Vol. 91, No. 3, 1979 December

BIOCHEMICAL

AND BIOPHYSICAL

RESEARCH COMMUNICATIONS

14, lW9

Pages

991-996

THE ENZYME "ALDEHYDE OXIDASE"

IS AN IMINIDM OXIDASE. l'(5') REACTION WITH NICOTINE A IMINIUM ION Svante

Department Receiyed

Brandlnge

and Lars

of Organic Chemistry, Stockholm, S-106

October

22,

Lindblom

Arrhenius Laboratory, 91 Stockholm, Sweden

University

of

1979

SUMMARY: The second of the two reaction steps involved in the metabolic transformation of (-)-nicotine to (-)-cotinine (2) (i.e., the oxidation of the intermediate 2) is mediated mainly, if not solely, by the enzyme aldehyde oxidase (EC 1.2.3.1). Of the molecular species that constitute 2, nicotine A1'(5') iminium ion (Za) appears to serve as the substrate. The enzyme has a strong affinity for Yj, as shown in a study on-the inhibition of the oxidation of 3-(aminocarbonyl)-1-methylpyridinium chloride. This study gave a value of Ki = = 6 uM ; Km = 2 pM (pH 7.4). Mainly in view of this finding, "iminium oxidase" seems to be a more adequate name than "aldehyde oxidase" for this enzyme. The metabolic bolite the

transformation

(-)-cotinine

(2)

intermediate

catalyzed

(-)-nicotine

in mammals is

2 (1,2).

by the

of

The first

cytochrome

a two-step reaction

P-450

(1)

system

into

its

reaction

step

seems

and the

principal

which

meta-

proceeds

via

to be hydroxylation

product

formed

initially

should

thus

be -2b (1,3). An equilibrium mixture (z), the main components of which are iminium ion -2a and the carbinolamine 2&, is then rapidly generated from -2b (3,4). Other species hypothetically present are -2c and 2cJ, but the amounts of these are so low that they could not be detected by NMR (4). The composition of

the

-2 is dependent librium level solution

on pH and on other of 2

but

approximately

The second Hucker aldehyde

was found

reaction

variables

to be approximately

90 % in 0.2 step

as well.

in the

M sodium

Thus,

at pH 7.4

75 % in unbuffered phosphate

transformation

buffer

of 1 into

mouse

by purified

liver

rabbit

A recently

microsomes, liver

developed

diperchlorate

(A)

the

oxidation

enzymatic

of

the

aldehyde method

02,

and NADPH, could oxidase (EC 1.2.3.1)

for

the

synthesis

indeed (5). (4)

(4). 2 was found

ion -2a has made it possible of -2 to -3 more closely. The results

by

to be "an --in situ

be oxidized

to -3

of a crystalline

iminium

equi-

aqueous

by a soluble enzyme which was suggested --et al. to be mediated oxidase" (1). In 1972, Hill --et al. showed that 2, generated

from 1,

the

hydro-

to investigate are

reported

below.

MATERIALS AND METHODS The 2a hydrodiperchlorate (2) was synthesized A controlTxperiment demonstrated that autooxidation a yield of 2 of less than 0.5 % was obtained after through a solution of 2 in sodium phosphate buffer, NMNA = 3-(aminocarbonyl)-l-methylpyridinium

as described previously (4). of 4 to 3 proceeded slowly; occa.sTonal-bubbling of O2 pH 7.4, for 30 min at

chloride 0006-291 991

Copyright All rights

X/79/230991

-06$01.00/O

@ I979 by Academic Press. Inc. of reproduction in anyform reserved.

Vol. 91, No. 3, 1979

BIOCHEMICAL

PY “’

2a

AND

BIOPHYSICAL

RESEARCH COMMUNICATIONS

0+/ y

1;“’

py\\’ c-J--N

-

PY “’

OH

0 c-s N dH3 3

AH, 2b

CHO py = 3-pyridyl

PY “’

Fig.

1. Metabolic

transformation

PY”’

of

0\ y

(-)-nicotine.

35 OC. For the enzymic reactions the hydrodiperchlorate of 2a and the corresponding dichloride served equally well as starting material. A?airiy concentrated solution of 2a hydrodiperchlorate (4) (> 10 ml4) decolorized 2,6-dichloroindophenol (i.e.~,6-dichloro-4[(4'-hyd~oxyphenyl)iminol-2,5-cyclohexadien-l-one) within 2Gin at 22 oC but, as seen spectropolarimetrically at 600 nm (6), no reduction of the indophenol by 4 took place during 10 min under the conditions used for the determination of Km (Fig. 3). Reduction of potassium hexacyanothis compound was not suitable as an artificial electron ferrate(II1) was faster; acceptor. Due to the instability of aqueous solutions of 2a hydrodiperchlorate, these were kept on ice and used within 1 h. Determinations of the relative amounts of 2a and 2b at equilibrium in 0.2 M sodium phosphate buffer at pH 7.45, 8.45, and 9.20 wereperformed using the polarimetric method (4). The results are oxidase (EC 1.2.3.2) was purchased from presented in Table 1. Milk xanthine Sigma and assayed as previously described (7). The activity towards -2a hydrodiperchlorate was measured as for aldehyde oxidase (see below). Purification of aldehyde oxidase. Rabbit liver aldehyde oxidase was partially purified (90 times with a 30 % yield) by carrying out four of the six purification steps described by Felsted et a1.(8), with the exception (9) that the enzyme was eluted from the calcium phosphate gel with 0.05 M instead of 0.01 M phosphate buffer, pH 7.8. Protein determinations were performed according to Kalckar (10). The enzyme was assayed using 3-(aminocarbonyl)-l-methylpyridinium chloride (N1Inethylnicotinamide, NMNA), which is oxidized to a mixture of pyridones (8), and by following the increase in optical density at 300 nm as previously described Changing the buffer from 0.05 to 0.2 M sodium phosphate did not affect (11). the assay result. This assay procedure could not be applied, however, to the crude liver preparation (6,000 x g supernatant); so in this case the hexacyanoferrate(II1) method (11) was used instead. To check this latter procedure it was also used together with the former in assays after the first purification step. Activity with 2a hydrodiperchlorate as substrate was determined by incubating a 0.42 mM solzion with the enzyme preparations in 0.2 M sodium phosphate buffer, pH 7.4, for 4 min at 20-22 'C. Under these conditions the reaction was linear with time for at least 6 min. After 4 min, 0.5 ml 5 M NaOH and 0.62 I.rmol lidocaine were added and then NaCl(s) to saturation. The solution was extracted

992

Vol. 91, No. 3, 1979

BIOCHEMICAL

AND BIOPHYSICAL

with methylene chloride (4 x 2 ml) and The yield of cotinine 2 was determined loo-120 mesh, 0.2 x 180 cm) using the experiments described below the partially otherwise stated, the experiments were buffer.

the combined by GLC (3 % lidocaine as purified run at 20-22

RESEARCH COMMUNICATIONS

extracts dried with Na2S04. JXR on Gas Chrom Q, an internal standard. In all enzyme was used and, unless oC in 0.2 Y sodium phosphate

RESULTS AND DISCUSSION It has been demonstrated of catalyzing the oxidation determine the

whether

specific

any other

activities

(NMNA) as well

by Hill et al. that aldehyde oxidase is capable of -2 to -3 (5). However, it is also of interest to soluble enzyme is involved. To this end we measured

towards

as towards

of aldehyde

oxidase.

Since

practically

constant

(1.0,

-2a hydrodiperchlorate, the ratio between 1.0,

respectively) and steps l-4, cluded that aldehyde oxidase of -2. Inhibitors (12), had a marked

oxidation quinone

is

mM -4. The effectiveness

parable.

Milk

/min

(EC 1.2.3.2)

oxidized

0.14

question

for

the

be correct,

enzyme

clearly

experiment.

Another

rate

remained preparation it

estradiol

the

and E-benzo-

of -2a hydrodipermM estradiol

with

0.06

as an inhibitor like

can be con-

enzyme mediating

of oxidation

which,

was com-

aldehyde

was found

oxidase,

is

to be much less

(2)

to -3. An amount of this oxidized approximately 1 nmol/ from

contrast,

purification

starting

sole,

e.g.,

(13),

distinguishable

In striking

nmol/min

the

was obtained

-2a hydrodiperchlorate 2.4 pmol/min of xanthine not

in the

partial

activities

procedure,

not

of p-benzoquinone

oxidase

oxidized an amount

autooxidation

substrate

on the

molybdenum-containing

of 4,

hardly

xanthine

if

the

specific

purification oxidase,

An 85 % inhibition

in oxidizing

which

the

chloride

during the

and 1.2

the main,

affect

and 0.42

enzyme which

1.2,

of aldehyde

(4)

effective

1.4,

during

chlorate

an unspecific

to 2.

3-(aminocarbonyl)-l-methylpyridinium

that

a solution

obtained

in an

of aldehyde

oxidase

of NMNA converted

0.17 nmol/min of -4 into -3. 2a 2d is actually serving as -enzyme. Hill et al. (5) favored can -2c but this hypothesis since it would be expected to lead to the formation of 4is which

-methylamino-4-(3'-pyridyl)-butanoic

of

the

forms

acid,

a compound

which

is

not

converted

(14) into -3 under the conditions used. Instead, we have obtained evidence that 2a is the substrate. The evidence for this comes from a study on the pH dependence of the inhibitory action of -2a hydrodiperchlorate (4) on the oxidation of NMNA catalyzed by aldehyde oxidase (Fig. 2). It was initially observed that with 1.0 mM NMNA, 0.83 mM 2, and aldehyde oxidase the oxidation of NMNA was inhibited the

virtually

the

values for inhibition

the

oxidation

K,,,

completely,

same as when NNNA was omitted.

whereas the rate of cotinine In the investigation of

the

formation was pH dependence,

NMNA were determined at pH 7.45, 8.45, and 9.20 (Table 1) and exerted by a constant amount of -2a hydrodiperchlorate (4) - on of NMNA was studied at various concentrations of the latter and

993

Vol. 91, No. 3, 1979

BIOCHEMICAL

AND BIOPHYSICAL

RESEARCH COMMUNICATIONS

1.0

0

Fig. 2. Inhibition by 2a hydrodiperchlorate (4) (0.083 mM) of the oxidation of NMNA catalyzed by aldehyde oxidase at three-different pH’s. The measured results were analyzed using a program @MDP3R, University of California, Los Angeles, USA) for unweighted non-linear regression analysis. The model fitted was that of a linear competitive inhibition. Constants Km for NMNA and Ki for i are presented in Table 1.

at the

same three

constants

Ki

tive

amounts

pH's

have Table 6.7

were

pH values.

The inhibition

calculated

for

from

of -2a and -2b at equilibrium estimated by measuring

been 1.

Calculation

x 0.85

= 5.7;

former set only is i = -2a hydrodiperchlorate. PR

5

the

kinetic

in 0.2 the

to be competitive data

M phosphate

optical

rotation

(Fig.

2).

buffer

and rela-

of various

as a function

of sets of Ki for & and 2. The.sets are obtained thus: x 0.15 - 1.0, etc. The approximate constancy of the evidence that 2a rather than2 is the true inhibitor. Ki i

for

Molar ratio --2a/2b

(N-0

Calculated for 2a (PM) -

Ki

Calculated K. for 2b ‘(PM) -

0.23

+ 0.05

6.7

-+ 1.0

85115

5.7

1.0

8.45

0.34

-+ 0.08

10.0

-+ 2.0

35/65

3.5

6.5

9.20

0.32

-+ 0.03

16.0

+ 2.0

15185

2.4

13.6

7.45

The

6.7

Km for

NMNA (a)

was found

994

of

Vol. 91, No. 3, 1979

BIOCHEMICAL

AND

BIOPHYSICAL

RESEARCH COMMUNICATIONS

Fig. 3. Lineweaver-Burk plot of the oxidation of 2a hydrodiperchlorate (4) to 2 catalyzed by aldehyde oxidase at pH 7.4 (22 'C). The reaction rate was measured by following (at 600 nm) the aldehyde oxidase catalyzed reduction of 2,6-dichloroindophenol (6). A Km of 2 PM was calculated from the intercept on the abscissa.

pH (Table

l)(4).

Values

thus

obtained

can be used

of Ki for

1). Since apparent -2a and -2b (Table undergo a 50 % and a two-fold increase,

only

studied,

it

undergo inhibitory

form,

then

its

low Km value

actually

the

lowest 15). buffer

If set

it

is assumed

Cotinine weakly

the

substrate

in the

oxidase

that

the

for

for

same is

NMNA

pH interval

NMNA does true

for

for

calculated for -2a (Table 2b. Thus, be the -2a should for the enzyme.

not

the

1) must be inhibitory

feature

of -2a in its interaction with aldehyde oxidase of 2 uM (pH 7.4), obtained by measuring the dichloroindoof

the

reduced

mM has previously using

(A)

max

of Ki values the set

activity

0.22

two new sets

values

respectively,

of aldehyde

form

of

the

K recorded for any substrate for m previously reported are about one order

, pH 7.8,

and only

affinity

lowest

values A K, of

the

than

also

striking

reductase

the

likely

and therefore,

phenol

that change.

as more

The most is

likely

any drastic

regarded form

is

to calculate

K, and v

does

inhibits

been obtained

the dichloroindophenol not

inhibit the

oxidation

the

for

method oxidation

enzyme aldehyde

This

oxidase;

is

the

of magnitude

higher

M phosphate

(9,

(16).

of -2a hydrodiperchlorate (d) 1.0 mM NMNA and 2.1 mM 3

of NMXA. At

the

shows an optical

and 9.5.

approximately

3).

NMNA in 0.05

a 20 % inhibition was observed. Substrate inhibition 0.3 mM -2a hydrodiperchlorate (A). The pH profile for region

(Fig.

between

pH 6.8

seems to begin oxidation

at about of -4 to -3

BIOCHEMICAL

Vol. 91, No. 3, 1979

Compound has been

i

which

of an aromatic expected

that

of several and that

(2a - hydrodiperchlorate) with respect to its

studied

al compounds

AND BIOPHYSICAL

contain

system iminium

saturated

seems to be the interaction

an oxidizable

have, ions

however, are

with

first

studied

principal

iminium

aldehyde

carbon-nitrogen been

the

RESEARCH COMMlJNICATlONS

before

oxidase.

double

bond

(9,15).

It

intermediates

in

that Sever-

as part is

to be

the metabolism

xenobiotics to the corresponding -N-heterocyclic oxidase plays an important role in the oxidation

aldehyde

ion

lactams of

these

intermediates. As already hyde

oxidase"

pointed is

good alternatives

out

inadequate have

by other for

been

for

Acknowledgements. cation

of

the

and Mr Johan supported

enzyme,

Dr Bengt

Montelius

by the

compounds

Swedish

for

Dr Anders Mannervik

advice Tobacco

but

of the

containing should

oxidase"

We thank

(9,17-19)

reasons, In view

lacking.

aromatic -2a and for several bond moiety, the name "iminium

investigators

various

for

in the

high

aid

affinity

name "aldebecause of the

a carbon-nitrogen

be more

Mzhl&r

the

has been kept

for

descriptive.

carrying

in using

spectrophotometric

enzyme double

out the

the purifi-

computer

work.

This

program, work

was

Company.

REFERENCES 1. 2. 3. 4. 5. 6. 7. 8. 9. 10. 11. 12. 13. 14. 15. 16. 17. 18. 19.

Hucker, H.B., Gillette, J.R., and Brodie, B.B. (1960) J. Pharmacol. Exp. Ther. 129, 94-100. Gorrod, J.W., and Jenner, P. (1975) Essays Toxicol. 6, 35-78. Murphy, P.J. (1973) J. Biol. Chem. 248, 2796-2800. L. (1979) Acta Chem. Stand. B33, 187-191. Brandlnge, S., and Lindblom, Hill, D.L., Laster, W.R., Jr, and Struck, R.F. (1972) Cancer Res. 32, 658-665. Rajagopalan, K.V., and Handler, P. (1964) J. Biol. Chem. 239, 2022-2026. Kalckar, H.M. (1947) J. Biol. Chem. 167, 429-443. Felsted, R.L., Chu, A.E.-Y., and Chaykin, S. (1973) J. Biol. Chem. 248, 2580-2587. Krenitsky, T.A., Neil, S.M., Elion, G.B., and Hitchings, G.H. (1972) Arch. Biochem. Biophys. 150, 585-599. Kalckar, H.M. (1947) J. Biol. Chem. 167, 461-475. Rajagopalan, K.V., and Handler, P. (1966) Methods Enzymol. 9, 364-368. Rajagopalan, K.V., Fridovich, I., and Handler, P. (1962) J. Biol. Chem. 237, 922-928. Bray, R.C. (1975) The Enzymes 12 B, 299-419. Hucker, H.B., and Gillette, J.R. (1960) Fed. Proc., Fed. Amer. Sot. Exp. Biol. 19, 30. Stubley, C., Subryan, L., Stell, J.G.P., Perrett, R.H., Ingle, P.B.H., D.W. (1977) J. Pharm. Pharmacol. 29, 77 P. and Mathieson, Branzoli, U., and Massey, V. (1974) J. Biol. Chem. 249, 4339-4345. Ref. 13, p. 352. H. (1978) Wiss. Z., Karl-Marx-Univ., Leipzig, Math.Kurth, J., and Aurich, -Naturwiss. Keihe 27, 47-54. @hkubo, M., and Fujimura, S. (1978) Cancer Res. 38, 697-702.

996