Specific α-bridge cleavage by heme oxygenase of [α-14C]deuterohemin IX, [α-14C]hematohemin IX and 2,4-diacetyl[α-14C]deuterohemin IX

Biochimica et Biophysica Acta, 704 (1982) 261-266 Elsevier Biomedical Press

261

BBA 31176

SPECIFIC a - B R I D G E CLEAVAGE BY H E M E OXYGENASE O F [ a - t 4 C ] D E U T E R O H E M I N IX, [ a - t4 C ] H E M A T O H E M I N IX AND 2,4-DIACETYL[ a - 14C ] D E U T E R O H E M I N IX IRENE REZZANO, MARIA L. TOMARO, GRACIELA BULDAIN and ROSALiA B. FRYDMAN *

Facultad de Farmacia y Bioquimica, Universidad de Buenos Aires, Junin 956, Buenos Aires (Argentina) (Received December 7th, 1981)

Key words: Bilirubin metabolism," a-Bridge cleavage; Heme oxygenase; Hemin; (Rat liver)

The enzymatic oxidations of [a-t4C]deuterohemin IX, [a-14C]hematohemin IX and 2,4-diacetyl[a-14C] deuterohemin IX were carried out by using a microsomal heine oxygenase system from rat liver in combination with the biliverdin reductase from the same origin. In every case the bilirubins formed were devoid of radioactivity, indicating the a-selective oxidation of the three hemins. Hematohemin IX was oxidized at the highest rate, followed by deuterohemin IX and 2,4-diacetyldeuterohemin. When the three heroins were preincubated with microsomal heme oxygenase in the absence of NADPH, and the latter was added after the preincubation period, it was found that the enzymatic oxidation of the heroins was inhibited. Therefore, for the maximal rate of oxidation both hemin and NADPH must he present simultaneously. In the presence of heroin IX (the natural substrate), the enzymatic oxidation of the synthetic heroins was inhibited. The oxidation of 2,4-diacetyldeuterohemin IX was the most inhibited, while the oxidation of hematohemin IX was affected to a much lesser degree. These results are in agreement with the higher affinity ( K m : 150/tM) of hematohemin IX for the enzyme, as compared to 2,4-diacetyideuterohemin IX (K m=660 /tM) and deuterohemin IX (K= = 330/t M).

Introduction Heme oxygenase is a microsomal enzyme which oxidizes heroin IX 1 (Fig. 1) to a-biliverdin. The latter is then reduce~l in mammals by a biliverdin reductase to a-bilirubin (Fig. 1), which is then released as its glucuronide conjugates into the biliary tract. Both heme oxygenase and biliverdin reductase were first isolated from rat liver [1] and were recently obtained in pure forms [2-5]. We studied the substrate specificity of heme oxygenase and found that only heroins which carry two vicihal propionic side chains at C 6 and C 7 are substrates of heme oxygenase [6,7]. In the case of

* To whom correspondence should be addressed. 0167-4838/82/0000-0000/$02.75 © 1982 Elsevier Biomedical Press

heroin XIII and heroin III where the selectivity of the bridge cleavage was examined it was found that the oxygenase was always a-selective, i.e., only the a-meso carbon was oxidized with exclusive formation of a-biliverdins [6,7]. Biliverdin reductase reduced the a-biliverdins derived from the enzymatic oxidation of the synthetic heroins to the corresponding a-bilirubins. The same results were obtained in vivo when heroins XIII and III were perfused through rat liver [8]. Deuterohemin IX 2 and several 2,4-diacyldeuterohemins IX were also found to be substrates of heine oxygenase, and the biliverdins formed were substrates of the biliverdin reductase. This was also the case of hematohemin • IX 3 and its higher homologues [7]. However, the a-seiective oxidation of these substrates was not established. The question of the selective oxidation

262

of the a-bridge of hemins by heme oxygenase is closely linked to the enzyme's mechanism. The chemical oxidation of heme IX pyridine hemochrome (the stable Fe 2+ form of hemin IX 1) produces all the four biliverdin isomers, and b]liverdin IX-a is preponderant in the mixture (32%) [9]. The same chemical oxidation of deuterohemin IX 5 also produces the four isomeric deuterobilivercnns, but the a-isomer is the minor product (10%), while the 8-isomer is the major product (43%) [9]. It is therefore of importance to establish whether the enzymatic oxidation of deuterohemin IX 2 is also a-selective, since the structural and eleci-ronic factors in deuterohemin IX 2 do not favor the formation of the a-isomer as is the case with heme IX. Since heme oxygenase is a non-heme protein [10], it is conceivable that the formation of the enzyme-substrate complex will be a hemoprotein, where the crevice of the enzyme imposes the a-selective oxidation which is a property of the heme binding site of the oxygenase. This bridge specificity imposed by the enzyme should prevail over the structural features of deuterohemin which direct the oxidation to the 8-bridge. In the case of the 2,4-diacetyldeuterohemin IX 4 (Fig. 1) which is also a substrate of heme oxygenase [7], the enzyme-imposed bridge specificity should also prevail over the deactivating effect of the electron withdrawing substituents, and an a-biliverdin should also be formed. In the case of hematohemin IX 3 where two hydrophilic substituents replace the-vinyl side-chains, an exclusive oxidation at the a-bridge will again lend support to a decisive influence of the heme binding site of the protein. These questions were examined by preparing aJ4C-labeUed deuterohemin IX 2, 2,4-diacetyldeuterohemin IX 4 and hematohemin IX 3 which were then used as substrates of the heme oxygenase system.

O

R

Herne o x y g e n a s e

, 1; R = C H = ¢ H 2 2 ; R =H

CO2H 3; R = C H O H C H 3 4; R = COCH 3

(~O2H

at - Bitiver'dins rEliliver din educt.ose

0

HNC)~

~R

L co~

2H

c(- B ; l i r u b i n s

Fig. 1. The enzymatic oxidation of hemins to a-biliverdins and their reduction to a-bilirubins. I, Hemin IX; 2, deuterohemin IX; 3, hematohemin IX; 4, 2,4-diacetyldeuterohemin IX.

were prepared by synthesis as outlined in Methods. All the other reagents and solvents were of analytical grade. TLC was performed on precoated silica gel 60 F 284 plates (Merck, 0.25 mm layer thickness).

Methods Synthesis of [a-14C]porphyrins and hemins. A Vilsmaier type formylation of pyrrole 5 with [14C]

H 5

H 6

Hoc

:oc H

8

H 7

,yf g; R.CO CH2ffn 10 ;R. C~K~

Materials and Methods •

Materials Heroin IX 1 and NADPH were purchased from Sigma Chemical Co. [a-14C]Deuterohemin IX 2, [a-14C]diacetyldeuterohemin IX 4 and [a-14(~] hematohemin IX 3 (Fig. 1) were obtained from the corresponding porphyrin dimethyl esters which

=Me ~ e

11

PMe=CH2CH2CO2CH3 • = C14

121 R °C02C~s 13; R=H 14; R ,COCH 3 15; R. CHOHCH3

Fig. 2. Synthetic outline for the preparation of [a-14C] deuteroporphyrin dimethyl ester _13_.

263

dimethylformamide (1 mCi) afforded the 2-formylpyrrole 6 (2.6.106 dpm/mg) (Fig. 2). The latter was reduced to the alcohol 7 with sodium borohydride, and 7 was condensed with the pyrrole 8 to afford the dil:,yrrylmethane 9 (9.9. 10.5 dpm/mg). Hydrogenolysis of the benzyl esters of 9 followed by decarboxylation and formylation afforded the 5,5'-diformyldipyrrylmethane 10, which was condensed with the 5,5'-dicarboxydipyrrylmethane 11. The resulting porphyrin 12 (1.2 • 105 dpm/mg) was decarboxylated by heating in 10% HC1 at 200°C during 8 h, and then reesterifled to obtain [a-~4C]deuteroporphyrin dimethyl ester 13 (1.4.105 dpm/mg). Porphyrin 13 was transformed into its copper chelate and acylated to 14 as described elsewhere [11]. The 2,4-diacetyldeuteroporphyrin IX 14 was reduced to 15 with sodium borohydride in methanol/methylene chloride. The hemins of [a-14C]deuteroporphyrin IX dimethyl ester 13, 2,4-diacetyl[a-14C]deuteroporphyrin IX d-i~nethyl, ester 14 and [a-14C] hematoporphyrin IX dimethyl ester 15 were obtained by incorporating iron as described elsewhere [12]. The hemin dimethyl esters were purified by low-pressure chromatography on a column packed with TCL silica gel 60 (Merck) using 5% methanol in chloroform. After saponification with 1 M KOH in methanol, the hemins were again purified by the same method using a mixture of chloroform/methanol/water (48 : 28 : 6 v/v). Microsomal heine oxygenase from rat liver. Wistar albino female rats (180-200g) were injected subcutaneously with a single dose of cobaltous cloride (160 mg/kg) and were later fasted for 19 h. They were then anesthetized with ether and the bleached livers were excised and repeatedly washed with an ice-cold saline solution. All further operations were carried out at 0-4°C. The livers were homogenized in a PotterElvehjem-type glass homogenizer in 3 vol. ice-cold 0.25 M sucrose solution containing 50 mM phosphate buffer (pH 7.4). The homogenate was centrifuged at 20000 × g for 15 rain and the supernatant was further centrifuged at 105000 × g during 60 min. The microsomal pellet was then resuspended in 1.5 vol. 0.15 M KC1 in 50 mM phosphate buffer (pH 7.4), and was again centrifuged at 105000 × g for 30 rain. This second pellet was

then resuspended in the same buffer to a volume so as to obtain a protein concentration of 15 mg/ml of suspension. This enzyme preparation was used as the microsomal heme oxygenase. It was usually used during the first 48 h when its activity with the synthetic iron-porphyrin was assayed. Its activity usually decreased to about 50% after 5 days storage at -20°C. Biliverdin reductase from rat liver. The 105 000 × g supernatant obtained from the microsomal preparation was fractionated by addition of ammonium sulfate (AS), and the 40-60% AS fraction was dissolved in 10 mM phosphate buffer (pH 7.4) and dialyzed against the same buffer. It was used as the biliverdin reductase. Assay of microsomal heine oxygenase. The standard incubation mixture contained, in a final volume of 200/~1; 10 /~mol Tris-HC1 buffer (pH 7.4), 140 nmol NADPH, 50 #1 of the microsomal heme oxygenase (0.75 mg protein), 50/~1 biliverdin reductase (0.42 mg protein, 44 nmol bilirubin formed/30min) and 70 nmol [a-14C]hemin. The latter (2 mg) was dissolved in 0.1 ml of 0.1 M sodium hydroxide and adjusted to pH 7.4 with phosphate buffer. The incubations were usually

10c

OC

~

4c

o

I

1 I I 3O 0O Time (rain) Fig. 3. Enzymatic oxidation rates of [a-14C]hemins. The incubations end measurements were carried out as described. A,, hematohemin 3, (3, deuterohemin 2; 0 , 2,~-diacetyldeuterohemin 4.

264 TABLE I SPECIFICITY O F T H E E N Z Y M A T I C B R I D G E C L E A V A G E The incubations were performed using [a- 14C]deuterohemin IX, [ a- 14C]hematohemin IX and 2,4-diacetyl[a- 14C]deuterohemin IX (400 nmol, spec. act. 7.5.104 dpm//~mol). Bilirubins were isolated by TCL. Substrate

[a-J4C]Deuterohemin IX 2 [a-14C]Hematohemin IX 3 Diacetyl[a- 14C]deuterohemin IX _4

Enzymatic system a

-- N A D P H Complete - NADPH Complete - NADPH Complete

Total radioactivity in the incubation (dpm)

CO loss (~)

a-Bilirubins nmol b

dpm

30 000 12 000 30 000 ! 0 500 30 000 ! 8 000

-60 -65 -40

none 240 none 260 none 160

17 14 21 c 17 20 19

a Incubated during 30 min. b Determined from the extracts of TLC plates. ¢ Backgrounddpm.

carried out for 30 rain at 37°C. Blanks omitting NADPH were run simultaneously. H e m e oxygenase activity was measured by the decrease of radioactivity in the incubation mixture as compared with the control runs where N A D P H was omitted. In the case of the experiments summarized in Table I and Fig. 3 a 5-fold incubation mixture as described above was used. Hemin consumption was measured on an aliquot taken at the end of the incubation in the runs described in Table I. In the case of Fig. 3, aliquots were taken at the indicated times. When bilirubins were estimated in the experiments described in Table I, the incubations were stopped by adding 1 ml 100 mM glycine-HCl buffer (pH 1.8) to 200 #1 of the incubation mixture, followed by 1.5 ml of an NaCI saturated solution containing 100 m g / m l of ascorbic acid. The solution was saturated with solid NaCI and extracted with chloroform. The chloroform extracts were evaporated, the residues were applied to silica gel plates, and developed with chlorof o r m / m e t h a n o l / w a t e r (48 : 28 : 6 v/v). The bilirubin bands were eluted from the silica with 10% methanol in chloroform and bilirubin concentration was measured by the A of absorbance between the maximum peak of the corresponding bilirubin (440 nm for hematobilirubin and 455 nm for the other two bilirubins), and the absorbance at 520 nm [7]. Standard concentration curves of

authentic samples of the corresponding bilirubins were used as references. The radioactivity of the eluted bilirubin bands was measured in a liquid scintillation counter using Bray's solution. The T L C runs also allowed the recovery of the nonoxidized hemins. Results

Specific enzymatic oxidation of the a-methine bridge of deuterohemin I X 2, 2,4-diacetyldeuterohemin I X 4 and hematohemin I X 3 [ ' a - 1 4 C ] D e u t e r o h e m i n - I X 2, [a-14C]diacetyl deuterohernin IX 4 and [a-14(~]hematohemin IX 3 were oxidized enzymatically to the corresponding bilirubins. The oxidation was performed by using the microsomal heme oxygenase system from rat liver in combination with the biliverdin reductase from the same origin. In every case the bilirubins formed were devoid of radioactivity (Table I), and were therefore the a-bilirubins (Fig. 1). The substitution of the vinyl side-chains by hydrogens (as in 2) did not affect the a-selectivity of heine oxygenase. The latter was not affected either by the substitution of the vinyls with electron-withdrawing side-chains (as in 4), or by nonhydrophobic side chains (as in 3). The use of hemins specifically labelled at the ot-meso position allowed an exact estimation of the amount of heroin oxidized by heme oxygenase.

265

The three heroins were good substrates of heme oxygenase and their relative oxidation rates were similar to those obtained by other methods [7]. The oxidation rates of [ct-m4C]deuterohemin 2, [a14C]diacetyldeuterohemin 4 and [ a ~ 4 C ] hematohemin 3 are shown in Fig. 3. Hematohemin IX 4 was the best substrate, followed by deuterohemin IX 2 and 2,4-diacetyldeuterohemin IX 3. An apparent Km~---150 ~M was found for hematohemin IX 4 in Tris-HC1 buffer (pH 7.4). A K m of 330 /~M was determined for deuterohemin IX 2 and of 660 FM for the 2,4-diacetyldeuterohemin IX 3. The Vm=, values were 33, 33 and 50 nmol heroin consumed/rag protein per 15 rain, respectively.

Inhibition of heme oxygenase by preincubation with the substrates in the absence of NADPH Our previous work [7] suggested that both hernin and NADPH must be present simultaneously for

the enzymatic oxidation to take place. When the microsomal heme oxygenase system was preincubated with heroin IX 1 in the absence of NADPH and the latter was added after the preincubation period, the enzymatic oxidation of the heroin was found to be inhibited. The synthetic [a-~4 C]hemins allowed a further examination of these results. The microsomal heine oxygenase system was preincubated with [a-14C]deuterohemin IX, [,~14C]hematohemin IX and 2,4-diacetyl[a-14C] deuterohemin IX in the absence of NADPH during different time periods. The system was then completed by addition of NADPH and incubated during 30 min. As can be seen in Fig. 4, the three heroins inhibited heine oxygenase activity by preincubation in the absence of NADPH. Hematohemin IX 3 was the strongest inhibitor, as could be expected-from its higher affinity for the enzyme. These results lend support to the suggestion [7] that both hem_in and NADPH must be present simultaneously for the enzymatic reaction to take

lOO

IOC

8o

8c

~ 6o

6C

~

4c

.|

.E

r~

0

10

20

30

Time of preincubotion (min) Fig. 4. Inhibition of heine oxygenase by substrate in the absence of NADPH. The enzyme was preincubated at the in-

dicated times with: A, hematohemin; O, deuterohemin; O, 2,4-diacetyldeuterohemin. NADPH was added after the preincubation period and the incubations were carried out during 30 rain. The control was the complete system without preincubation.

20--

30 Hemin ~

90 ( nmol )

150

Fig. 5. Effect of heroin IX on the enzymatic oxidation of the synthetic [a-14C]hemins. The complete system with the addition of the indicated amounts of heroin IX was incubated as described. The substrates used were: A, [a-m4C]hematohemin; O, [a-14C]deuterohemin; 0 , 2,4-diacetyl[a-14C]deuterohemin. The control was the complete system without addition of heroin IX.

266

place. The inhibition is obviously specific, since it decreases with the decreasing affinities of the hemins for the oxygenase.

Effect of heroin I X on the enzymatic oxidation of [a-t4C]deuterohemin IX, [a-HC]hematohemin I X and 2, 4-diacetyl [a- H C ]deuterohemin I X When the enzymatic oxidation of the labelled synthetic hemins was measured in the presence of the natural substrate (heroin IX 1), it was possible to measure the inhibitory effect of the latter on the enzymatic oxidation of the synthetic hemins. It was found that hematohemin IX 3 was affected to a lesser degree than the other two hemins by the presence of heroin IX 1 (Fig. 5). Even in the presence of equivalent amounts of hemin IX 1 and hematohemin IX 3, the enzymatic oxidation o f the latter decreased only by 35%. The enzymatic oxidation of 2,4-diacetyldeuterohemin IX 4 was 80% inhibited under the same conditions (l~ig. 5). These results are in agreement with those previously found on the strong inhibitory effect of hematohernin IX 3 on the enzymatic oxidation of hemin IX 1, both in vitro [7] and in vivo [8]. Discussion

This report conclusively shows that heme oxygenase oxidises the a-methine bridge of hemins irrespective of the nature of the side-chains in rings A and B. The structure of the crevice where the vinyl side-chains of heme IX 1 bind is as yet unknown, but it has clearly a grea/-elasticity, since the specific oxidation at the a-bridge is not affected either by the absence of the vinyl groups or by its substitution by hydrophilic or deactivating side-chains. The synthesis of [a-J4C]hemins allowed the confirmation of observations on the mechanism of heme oxygenase which were suggested on the basis of other measurements. The most remarkable effect is the need for the simultaneous presence of heroins, NADPH and enzyme to achieve the enzymatic oxidation to a-biliverdins.

In the absence of the reducing agent an inactivation of the enzyme takes place (Fig. 4), probably due to the binding of the non-reduced hemin to the protein. This is a specific effect which can only be found with the heroins that are substrates of heme oxygenase (see also Ref. 7). Hematohemin 3 competes very efficiently with the oxidation of-the natural hemin IX 1. This result lends support to the data obtained by perfusing rat liver with hematohemin and hemin IX [8], where an inhibition of the degradation of the natural heroin was found. Acknowledgments This work was made possible by a grant from the National Institutes of Health (U.S.A.), and in part by the Fundacirn Lucio Cherny. References I Tenhunen, R., Marver, H.S..and Schmid, R. (1969) J. Biol. Chem. 244, 6388-6394 2 Maines, M.D., Ibrahim, N.G. and Kappas, A. (1977) J. Biol. Chem. 252, 5900-5903 3 Yoshida, T. and Kikuchi, G. (1979) J. Biol. Chem. 254, 4487-4491 4 Noguchi, M., Yoshida, T. and Kikuchi, G. (1979) J. Biochem. (Tokyo) 86, 833-848 5 Krishnan Kutty, R. and Maines, M.D. (1981)J. Biol. Chem. 256, 3956-3962 6 Frydman, R.B., Awruch, J., Tomato, M.L. and Frydman, B. (1979) Biochem. Biophys. Res. Commun. 87, 928-935 7 Frydman, R.B., Tomato, M.L., Buldain, G., Awruch, J.. Diaz, L. and Frydman, B. (1981) Biochemistry 20, 51775182 8 Awruch, J., Lemberg, A., Frydman, R.B. and Frydman, B. (1982) Biochim. Biophys. Acta 714, 209-216 90'Carra, P. (1975) in Porphyrins and Metalloporphyrins (Smith, K.M., ed.) pp. 123-153, Elsevier, Amsterdam 10 Yoshida, T. and Kikuchi, G. (1978) J. Biol. Chem. 253, 4224-4229 I1 Brockmann, H., Bliesner, K.M. and Inhoffen, H.H. (1968) Justus Liebigs Ann. Chem. 718, 148-161 12 Cavaleiro, J.A.S., Rocha Gonqalvez, A.M.d'A., Kenner, G.W. and Smith, K.M. (1974) J. Chem. Soc. Perkin Trans I. 1771-1781