Purification and Characterization of Phthalate

0 downloads 0 Views 2MB Size Report
Aug 28, 1986 - An enzymatic system has been isolated that catalyzes dihydroxylation of phthalate to form 1,2-dihydroxy-. 4,5-dicarboxy-3,5-cyclohexadiene ...
THEJOURNALOF BIOLOGICAL CHEMISTRY 0 1987 by The American Society of Biological Chemists, Inc

Vol. 262, No. 4, Issue of February 5,pp. 1510-1518, 1987 Printed in U.S.A.

Purification and Characterizationof Phthalate Oxygenase and Phthalate Oxygenase Reductase fromPseudomonas cepacia” (Received for publication,August 28, 1986)

Christopher J. Batie, Edward LaHaie, and David P. BallouS From the Departmentof Biological Chemistry, Universityof Michigan, Ann Arbor, Michigan48109

An enzymatic system has been isolatedthat catalyzes cis-dihydrodiols. Next, NAD-coupled dehydrogenations yield dihydroxylation of phthalate to form 1,2-dihydroxy- catechols (3, 7). Thus, unactivated aromatic compounds are 4,5-dicarboxy-3,5-cyclohexadienewith consumption converted into much more reactive catechols with only a net ofNADH and 0,. This system is comprised of two consumption of oxygen. proteins: a flavo-iron-sulfur protein with NADH-dePhthlalate’isanunactivatedaromatic chemical that is pendent oxidoreductase activity and a nonheme iron released into the environment by industrial as well as by protein with oxygenase activity. Phthalate oxygenase natural processes ( 5 ) . Ribbons and his associates (and others) is a large (approximately 217 kDa) protein composed (5, 15-17) have isolated bacteria thatwill utilize phthalate as of apparently identical 48-kDamonomers. The active sole carbon source and have resolved the metabolic pathways enzyme has one Rieske-type [2Fe-2S] center and one mononuclear iron/monomer. Removal of the mononu- of phthalate degradation. In Pseudomonas cepacia (Scheme give phthalate 4,5clear iron by incubation with EDTA or with o-phe- I)phthalateisfirstdihydroxylatedto (1,2-dihydroxy-4,5-dicarboxy-3,5-cyclohexadinanthrolineinhibitsoxygenation;ferrous ion com- dihydrodiol pletely restores activity. No other metalsare effective. ene)(step 1); thisproduct is dehydrogenatedto give 4,52 ) . Decarboxylation Phthalate oxygenase is specific for phthalate or other dihydroxyphthalateandNADH(step closely related compounds. However, only phthalateis (step 3) then yields protocatechuate, a central metabolite in tightly coupled toNADH oxidation and 0,consumption the metabolism of aromatic compounds (51);protocatechuate with a stoichiometry of 1:l:l.Phthalate oxygenase is dioxygenase catalyzes the opening of the benzene ring (step chemically competentto oxygenate phthalate when ar- 4)to produce P-carboxy-cis-cis-muconate (15,18). tificially supplied with reducing equivalents and 0,. Keyser and Ribbons (5,20) resolved the enzyme system Phthalate oxygenase reductase is required, however, that catalyzes dihydroxylation of phthalate (Scheme I, step is a mon- 1)into two protein components (FactorsA and C). In contrast for efficient catalytic activity. The reductase omeric34-kDaflavo-iron-sulfurproteincontaining tothe ring-fissiondioxygenases (suchasprotocatechuate a dioxygenase (step 4)), phthalate oxygenase required NADH FMN and a plant-ferredoxin-type [2Fe-2S] center in 1:l ratio. Phthalate oxygenase reductase is specific for as a cofactor. Supplementation of the enzyme with ferrous NADH but canpass electrons toa variety of acceptors, including: phthalate oxygenase, cytochrome e, ferri- ion was required for maximal activity, indicating iron as a cyanide, anddichlorophenolindophenol. This systemis cofactor (5).The UV-visible absorption spectrum of Factor A No similar to other bacterial oxygenase systems involved suggested that it contained an iron-sulfur center. evidence in aromatic degradation including: benzoate dioxygen- for participation of a cytochrome P-450 was found (20). We report here the purification and characterization of this ase, toluene dioxygenase, benzene dioxygenase, and 4methoxybenzoate demethoxylase. However, phthalate phthalate oxygenase and of itselectrontransferprotein, oxygenase can be isolated in large quantities and is phthalate oxygenase reductase, from P. cepacia.* The active oxygenase has one Rieske-type [ZFe-ZS] center and one admore stable thanmost other suchsystems. ditional Fez+ (mononuclear iron(30)) per monomer; it corresponds to FactorA (referred toabove). The electron transfer protein has both FMN anda plant-ferredoxin-type [2Fe-2S] Soil bacteria play an importantrole in ecology by degrading center on a single polypeptide. It is apparently the same as a wide variety of aromatic compounds. Such compounds are Factor C described by Keyser and Ribbons, although both the released into the environmentlargely by plants as secondary apparent molecular weights and the absorbance spectra differ metabolites and as lignin breakdownproducts (1, 2 ) . The substantially. metabolism of these compounds constitutesa significant porPhthalate oxygenase is oneexample of a class of oxygenases tion of the carbon flux in nature (3, 4). that catalyzes formation of cis-dihydrodiols. All enzymes of Bacterial degradationof aromatic compoundsgenerally prothis class that have beenisolatedhave spectral properties ceeds by a series of oxygenations. Firstthecompoundis convertedto a catechol. Secondary dioxygenationscleave The abbreviationsused are: dihydrodiol orphthalate 4,5-dihydrocarbon-carbon bonds, thus opening the ring and destroying diol, 1,2-dihydroxy-4,5-dicarboxy-3,5-cyclohexadiene; HEPES, 4-(2aromaticity (4-6). Gibson and co-workers (7-14)have shown hydroxyethy1)-I-piperazineethanesulfonicacid; Fd, ferredoxin; SDS, that soil bacteria convert unactivated aromatic compounds tosodium dodecyl sulfate; [2Fe-2S], 2 iron, 2 sulfur nonheme iron center; catechols via a two-step process. First, dihydroxylationsyield DTNB, 5,5’-dithiobis-(2-nitrobenzoicacid); HPLC, high performance liquid chromatography. This strain was originally classified as Pseudomonas fluorescem * This work was supported by National Institutes of Health Grant GM20877. The costs of publication of this article were defrayed in PHK (20). It has also been describedas Pseudomonas putida (18).It has since been reclassified as P. cepacia on the basis of detailed part by the payment of page charges. This article must therefore be hereby marked“aduertisement” in accordance with 18 U.S.C. Section nutritional studies (Dr. R. Olsen, personal communication). This is the same organism used in isolation of protocatechuate 3,4-dioxygen1734 solely to indicate this fact. ase by our laboratory (18). 2 To whom correspondence should be addressed.

1510

1511

Purification of a PhthalateOxygenase System NADH

WAD

0-

WAD

1

NADH

2

4

3 SCHEME I

very similar to the Rieske-type [2Fe-2S] centers which participate in the electron transfer chainsof mitochondria, chloroplasts, and bacteria (21-26). A Rieske protein from Thermus thermophilushasrecentlybeen shown t o have a[2Fe-2S] 6.6 center substantially different from that of other [2Fe-2S]containingproteins,suchasspinach ferredoxin (19). The latter have four cysteines symmetrically arranged around the [2Fe-2S] center. In contrast, theRieske protein has only two cysteines/[2Fe-2S] center, ruling out this symmetricalgeom- 0 6.0 etry. Thus, theRieske center constitutes a new type of [2Fe0 2S] center. Phthalate oxygenase seems to bemore stable than a 0 most other systems in this class and is available in abundant -1 quantity from phthalate-grown P. cepacia. These character4.6 istics make the systemwell suited for mechanistic and structural study by a variety of methods including spectroscopic and rapid reaction techniques. EXPERIMENTALPROCEDURESANDRESULTS3

4.0

4

I

I

I

I

0.1

0.2

0.3

0.4

Physical and Chemical Characterization-Purified phthalK8V ate oxygenase and phthalateoxygenase reductase eachshowed only a single band on sodiumdodecyl sulfate-polyacrylamide gel electrophoresis, indicating that each of the nativeenzymes contains only one type of polypeptide. The estimated molecular weights of the denatured proteins were 48,000 and 34,400, respectively (Fig. 1).Gel filtration was then used to estimate 4.8 the native molecular weights of the two enzymes. Phthalate oxygenase reductase was foundt o be a monomer: M, = 34,000 on Sephadex G-100 and 34,300 on Sephacryl S-300 (Fig. 2, A and B). The oxygenase was excluded from Sephadex G-100, but gel filtration on Sephacryl S-300 indicated M , = 217,000 (Fig. W). Hence phthalate oxygenase is most probably a n a4 tetramerinthenativestate.The differencebetween the calculated tetramersize (192,000) and thatobserved (217,000) is within the uncertaintyof these methods. Flavin and[2-Fe-2Sl Content-Phthalate oxygenase reduc4.2 4 \ tase, which is orange, has an absorbance spectrum consistent with the presenceof both a flavin and a plant-type ferredoxin (compare Fig. 3 with Fig. 4 (46)). The spectrum has peaks at I 330 and 462 nm.Amongseveral preparations of purified 0.2 0.3 0.4 0.6 0.8 0.1 ( 3 reductase the ratioof AZm:A462varied from 3.1 to 3.4. K.” Upon boiling, the phthalate oxygenase reductase solution FIG. 2. Determination of native molecular weights by gel turned yellow and the protein precipitated; loss the in absorb- filtration. A , Sepbacryl S-300 elution. Standards were myoglobin, ance is most probably due to destruction of a [2Fe-2S] center bovine serum albumin, bovine immunoglobulin G, bovine catalase, (see below). The spectrum of the chromophore remaining in glutamate dehydrogenase, horse ferritin, bovine thyroglobulin. Samsolution resembled closely that of FMN (Fig. 3). HPLC and ples and standards were chromatographed separately on a 1.5 X 90fluorimetric analyses,as described under “Experimental Pro- cm Sephacryl S-300 column equilibrated with 50 mM KPi, pH 6.8. cedures,” also indicated that theyellow color of the superna- Ferricyanide and blue dextran 2000 were used to determine Viand tant solution was due to FMN. Neither FAD nor riboflavin VO.B , Sephadex G-100 elution. Standards were horse cytochrome c,

-I

~

1

Portions of this paper (including “Experimental Procedures,” part of “Results,” Tables I and 11, Fig. 1, Scheme 11, and Footnotes 4 and 5) are presented in miniprint at the end of this paper. Miniprint is easily read with the aid of a standard magnifying glass. Full size photocopies are available from the Journal of Biological Chemistry, 9650 Rockville Pike, Bethesda, MD 20814. Request Document No. 86M-2976, cite the authors, and include a check or money order for $6.00 per set of photocopies. Full size photocopies are also included in the microfilm edition of the Journal that is available from Waverly Press.

,

I

1

myoglobin, a-chymotrypsinogen, ovalbumin, and bovine serum albumin. All samples were eluted as in A on a 2.5 X 89-cm Sephadex G-100 (medium) column. PO, phthalate oxygenase; POR, phthalate oxygenase reductase.

could be detected. The protein precipitatewas white, indicating that nosignificant amount of chromophore was bound. The molar absorbance coefficient of phthalate oxygenase of the amountof flavin reductase was established on the basis released from denatured protein. FMN was measured by all

1512

Purification of a Phthalate Oxygenase System A

1.6

-

c

-hE,

-E

1

20

Gz

A

a

m LT 0 v) m

Oxidized PO

0.6

10

a

300 600

eo0

400

I

I

I

I

400

600

600

700

E

Wavelength (nrn)

700

WAVELENGTH (nm) FIG. 3. Absorption spectra of phthalate oxygenase reduc-, purified reductase in 50 mM KPi, pH 6.8; - - -, tase (POR). same sample after 3 min a t 90 “C followed by centrifugation. Assuming that thevisible absorbance of the heat-treated enzyme is entirely due to FMN, molar absorbance coefficients were assigned to native reductase. TABLEI11 Iron, acid-labile sulfide, and flavin content Phthalate oxygenase and phthalate oxygenase reductase concentrations were determined from their visible absorbance spectra as described under “ExDerimental Procedures.” Enzvme

Iron”

S2-b

o 4 0.7

0.6

1

FMN‘

mollmol enzyme monomer

Phthalate oxygenase reductase Phthalate oxygenase as isolated EDTA dialyzedd

1.91 2.07 2.70 2.00

2.27 2.12

1.0 0.0 0.0

Iron was determined by the ferrozine method described under “Experimental Procedures.” Acid-labile sulfide was determined as by Rabinowitz (39). Flavin content was determined spectrophotometrically after heating the enzymes to 90 “C and centrifuging to remove precipitated protein. Phthalate oxygenase and phthalate oxygenase reductase were dialyzed versus 0.1 M HEPES, pH 8.0, 10 mM EDTA.

0.4

I /

0.3 0

1

I

I

,

I

I

60

100

160

200

260

300

:

0

Ferricyanide (nrnol)

0.4

three analytical procedures described under “Experimental Procedures”; enzyme was denatured either by acid precipitation with trichloroacetic acid or by thermal denaturation. All * Q procedures indicated that per unit of absorbance at 462 nm, 0.3 a 42 PM flavin was released. If phthalate oxygenase reductase has one FMN/monomer of active enzyme (an assumption generally supported by the ratio of FMN to protein6), then 0.2 e462 = 24 x IO3M” cm”. Phthalate oxygenase reductase also has one [2Fe-2S] center/enzyme. Iron and acid-labile sulfide were found in a ratio of 2 each/flavin (Table 111).E P R spectra of phthalate oxygen0.1 I I asereductase previouslyreduced by 1 NADH/flavinhad 0 60 80 20 40 features with g values of 2.041, 2.008, 1.949, and 1.900 (54). NADH (nmol) The featureat g = 2.008 indicated the presenceof the neutral FIG.4.. Absorption spectra of phthalate oxygenase (PO). A: (blue) flavin semiquinone. The other signals are similar to -, UV-visible spectra of oxidized phthalate oxygenase (2.2 ml, 4.1 those of spinach ferredoxin (60, 61). Hence, phthalate oxyI

6 T h e protein content of phthalate oxygenase reductase, when assayed by the Lowry procedure (43), indicated an FMN:polypeptide ratio of 0.75, a number which rules out the possibility that the flavoprotein has more than 1 FMN/molecule. The deviance from a ratio of 1.0 is most likely due to thevariability in the response of the Lowry method to different proteins (43). It may, however, indicate the presence of some apoprotein.

I

mg/ml) a t ambient temperature in 100 mM HEPES, pH 8.0, 20 mM EDTA, and 260 nM FMN; - - -, reduced enzyme, the same as above but after 20 min of illumination with a 2000-watt tungsten-halogen lamp at a distance of about 12 inches. Prior to the experiment the enzyme was incubated with 16 mM potassium ferricyanide. After 5 min, ferricyanide was separated from the enzyme bygel filtration through Sephadex G-25 in 100 mM HEPES, pH8.0. EDTA and FMN were added as indicated, and the sample wasmade anaerobic as

1513

Purification of a PhthalateOxygenase System TABLEIV Dihydrodiol formation by phthalate oxygennse without reductase Reactions were carried out in a volume of 1.1 ml bufferedwith 0.1 M HEPES, pH 8.0, at 25 "C. Where indicated the reaction mixture contained phthalate oxygenase Type I (109 nmol), 5 mM phthalate, Fez+ (110 nmol), methyl conditions with viologen (200 nmol), spinach ferredoxin (110 nmol). Samples werereducedunderanaerobic dithionite as described in Ref. 55; 1 eq of dithionite was taken as that required to reduce phthalate oxygenase with only minimal spectral contributions from reducedmethyl viologen. Oxidation states of the reactants were monitored spectrophotometrically. After reduction the reactants were aerated. Upon the return of the spectrum of oxidized phthalate oxygenase, the sample was filtered through a Centricon-lG filter. The filtrate was assayed for dihydrodiol by adding NAD (0.2 or 1 mM final) and phthalate 4,5-dihydrodiol dehydrogenase (1.5 units) and then monitoring the increase in A3a. We assumed one NADH produced per product molecule. Reaction components Experiment

1 2 3 4 5

Phthalate oxygenase

+ + + +-

Fe2'

-

+ -

+

Reducing

Methyl viologen

+ + + + +

Spinach ferredoxin "

-

-+

equivalents of dithionite

1 0.221 2 3 1

Product yield

nmol

5.8 24.0 19.2 24.2 0.0

molfmol of oxygenase ___

0.05 0.18 0.23 0.00

of genase reductase has two redox centers: FMN and an ferre- facilitates oxygenation. In all experiments the spectrum oxidized enzyme returned very slowly after aeration of the doxin-like [2Fe-2S] center. T h e absorbance spectra of oxidized and reduced phthalate reduced oxygenase (the reaction rate was 0.01 s" a t 25 "C). oxygenase (Fig. 4) were very similar to those of the Rieske- In contrast, at 25 "C, V,,, of phthalate oxygenase is approxitype [2Fe-ZS] protein from T. thermophilus (see Fig. 4 (19)). mately 17 s-'. Under standard assay conditions of 0.18 VM E P R spectra of reduced phthalate oxygenase (52) were also phthalate oxygenasereductase, which is not sufficient for very similar to thoseof the Rieske proteinof T. thermophilus optimal rates, phthalateoxygenase has a turnover number of (19). Thus, phthalateoxygenase contains a Rieske-type [2Fe- about 0.8 s-'. 2S] center. Other iron-sulfur-containing oxygenases isolated In catalysis, phthalate oxygenase is specific for phthalate from pseudomonads have similar visible and EPR spectra oxygenase reductase as its reductant. When phthalate oxy(21-29); thus, this center iswidely distributed in nature. genasewas incubatedwithphthalate, 02,andNADH (or In the purified phthalate oxygenase (with phthalate reNADPH) and otheroxidoreductases including benzoate oxymoved by gel filtration) the ratio of A2M):A462 was 14. Upon genase reductase from Pseudomonas aruilla, cytochrome Pboiling, the enzyme lost all visible color; neither flavin nor 450 reductase from rabbit liver microsomes, and the spinach heme couldbe detected in the supernatant solution. The molar ferredoxin-ferredoxin-NADPreductasesystemnoproduct absorbance of phthalate oxygenase was determined by titra- could be detectedby HPLC analysis (Method11). In addition, tions of reduced enzyme with ferricyanide (Fig. 4B) and by in single-turnover experiments inclusion of reduced spinach reductive titrations with NADH (withcatalyticreductase, ferredoxin increased neither the rate of oxidation of phthalate Fig. 4C) and with dithionite.If we assume that each [2Fe-2S] oxygenase [2Fe-2S] center nor the yield of product (Table center will take up one electron (60, 61), then the indicated IV). These data indicate that phthalateoxygenase reductase absorption coefficients (per molar [2Fe-2S]) of oxidized not only donates electrons to the oxygenase but that it also phthalate oxygenase a t 464 nm were 7800,7700, and 8000 M" interacts with the oxygenase in some manner toincrease both cm", respectively. A value of 7800 M" cm" was taken as the the rate andefficiency of oxygenation. concensus value and used in subsequent determinations of Phthalate oxygenase requires an additional iron atom, rethe concentrationof [2Fe-2S] center. ferred to as mononuclear iron (30), in addition to the 12FeIron and sulfide analyses of phthalate oxygenase that had 2S] center, for activity. When phthalate oxygenase was isobeendialyzed againstEDTAintheabsence of phthalate lated in the absence of phthalate, its activity when assayed indicated 2 Fe and 2 S2-/[2Fe-2S] (Table 111). The protein without supplemental Fez+ was only 30% of that observed concentration of phthalate oxygenase solutions of known when Fe2+was added to the assay. However, phthalate oxyabsorbanceindicatedthepresence of 0.9 Rieske center/ genase isolated in the presence of phthalate and dithiothreitol 48,000-dalton phthalate oxygenase monomer. These data inhad 70-80% of full activity when assayed in the absence of dicate that phthalate oxygenase has one Rieske-type [2FesupplementalFez+.EDTAando-phenanthrolineinhibit 2S] center/monomer. phthalate oxygenase. Of several metal ions tested, only Fez+ Catalytic Properties-The Rieske-type iron-sulfur protein and Fe3+ were effective instimulatingactivityin EDTAwas identified asthe oxygenase by showing that,inthe dialyzed phthalate oxygenase (Table V), ferrous ion being absence of phthalate oxygenase reductase,the oxygenase most effective. Cu2+ and Zn2+ inhibit phthalate oxygenation. would oxygenate phthalatewhensupplied with reducing Phthalate oxygenase requires one Fe2+/[2Fe-2S]for maxiequivalents (Table IV). Either addition of excess dithionite of mal activity; this conclusion is based on two types of experior excess Fez+ (TypeI oxygenase was used in which ~ 7 0 % the enzyme already has mononuclear iron bound,see below) ments. In Experiment 1, addition of o-phenanthroline to the assay mixture in a concentration 2.5-fold that of phthalate described inRef. 55. B, titration of photoreducedphthalate oxygenase oxygenase (Type 11) caused 85-95% inhibition of oxygenase with ferricyanide.Enzyme in A was titrated with 4.46 mM ferricyanide activity (Fig. 5). About 1 mol of Fe2+/mo1 of [2Fe-2S] was after photoreduction. Spectra were recorded after each addition. C, required to saturate the o-phenanthroline; an additional molar titration of phthalate oxygenase with NADH. Phthalate oxygenase (3.31 mg) and phthalate oxygenase reductase (0.25 nmol) were equil- equivalent of Fez+causednearly maximal reactivation of ibrated with 0.1 M HEPES, pH 8.0, and diluted to a volume of 2 ml. phthalate oxygenase. These data indicate that the mononuclear iron is tightlybound, KD < 1 PM, and required for The enzymes were titrated with 1.33 mM NADH underanaerobic conditions as described in Ref. 55. activity. In Experiment 2 phthalate oxygenase samples with

Phthalate a Purification of

1514

TABLEV Effect of metal ions on activity of EDTA-dialyzed phthalate oxygenase Phthalate oxygenase was dialyzed overnightversus 0.1 M HEPES, pH 8.0, 0.2 mM EDTA. Assays were conducted at 25 “C with 0.244 p~ phthalate oxygenase reductase and0.341 p~ phthalate oxygenase. Added metals were at 100 p~ (stock solutions were 10 mM in 0.1 M HCL). Salts used were Fe(NH4)z(S04)z,NiCI,, NaZMo04,ZnClz, CuSO,, MnC12, MgClZ,CaCI2, CoSO,, FeCb. Metal ion added to assav

Activity NADHjminjoxygenase

None Fez+ Fe3+ Ni2+ Co2+

MnZ+ MOO:cu2+

Znz+

12.2 56.2 18.3 8.3 8.7 8.6 5.1 1.6

Oxygenase System TABLEVI Correlation of iron content and activity Assays were conducted as described under “Experimental Procedures,” either with or without the supplemental Fez+.All oxygenase samples were equilibrated with bufferby passage over a Sephadex G25 column just prior to the assays. The gel was equilibrated with 0.1 M HEPES, pH 8.0, and 5 mM phthalate for samples 1 and 3; 0.1 M HEPES, pH 8.0, was used in preparing sample 2. Iron assays were performed by the ferrozine method described under “Experimental Procedures.” Sample 1, oxygenase (1 mM) was thawed just prior to the experiment. Sample 2, oxygenase (0.7 mM) was incubated overnight at room temperature with0.1 M HEPES, pH 8.0,lO mM EDTA. Sample 3, oxygenase (0.8 mM) was incubated with 2 mM Fe2+,5 mM phthalate, 10 mM EDTA, and 0.1 M HEPES, pH 8.0, at 4 ”C overnight. Sample

Description

0.58 1 2.80As isolated 2 EDTA treated 3 Iron treated

Iron/oxygenase

(-Fe,+ Activity Fe)

2.00

0.10 0.85

2.96

4.4

12.2 11.9

from 2 to 3 total iron atoms/oxygenase monomer. Additional iron (more than3 total) would bind to phthalateoxygenase if EDTA were omitted from the incubation mixture, but the 260 specific activity was not altered. Together these data indicate 0 that the active form of phthalate oxygenase has one mononuclear iron and one[ 2Fe-2SI center/monomer. Substrate Stoichiometry-The stoichiometry of NADH: 02:phthalate consumed during catalysis by phthalate oxygenase was determined by following NADH oxidation (spectrophotometrically) or O2 consumption (by oxygen electrode) when limiting phthalate was added to an otherwise complete assaymixture. A 1:1:1 stoichiometry of NADH: 02:phthalate was indicated. In an analogous experiment, when NADH was the limiting reagent and the disappearance of phthalate was monitored by HPLC (Method I) a 1:l ratio of NADH:phthalate was also observed. These observations confirm those reported by Keyser et al. (5,20) and indicate that oxygenation of phthalate by the phthalate oxygenase-phthalate oxygenase reductase system is tightly coupled to NADH oxidation andoxygen activation. It is interesting to note that I I , in the absenceof the reductase, theoxygenation of phthalate 0 1 2 3 4 ismuch lesscoupled toreduction of the oxygenase. For Fe / Phthalate Oxygenase (mol/rnol) of Table IV, if the reaction were example, in the experiments tightly coupled we would have expected a t least 0.5 mol of FIG. 5. Iron dependence of phthalate oxygenase activity. ) incubated for 2 min in an Phthalate oxygenase type I1 (4.13 p ~ was product/mol of oxygenase (2 reducing equivalents required assay mixture consisting of 10 p M o-phenanthroline, 100 p~ sodium per reaction). Instead, only 0.2 mol of product was determined. ascorbate, 100 NM NADH, and 100 mM HEPES, pH8.0. Fez+,as Hence,thereductase also promotes a tighter coupling of indicated, and phthalate oxygenase reductase (0.17 p ~ were ) added phthalate oxygenation. and activity assayed as phthalate-dependent NADHoxidation. Substrate Specificity-The substrate specificity of phthalPhthalate concentration was 1 mM. Oxygenasefromtwo different ate oxygenase was tested by looking forstimulation of NADH preparations was used in the experiment ( 0 , O ) . consumption in the presenceof phthalate oxygenase, phthalvarious iron contents were prepared and theirspecific activi- ate oxygenase reductase, andFe2+(Table VII). Only aromatic ties determined in the presence and in the absence of supple- (or heteroaromatic) compounds with vicinal anionic groups mental Fez+ (Table VI). Iron-depleted phthalate oxygenase stimulatedoxidation of NADH. Theanions need not be had only 10% of full activity when assayed in the absence of carboxyl groups since o-sulfobenzoate also gave activity with Fez+. Enzyme with about 3 Fe/monomer (2 for the [2Fe-2S] phthalate oxygenase. Dicarboxypyridines and 4-chlorophthalcenter and1 for the mononuclear site) had nearly full phthal- ate also stimulated NADH oxidation. Oxygenation was tightly coupled to NADH oxidation only ate oxygenase activity when assayed in the absence of Fez+, The mononuclear iron in this sample was retained by the with phthalate as a substrate; 3,4-dicarboxypyridine was unenzyme, even though the enzyme-substrate mixture was in- modified by phthalate oxygenase, eventhough itwas effective cubated overnight with 10 mM EDTA. Thus, in the presence in stimulating NADH oxidation. 2,3-Dicarboxypyridine and of substrate, theFe2+-oxygenase complexmust be very tight. 4-chlorophthalate stimulated oxygen consumption and were Other samples with intermediate iron/oxygenase ratios (be- modified to more polar substances (as determinedby affinity tween 2 and 3) were prepared by incubating EDTA-treated for C-%silica columns, Method 11), but both substrates exoxygenase with various levels of Fez+inthepresence of hibited a stoichiometry of NADH oxidized to substrate oxy1.Tetrachlorophthalate and tetraphthalate, treating with EDTA, and thenremoving unbound genated much greater than stimulated NADH oxidation only very iron and EDTAby gel filtration. Thespecific activity of these bromophthalate samples was proportional to the iron content over the range weakly. M e Caz+

Purification of a Phthalate Oxygenase System

1515

Our observationsregarding the phthalateoxygenase system are ingeneral agreement with the preliminary work of Keyser and Ribbons (5,ZO). We found a two-component system; the Substrateb Activity” with Substrate spectrum of our phthalate oxygenaseagrees well with the uhthalate (yes or no) spectrum of their Factor A. When we analyzed our purified % enzymes by sodium dodecyl sulfate-polyacrylamide gel elec100 Y o-Phthalate trophoresis, however, we observedonlyone band each for 0 m-Phthalate 1 phthalate oxygenase andphthalate oxygenasereductase, p-Phthalate 1 o-Homophthalate whereasthey observedmultiple bands.Proteolysisduring 0 Benzoate purification may explain some of thebands observedby 0 o-Carboxybenzaldehyde Keyser and Ribbons. We have also observed additional low 7 Y o-Sulfobenzoate molecular weight bands in phthalate oxygenase preparations 0 Anthranilate that were isolated without the protease inhibitor phenylmeth69 Y 2,3-Dicarboxypyridine ylsulfonyl fluoride in the homogenization buffer; similar pat64 N 3,4-Dicarboxypyridine 79 Y 4-Chloro-o-Phthalate terns were observed on analysis of partially pure samples 2 Tetrachloro-o-Phthalate stored at -20 “C for several months. 2 Tetrabromo-o-Phthalate Phthalate oxygenase binds Fez+ tightly at the mononuclear ____ Assays were conducted as described under “Experimental Proce- site. This is indicated by the relatively sharp break in the dures,”but with 0.14 p~ reductase. titration curve in Fig. 5, which is consistent with a X, of less Product formation was assayed by HPLC Method11. A compound than 1p ~Phthalate . seems to increase the affinityof oxygenwas considered a substrate if reaction with enzyme resulted in a new ase for Fez+as indicated by the retentionof mononuclear iron more polar component and a decrease in substrate. throughout an overnight incubation with EDTA (Table VI, Experiment 3). We have found that EDTA and o-phenanTABLEVI11 throline readily remove the mononuclear iron when incubated Phthalate oxygenase reductase activities with oxygenase in the absenceof substrate. The otherRieske All assays were monitored spectrophotometrically and conducted [2Fe-2S] center-containing oxygenases also have a requireat 25 “C in aerobically equilibrated 0.1 M HEPES buffer, pH 8.0, containing 100 p~ NADH and reductaseas indicated. Other reagents ment for Fez+ activity (5, 22, 25-28, 30, 33); these enzymes were: Experiment 1, 25 pM phthalate oxygenase, 5 mM phthalate, 50 either are isolated with mononuclear iron bound or require W M Fez+;Experiment 2, no additions; Experiment 3,50 WMcytochrome reconstitution with Fez+ for maximal activity. It seems likely c; Experiment 4, 1 mM potassium ferricyanide; Experiment 5, 45 p M that this site, rather than the [2Fe-2S] center, is locus the of dichlorophenolindophenol. Experiment 1 was monitored at 380 nm oxygen activation. A more complete characterization of the which is isosbestic for the oxidized and reduced forms of phthalate mononuclear iron binding site, especially with regard to the Oxygenase. ExDeriments 2.. 3.. 4, . and 5 were monitored at 340.. 550,. substrate binding site, would greatly aid in describing the 42O;and 600 nm, respectively. mechanism of this enzyme. Experiment Electron acceptor Reductase Turnover Phthalate oxygenase is specificfor aromatic compounds nM NADHfmin similar to phthalate. We have identified several compounds 1 Phthalate oxygenase 5080 18.6 that are substrates or effectors. For example, substitution of 174 3 2 Oxygen chlorine at the 4-position(which is normallyoxygenated), 1.74 5300 3 Cytochrome c decreases the ratio of product formed to O2 consumed. In 4 Ferricyanide 1.74 4140 contrast, 3,4-dicarboxypyridine (the analogue in which nitro3300 5 Dichlorophenolindophenol 1.74 gen is substituted for carbon at position 4) stimulatesoxygen reduction at a rate similar to that caused by phthalate, but Phthalate oxygenase reductase is an efficient catalyst of this analogue is not oxygenated. 2,3-Dicarboxypyridine, howelectron transfer from NADH to a wide variety of electron ever, is oxygenated, albeit at a rate lower than thatof oxygen acceptors(TableVIII).Theturnovernumber with cyto- reduction. We plan to characterize thecompounds produced chrome c washigher than with phthalate oxygenase. The from these phthalate analogues (and those from other anareductase reduces O2 at a very low rate, a common character- logues, astheyareidentified).Thesestudiesshouldhelp istic of electron transferases(53). Phthalate oxygenase reduc- delineate the stearic, ionic, and electronic features that: (a) tase is specific for NADH, as NADPH did not support meas- facilitateformation of areduced activated oxygen species urable activity with either phthalate oxygenase or cytochrome (common to effectors and substrates) and( b ) enable oxygenC. ation (found only on substrates). Phthalate oxygenase bears a strong resemblance to several DISCUSSION other oxygenases isolated from pseudomonads and occasionThe two components of the phthalate oxygenase system ally from othermicroorganisms. The oxygenases in thisgroup have been purified. A 217-kDa protein made up of 48-kDa have Rieske-type iron-sulfur centers (as indicated by their a requiremonomers was found t o have oxygenase activity. The active distinctive EPR andvisible absorbance spectra) and for activity. Each of these systems oxygenase has botha Rieske-type [2Fe-2S] center and another ment for an additional iron dissociable iron, apparently Fez+. Removal of this Fez+ realso includes a flavo-iron-sulfur electron transfer system (on sulted inreversible inactivation of the oxygenase. one or more separate proteins) that supplies electrons from Although the reducedform of the isolated oxygenase is NADH for the oxygenation. The Rieske-type iron-sulfur oxcompetentto effect phthalate oxygenation, theassociated ygenases are most commonly dioxygenases that produce cisoxidoreductase is required for catalytically significant ratesof diols. Examples of such systems include the oxygenases for: oxygenation. This electron transfer protein has both FMN benzene (22-24), benzoate (28), toluene (25,26), naphthalene and a chloroplast-ferredoxin-type [2Fe-2S] center associated (271, and pyrazon (27). In addition, a monooxygenase conwith a 34-kDa monomeric polypeptide. The oxidoreductase taining a Rieske-typeiron-sulfurcenterhasbeenstudied catalyzes electron transfer from NADH to a wide variety of extensively by Bernhardt and colleagues (29, 30, 34, 35): 4electron acceptorsincluding phthalate oxygenase, cytochrome methoxybenzoate demethoxylase (putidamonooxin).The twoc, ferricyanide, and dichlorophenolindophenol. component vanillate 0-demethylase activity investigat,ed by TABLE VI1 Substrate specijicity ofphthulate oxygenase

1516

Purification of a Phthsalate Oxygenase System

Ribbons (62) may also belong to thisclass of enzymes. These oxygenases are generally large proteins of greater of sodium dodecyl than 100 kDa butdissociate in the presence sulfate into smaller peptides. Phthalate oxygenase appears to be a tetramer of identical 48-kDa monomers. Iron and S2content are consistent with one [2Fe-2S] center/monomer. Several of the enzymes of this class have two types of subunits with one Rieske center/a@ unit (22-29, 47). Tertiary structures for these enzymes reported include a& (21, 26, 27,48) or a& (47). Putidamonooxin is apparently unique in having only 2 subunits/holoenzyme (29). The oxidoreductase component of the phthalateoxygenase system has been found to have both FMN and an [2Fe-2S] center bound to amonomeric 34-kDaprotein.Phthalate oxygenase reductaseresembles the putidamonooxin reductase, which has both FMN and an [ZFe-ZS] center in a 42kDa protein (30,34). Both benzoateoxygenase reductase (31, 32) and methanemonooxygenase reductase (49,50)have FAD and an[2Fe-2S] center on one peptide. The pyrazon,benzene, in and tolueneoxygenase systems have two separate proteins their electron transfer chains: one with flavin and one with an iron-sulfur center (22-25,27). Naphthalene oxygenase may have a similar two-component electron transfer system(33). The phthalate oxygenase system from P. cepacia provides an excellent opportunity to characterizemechanistically this biologically important class of enzymes. Both the oxygenase and the reductase can be readily purified in good yield from P. cepacia. Both enzymes are stable underaerobic conditions at 4 "C; the oxygenase retains good activity at 25 "Cfor hours. Steady state and rapid reaction kinetic studies are now in progress. In addition, genetic studies have commenced. We have previously reported crystallization of the reductase (54) and spectroscopic studies on the oxygenase (52). Acknowledgments-We would like to thank C. P. Mountjoy for his assistance and advice during fermentation and purification. Drs. J. Fee and T. Yoshida now at Los AlamosNational Laboratory recorded the EPR spectra of the oxygenase and the reductase. We are grateful for the following gifts: ferredoxinand ferredoxin-NADP reductase from Dr. H. Kamin of Duke University, cytochrome P-450 reductase from Dr. M. J. Coon of the University of Michigan, 4-hydroxyphthalate from Dr. J. Powlowski of the University of Michigan, and 4,5dihydroxyphthalate and the P. cepaciu strain from Dr. D. Ribbons of Imperial College, London.

1. 2. 3. 4. 5.

6. 7. 8.

9. 10.

REFERENCES Cartwright, N. J., and Smith, A. R. W. (1967) Biochem. J. 102,826-841 Clarke P. H. (1982) Antonie Leeuwenhoek J. Microbiol. 48,105-130 Dagle;, S. (1975) Am. Sci. 63,681-689 Wood, J . M. (1982) Enuiron. Sei. Technol. 16,291A-297A Keyser, P. K., Pujar, B. G., Eaton, R. W., and Ribbons, D.W. (1976) Enuiron. Health Perspect. 18, 159-166 Gibson, D. T. (1972) Crit. Reu. Microbiol. 1, 199-223 Jeffrey, A. M., Yeh, H. J. C., Jerina, D. M., Patel, T. R., Davey, J. F., and Gibson, D. T. (1975) Biochemistry 14,575-584 Jerina, D. M.,,Daly, J. W., Jeffrey, A. M., and Gibson, D. T. (1971) Arch. Biochem. Rlophys. 142, 394-396 Gibson, D. T., Hensley, M., Yoshioka, H., and Mabry, T. J. (1970) Biochemistry 9, 1626-1630 Akhtar, N. M., Boyd, D. T., Thompson, N. J., Koreeda, M., Gibson, D. T., Mahadevar, V., and Jerina, D. M. (1975) J . Chem. SOC.Perkins Trans.I,

2506-2511 11. Laborde, A. L., and Gibson, D. T. (1977) Appl. Enuiron. Microbiol. 34, 783-790 12. Koreeda, M., Akhtar, A. N., Boyd, D. R., Neil, J. D., Gibson, D. T., and Jerina, D. M. (1978) J. Org. Chem. 43,1023-1027 13. Kobal, V.M., Gibson, D. T., Davis, R. E., and Garza, A. (1973) J. Am. Chem. SOC.95,4420-4421 14. Gibson, D. T., Cardini, G. E., Maseles, F.C., and Kalio, R.E. (1970) Biochemistry 9,1631-1635 15. Ribbons, D. W., and Evans, W. C. (1960) Biochern. J. 76,310-318 16. Nakazawa, T., and Hayashi, E. (1977) J. Bucteriol., 131,42-48 17. Pujar, B. G., and Ribbons, D. W.(1985) Appl. Enuwon. Mmobcol. 49,374376 18. Bull, C., and Ballou, D. P. (1981) J. Biol. Chem. 256,12673-12680 19. Fee, J. A,, Findlmg, K. L., Yoshlda, T., Hllle, R., Tarr, G. E., Hearshen, D. O., Dunham, W. R., Day, E. P., Kent, T. A,, and Miink, E. (1984) J. Biol. Chem. 259, 124-133 20. Keyser, P. K. (1976) Ph.D. Dissertation, University of Miami 21. Ensley, B. D., and Gibson, D. T. (1983) J. Bucteriol. 155,505-511 22. Axcell, B. C., and Geary, P. J. (1975) Biochem. J. 146,173-183 23. Crutcher, S. E., and Geary, P. J. (1979) Biochem. J. 177,393-400 24. Geary, P. J., Saboowalla, F., Patil, D., and Cammack, R. (1984) Biochern. J. 217, 667-673 25. Yeh, W. K., Gibson, D. T., and Liu, T.-N. (1977) Biochem. Biophys. Res. Commun. 78,401-410 26. Subramanian, V., Liu, T.-N., and Gibson, D. T. (1979) B k h e m . Biophys. Res. Commun. 91, 1131-1139 27. Sauber, K., Frohner, C., Rosenberg, G., Eberspacher, J., and Lingens, F. (1977) Eur. J. Biochem. 74,89-97 28. Yamaguchi, M., and Fujisawa, H. (1980) J. Biol. Chem. 255,5058-5063 29. Bernhardt, F.-H., Heymann, E., and Traylor,P. S. (1978) Eur. J. Biochem. 92,209-223 30. Twilfer, H., Bernhardt, F.-H., and Gersonde, K. (1981) Eur. J. Biochem. 119,595-602 31. Yamaguchi, M., and Fujisawa, H. (1978) J. Biol. Chem. 253,8848-8853 32. Yamaguchi, M., and Fujisawa, H. (1981) J. Biol. Chern. 256,6783-6787 33. Ensley, B. D., Glbson, D. T., and Laborde, A. L. (1982) J. Bucteriol. 149, 948-954 Pachowsky, H., and Staudinger, H. (1975) Eur. J. 34. Bernhardt, F.H., Biochem. 57,241-256 35. Bernhardt, F.-H., and Kuthan, H. (1983) Eur. J. Biochem. 130,99-103 36. Margoliash, E., and Walasek, 0.F. (1967) Methods Enzymol. 10,339-348 37. Massey, V. (1957) J. Biol. Chem. 229,763-770 38. Carter, P. (1971) Anal. Biochem. 40,450-458 39. Rabinowitz, J. C. (1978) Methods Enzymol. 53,275-277 40. Faeder, E. J., and Siege], L. M. (1973) Anal. Biochem. 53, 332-336 41. Whitby, L. G. (1953) Biochqn. J. 54,437-442 42. Goa, J. (1953) S c a d . J. Cltn. Lab. Inuest. 6 , 218-222 43. Lowry, 0.H., Rosebrough, N. J., Farr, A. L., and Randall, R. J. (1951) J. Biol. Chem. 193,265-275 44. Kolthoff, I. M., Shore, W. S., Tann, B. H., and Matsuoka, M. (1965) Anal. Biochem. 12.497-508 45. Riddles, P. W., Blakeley, R. L., and Zenner, B. (1983) Methods Enzymol. 9 1,49-60 46. Batie, C. J., and Kamin, H. (1981) J. Biol. Chem. 256,7756-7763 47. Yamaguchi, M., and Fujisawa, H. (1982) J. Biol. Chem. 257, 12497-12502 48. Ziffer, H., Kabato, K., Gibson, D. T., Kobal, V. M., and Jerina, D.M. (1977) Tetrahedron 33,2491-2496 49. Colby, J., and Dalton, H. (1979) Biochem. J. 177,903-908 50. Lund, J., and Dalton, H. (1985) Eur. J. Biochem. 147, 291-296 51. Chapman, P. J. (1972) in Degrdation of Synthetic Organic Molecules in the Biosphere, pp. 17-55, National Academy of Sciences, Wash. D. C. 52. Cline, J. F., Hoffman, B. M., Mims, W. B., LaHaie, E., Ballou, D. P., and Fee, J. A. (1985) J. Biol. Chem. 260, 3251-3254 53. Massey, V., Muller, F., Feldberg, R., Schuman, M., Sullivan, P. A,, Howell, L., Mayhew, S. G., Matthews, R. G., and Foust, G. P. (1969) J. Biol. Chem. 244,3999-4006 54. Correll, C. C., Batie, C. J., Ballou, D. P., and Ludwig, M. L. (1985) J . Biol. Chem. 260, 14633-14635 55. Williams, C.H., Jr., Arscott, L.D., Matthews, R. G., Thorpe, C., and Wilkinson, K. D. (1979) Methods Enzyml. 62,185-198 56. Tarr, G. E. (1986) in Methods of Protein Microcharacterization (Shively, J. E., ed), pp. 155-194, Humana Press Inc., Clifton, NJ 57. Holmquist, B., and Vallee, B. L. (1973) Biochemistry 12,4409-4417 58. Laemmli, U. K. (1970) Nature 227,680-685 59. Taylor, B. F., and Ribbons, D.W. (1983) Appl. Enuiron. Microbiol. 46, 1276-1281 60. Palmer, G. (1973) in Iron-Sulfur Proteins (Lovenberg, W., ed) Vol. I, pp. 285-325, Academic Press, Orlando, FL 61. Orme-Johnson, W. H., and Sands, R. (1973) in Iron-Sulfur Proteins (Lovenberg, W., ed) Vol. 11, pp. 195-238, Academic Press, Orlando, FL 62. Ribbons, D. W. (1970) FEES Lett. 12, 161-165

Purification of a Phthalate Oxygenase System

M: Weber, personal communication. Dr. R. Hiile, persona1 communication.

1517

1518

Purification of a Phthalate Oxygenase System

a b c a b d a b6.2 8.6 6.S 1S.6 12.b

25.2 7.5 8.0

17.2

lS.9 21.9 6.9

11.0 10.9

27.7

2b.3

17.7

11.8

IC.) 29.0 11.1 3.9 18.I

11.1 21.3 6.7

36.7

17.8 B7.8 12. I 20.6 20.6

17.7

5.3

12.6

7.0

21.1 22.1