fluorochem Lmtd. Derbyshire.
Walsh TA & Ballou DP (1983) Halogenated protocatechuates as
substrates for protocatechuate dioxygenase from Pseudomonas
cepacia. J. Biol. Chem. 258: 14413–14421
Walsh TA, Ballou DP, Mayer R & Que L Jr (1983) Rapid reaction
studies on the oxygenation reactions of catechol dioxygenase. J.
Biol. Chem. 258: 14422–14427
Oltmanns RH, Müller R, Otto MK & Lingens F. (1989) Evidence for
a new pathway in the bacterial degradation of 4-fluorobenzoate.
Appl. Environ. Microbiol. 55: 2499–2504
Peelen S, Rietjens IMCM, Boersma MG & Vervoort J (1995) Con-
version of phenol derivatives to hydroxylated products by phenol
hydroxylase from Trichosporon cutaneum. A comparison of re-
gioselectivity and rate of conversion with calculated molecular
orbital substrate characteristics. Eur. J. Biochem. 227: 284–291
Prucha M, Peterseim A, Timmis KN & Pieper DH (1996) Mu-
conolactone isomerase of the 3-oxoadipate pathway catalyzes
dechlorination of 5-chlorosubstituted muconolactones. Eur. J.
Biochem. 237: 350–356
Wray V (1983) Fluorine-19 nuclear magnetic resonance spec-
troscopy. In: Webb GA (Ed) Annual Reports on NMR Spec-
troscopy, Vol 14. (pp 149–191). Academic Press Inc., London
biod533.tex; 13/01/1999; 14:06; p.12