NADPH-flavin reductase in human erythrocytes and the reduction of methemoglobin through flavin by the enzyme

Vol. 76, No. 1, 1977

BIOCHEMICAL

NADPH-FLAVIN

AND BIOPHYSICAL RESEARCH COMMUNICATIONS

REDUCTASE IN HUMAN ERYTHROCYTES AND THE REDUCTION OF

METHEMOGLOBIN THROUGH FLAVIN BY THE ENZYME

Toshitsugu

Yubisui,

Takasumi

Matsuki,

Kiyoo

and Yoshimasa Department of Biochemistry, Medical Technology, School

Received

April

Tanishima",

Masazumi

Takeshita",

Yoneyama

School of Medicine, and *Department of Paramedicine, Kanazawa University, Kanazawa, Japan

of

1,1977

SUMMARY A NADPH-dehydrogenase of human erythrocytes was exhaustively purified to a homogeneous protein judging from the electrophoresis on a polyacrylamide gel in the presence of sodium aoaecyl sulfate. Studies on the specificity for the electron acceptor of this enzyme suggest that flavins serve as the natural and direct electron acceptor. The enzyme showed a broad specificty for-slavins and the Michaelis constants for flavins were estimated to be 5 X 10 M for both FMN and riboflavin. Rapid reduction of methemoglobin by the enzyme in the presence of flavin was demonstrated, and the reduction was explained by the reduction of flavin by the enzyme, and subsequent non-enzymatic reduction of methemoglobin by the reduced flavin. The therapeutic significance of flavins was discussed with reference to the flavin reductase activity in hereditary methemoglobinemia. INTRODUCTION During

the

dehydrogenase purified

of human erythrocytes

can be replaced

by methylene

Further

studies

activity

other

ductase

activity

via

purification,

however,

activity.

The lability

be

reason

a major

enzyme activity enzyme which

Copyright All rights

the

0 1977

apparently

the diaphorase

methylene

blue.

resulted of the

In this

in the presence

of flavin

communication it

the

to observe we report

as the

(1).

by the

has flavin

from the

activity

failed

the

re-

or the m&hemoglobin

of flavin

reductase have

methemoglobin

enzyme we purified activity

in erythrocytes

by Academic Press, Inc. in any form reserved.

of reproduction

flavin

a NADPHand that

to reduce

Removal

that

by flavin

in an enzyme without

characterized

of flavins

blue the

we found

stabilized

rapidly

that

why many workers (2,3).

role

revealed than

reductases

is well

methemoglobin

ductase

discuss

on methemoglobin

enzyme can reduce

The flavin enzyme.

studies

flavin

in relation

re-

buffer

during

flavin

redcuctase

of the

enzyme may

the some

physiological properties

reductase.

of the

We also

to the reduction

of

174 ISSN

0006-291

X

Vol. 76, No. 1, 1977

methemoglobin

MATERIALS

in hereditary

BIOCHEMICAL

AND BIOPHYSICAL

RESEARCH COMMUNICATIONS

methemoglobinemia.

AND METHODS

Normal human red cells were obtained from the central laboratory of the Tokyo Red Cross Blood Transfusion Service. B-NADH, cytochrome c, FAD and FMN were obtained from Boehringer Manheim, NADPH from Sigma Chemicals Co., and other reagents from commercial sources. Superoxide dismutase purified from cow blood according to the method of McCord and Fridovich (4) was kindly supcow adrenodoxin, plied by Dr. A. Tomoda in our laboratory. Frog biopterin, and spinach ferredoxin were also generously presented by Dr. M. Akino of Tokyo Metropolitan University, by Dr. E. Itagaki and Dr. K. Suhara of Kanazawa University, and by Dr. M. Shin of Kobe Yamate Women's College, respectively. Spectrophotometric determinations were performed with a Hitachi spectrophotometer model 124 fitted with a recorder model 056. NAD(P)H-diaphorase activity was determined as described by Sugita et al. (5) in 2.0 ml of 50 s&l NAD(P)H-flavin reductase activity during the puriphosphate buffer (pH 7.5). fication was determined by following the reduction of methemoglobin in the mixture contained 100 umoles presence of flavin at 576 nm. Two ml of reaction phosphate buffer (pH 7.5), 200 nmoles NAD(P)H, 200 nmoles flavin, 1 umole EDTA, 100 nmoles methemoglobin and an appropriate amount of enzyme. The reaction was started by the addition of enzyme and performed at 23-25O. The enzyme activity was calculated using a millimolar extinction coefficient of 11.9 at 576 nm. Purification of enzyme --- NADPH-dehydrogenase was purified from human erythrocytes by the modified method of previous workers (5-8). Red cells were centrifuged at 1,500 X g for 5 min to remove plasma proteins and buffy coats, with isotonic saline at least three times. The washed cells and then washed were hemolyzed by adding 3 volumes of 20 uM FMN, and stroma was removed by high speed centrifugation. To the supernatant an additional 1.5 volumes of 20 uM FMN was added, and the hemolysates thus prepared were applied on a DEAE-cellu lose column which had been equilibrated with 2 mM phosphate buffer (pH 7.0) reductase fraction eluted from containing 20 uM FMN. NADPH-methemoglobin the column by 100 mM phosphate buffer (pH 7.0) containing 20 uM FMN were combined, concentrated on a PM 10 Diaflo membrane, and applied on a Ultrogel AcA 54 column which had been equlibrated with 50 mM phosphate buffer (pH 7.5) containing 20 uM FMN and 1 mM EDTA. The NADPH-methemoglobin reductase fractions eluted from the column were pooled, concentrated as described above, and dialyzed against 10 mM phosphate buffer (pH 7.5) containing 20 uM FMN and 1 mM EDTA. The dialyzed enzyme was applied on a DE-32 column which had been equilibrated with the same buffer used for the dialysis. The NADPH-methemoglobin reductase was eluted from the column during the wash with the equilibrating buffer, while NADH-cytochrome b reductase was eluted by the linear gradient of phosphate. The NADPH-enzymg5was fractionated with ammonium sulfate, and precipitates appearing between 40 to 70% saturation were collected. Precipitates were dissolved in a minimum volume of buffer, and again applied on the Ultrogel column as described above. The NADPH-enzyme was concentrated on a PM 10 Diaflo membrane and dialyzed against 10 mM phosphate buffer (pH 6.8) enzyme was applied on a CMcontaining 20 uM FMN and 1 mM EDTA. The dialyzed 32 column which had been equilibrated with the same buffer used for the dialysis. By washing the column with the equlibrating buffer almost all the colored protein was eluted and the NADPH-enzyme was eluted by the linear gradient of phosphate from 10 to 100 mM (pH 6.8). The enzyme was concentrated on a membrane and again treated with Ultrogel, and then further purified by electrophresis with carrier ampholite of pH range from 7-9. The electrophoresis was performed at 4O-6" for 38-40 hrs. The enzyme was concentrated and the carrier ampholite was removed by gel filtration. Protein was determined by the method of Lowry et al. (9).

175

Vol. 76, No. 1, 1977

BIOCHEMICAL

AND BIOPHYSICAL RESEARCH COMMUNICATIONS

Table

I

Summary of purification Total activity

Fractions 1 2

Hemolysates DEAE-cellulose 3 Ultrogel AcA 54 4 DE-32

(a)

380

1.16 z:.;

5 (NH412s04 (40-m%) 6 Ultrogel AcA 54

63:o

7

16.2

Protein (g)

Specifictb) activity

4070 46.3 10.7 8.2 4.3

0.093

58.1

CM-32 8 Ultrogel AcA 54 9 Electrofocusing

6.03

(I) 100

2.51 6.88 8.0 14.7 24.6 296 333 281.7

2.63 0.055 0.035 0.021

11.8

Yield

30.5 19.4 17.2 16.6 15.3 4.3 3.1 1.6

a) Total activity; pmoles hemoglobin reduced / min. b) Specific activity; nmoles hemoglobin reduced / min / mg. of methemoglobin was followed at 576 nm in the assay mixture as described in "METHODS".

The reduction 100 pi4 FMN in the

presence

of

RESULTS Table

I shows the

human erythrocytes.

of

NADH-flavin diaphorase ed

reductase activity

sulfate

istic

1% of those

of flavin,

final

to unity.

respect

Purity gel

The ratio

of the in the

final

to the

presence

presence

of NADPH to while

of sodium

on the

gel

showing

absorption

maxima

the

enzyme was easily

removed

enzyme.

in the

The resulting absorbance

was 50%, and after

176

of the gel

doas

at 443, by gel

enzyme had an absorption

at 443 and 370 nm decreased

The yield

of

of FMN had a character-

of a flavoprotein flavin

that

enzyme was exsmin-

band was observed in the

from

and had a specific

enzyme was 4.5-5.0,

one distinct

at 276 nm, and the of untreated

with

reduced/min/mg.

The enzyme purified

spectrum

electrophoresis

presence

by the

or by electrophoresis.

maximum only

after

and only

and 276 nm; however,

filtration

3,500-fold

on a polyacrylamide

1.

absorption

methemoglobin

was close

(lo),

shown in Figure

in the

activity

by electrophoresis

decyl

370,

reductase nmoles

330

of a NADPH-dehydrogenase

The enzyme was purified

NADPH-methemoglobin activity

summary of purification

filtration

flavin

reductase more than

to about activity

lOO%, respec-

Vol. 76, No. 1, 1977

BIOCHEMICAL

AND BIOPHYSICAL

RESEARCH COMMUNICATIONS

Figure 1. Polyacrylamide gel electrophoresis of the flavin reductase in the presence of sodium dodecyl sulfate. Fifteen and 30 !Jg of the purified enzyme was applied on the left and right gels, respectively. The electrophoresis and staining of protein were performed according to the method of Weber and Osborn (10).

tively.

Figure

and methylene

2 shows the blue

on the

tion

of methemoglobin

only

barely

greatly

had no activity,

reduction

of methemoglobin

reduction

through

two times

however,

activity

half

through

by the

nM than

177

enzyme.

flavins

as well (curves

almost

methylene those

through

b-e).

rates the blue

The reducacrrier

as methylene

was blue

Boiled

was also

The initial

or FMN were while

FMN and FAD),

of an electron

b reductase -5

flavin.

of these, at 0.5

(riboflavin,

absence

of methemoglobin

50 uM riboflavin

FAD was about higher

in the

and the NADH-cytochrome

reduction

through

a);

flavins

of methemoglobin

enzyme

(curve the

of three

reduction

by the

detectable

stimulated

effects

enzyme

inactive

on the

of methemoglobin

same, and that showed

flavins.

more than Table

II

Vol. 76, No. 1, 1977

BIOCHEMICAL

AND BIOPHYSICAL RESEARCH COMMUNICATIONS

1

4

Tirni(tnin3)

Figure 2. Reduction of methemoglobin by the flavin reductase through flavin and methylene blue. Reduction of methemoglobin by the flavin reductase was followed spectrophotometrically at 576 nm as described in "METHODS". Curve a) control (no addition of flavin or methylene blue), b) 100 nmoles FMN added, c) 100 nmoles riboflavin added, d) 100 nmoles FAD added, e) 1 nmole methylene blue added.

shows the electron were

specificity donor.

found

to serve

the

hemin,

(DCIP)

be about because activity

to be about

FAD, and 2 UM for

the

As the

reduced

NADPH. direct

flavin

was determined presence

quickly

conveniently

in the

rectly

observed

anaerobically

of the

absorption

spectra

of the

of flavin.

for

both for

flavins

by following

the

of FMN during

the

178

and FMN, 120

of flavin

by air,

in Figure

electron

and NADPH of

riboflavin

reoxidized

as shown

and

as an direct

DCIP was also

reduction

The reduction

hemoproteins

or 2,64ichlorophenolindo-

Km values 50 PM for

flavins

(adrenodoxin

inactive

blue

NADPH as an only

but

proteins

all

The Km value

assay is

globin

iron

were

acceptor.

examined,

acceptors,

nonheme

such as methylene

as a good

enzyme with

substances

electron

b+ and c),

estimated

of this

natural

efficacious

ayes

served

4 uM.

acceptor

GSSG, and menadione

Artificial

enzyme were

UM for

as the

cytochrome

ferredoxin), acceptor.

electron

Among the various

(methemoglobin,

phenol

for

estimated is troublesome

the

flavin

reduction

reductase

of methemo-

of flavin,

however,

3.

3 shows the

reduction

Figure by the

to

was di-

enzyme with

change NADPH

Vol. 76, No. 1, 1977

BIOCHEMICAL

AND BIOPHYSKAL

Table

Specificity

of the for

Electron

RESEARCH COMMUNICATIONS

II

NADPH-dehydrogenase Various

Electron

of Human Erythrocytes Acceptors.

Concentration (WI)

acceptors

(AA 340 nm/min)

x 5

(0 ii

None Methemoglobin Cytochrome h5 Cytochrome c Biopterin Adrenodoxin Ferredoxin Hemin FMN FAD Riboflavin GSSG Menadione

50 50 50 50 50

0 681, 0 037, 0 060, 0 0008, 0.002,

0.079 0.034 0.051 0.0015

Methylene blue 2,6-Dichlorophenolindophenol Ferricyanide

50 50 50

0.132,

0.115

0.716, 0.0096,

0.720 0.003

20 20 20 20 20 20 20

Iii (0) (0) (0)

0.002

O.OOk

The enzyme activity was determined by following the absorbance of NADPH For weak activity at 340 nm with the electron acceptor given in the Table. The the initial rate was followed by using the expanded scale of the recorder. change in valuell within more than absorbance change "(0)" means "no detectable The enzyme (72 ug protein) of another preparation 7-8 min after the reaction. The specific activity from that described in Table I was used for the assay. was 35.5 nmoles/min/mg with FMN as an acceptor, and 194.4 nmoles/min/mg with 2,6-dichlorophenolindophenol as an acceptor.

as an electron

donor.

enzyme added that

to the mixture.

flavin

serves

rather

than

(80%

inhibition)

Superoxide flavin the globin

The rate

the

dismutase

reaction by the

These

as a natural

as a tightly

by the

of reduction

bound

reduction did

not

was dependent

results

and direct

described electron

cofactor.

Atebrin

of methemoglobin affect

the

enzyme.

This

result

suggests

of flavin

with

NADPH is

not

enzyme.

179

that involved

amount

above apparently acceptor

of this

flavin

enzyme

0; which in the

inhibited

by the

of methemoglobin

enzyme.

through

may be derived reduction

of suggest

at 0.5 mM strongly through

reduction

on the

from

of1 methemo-

Vol. 76, No. 1, 1977

BIOCHEMICAL

AND BIOPHYSICAL RESEARCH COMMUNICATIONS

0.1 -

360

400

440

1

460

Wavelength

520

( nm 1

Figure 3. The time change of the absorption spectra of FMN during reduction by the flavin reductase. The reduction of FMN by the enzyme with NADPH as an electron donor was followed under anaerobic conditions. A Thunberger's tube whose main compartment contained 28 !JM FMN in 1.95 ml of 50 mM phosphate buffer (pH 7.5) and 106 ug of the purified enzyme, and the side arm compartment which The tube was evacuated contained 0.05 ml NADPH was sealed with vacuum grease. carefully and the gas phase was exchanged with Q-gas (helium-isobutane (99.05 : 0.95)) at least three times. After mixing the NADPH with the reaction mixture in the cell, the reduction of FMN was followed at an appropriate interval of 1) Before mixing, 2) 5 min, 3) 15 min, 4) 30 min, 5) 60 min time as indicated. after mixing.

DISCUSSION Many controversial methemoglobin

reductase

or whether

free

and those

problems

studies diaphorase However, genase" no flavin activity.

results

flavin

was demonstrated activity the

is the

in the

to have the

activity

activity,

is very

substrate

likely

is

that

our

NADH- and NADPH-

a constituent

reduction

flavin

reductase

of the

was observe4

only

enzyme,

does not

contain

flavin

of the

enzyme,

judging

180

other via

enzymes

which

with

than

the

results

but

blue.

had

(5-8)

have strong

as a cofactor,

the

"NADPH-dehydro

diaphorase

seems to correspond

from the

3)

methylene

of human erythrocytes the

(2,

in the present

activity activity

both

enzyme (sf,

of methemoglobin

recluctase

b+ reductase although

on both

The enzyme purified

settled.

NADH-cytochrome

"NADPH-dehydrogenaseV, flavin

not yet

reductase

reductase It

involved

reported

flavin

or the methemoglobin

flavin

whereas

as to whether is

are

have been

that

to the free

of spectral

and

Vol. 76, No. 1, 1977

kinetic

BIOCHEMICAL

studies.

globin

The flavin

non-enzymatically,

through

flavin

reduced and the

The sequence

msec)(“).

of the

other

luciferase respect molecular 22,000

by disc

bacterial

the

overall

possibility flavins

seem to be low

they

diatary

are

enriched

the major

as one of the however, patients.

erythrocyte

in the for

the

sense

the

whose

methemoglobin

flavin

the

of riboflavin,

a natural

are

will

component

that

there

blue

patients

has been

form than

in erythrocytes

a

by flavins with

here-

b re-5 administered

side-effects

be better

is of

The agent,

NADPH-dehydrogenase.

permeable

in

in erythro-

NADH-cytochrome

produces the

involved

concentrations

To the

methylene

and

(g-12).

to be activated

the

to be about

sulfate,

of flavins

lack

sometimes

to such patients

dodecyl

on the

expected

to activate

enzyme with and to their

enzyme suggests

erythrocytes

to their

flavins,

23,000

by some method.

effective,

is

for

in erythrocytes,

by our

reductase,

agents

membrane, that

of about

enzyme is

flavin to our

of sodium

components

in erythrocytes

The administration

15

reductases

enzyme was estimated

concentrations

greatly

= about

flavin

reduced

Km values

Although

therapeutic

although

properties

depending

(13),

that

similar

of our

the

fact

the

weights the

in the

enzyme may function

methemoglobinemia,

ductase,

the

the

(tl,2

methemo-

---*Methemoglobin.

presence

of methemoglobin

in erythrocytes.

cytes if

that

in the

flavins,

reduction

reduce

NADPH to methemoglobin

provide

weight

have molecular that

lies

and high

electrophoresis

Our finding

from

--+Flavin

have closely

The molecular

enzymes

to be fast

transfer

which

specificity

weights.

enzyme can in turn

was found

interest

(9-K?),

as a substrate,

RESEARCH COMMUNICATIONS

as follows;

great

bacteria

to the broad

rate

reductase

hand,

of some luminous

by the

electron

can be explained

NADPH --+NADPH-flavin On the

AND BIOPHYSICAL

in the of flavin

methylene

through blue

and harmless

patients.

ACKNOWLEDGMENT tion

This work for Study

'a) T. Yubisui,

was supported of Metabolism unpublished

in part by grand-in-aid and Disease. result.

181

from

the Japanese

Associa-

Vol. 76, No. 1, 1977

BIOCHEMICAL

AND BIOPHYSICAL

RESEARCH COMMUNICATIONS

REFERENCES

1)

Yubisui,

T.,

Okamoto,

H.,

Tanishima,

K.,

Takeshita,

4) 5)

CRC Press, Cleveland, Ohio. MacCord, J.M., and Fridovich, A.M. (1976) J. Biol. Sugita, Y., Nomura, S., and Yoneyama, Y. (197m.

M.,

and Yoneysma, Y. Kyoto. (Yoshikawa, Tokyo.

85, (1976) The 16th Ineternational Congress of Hematology, Abst., Cellular and MoleculaFBiology of Erythrocytes 2) Jaffe, $X.(1976) H and Rapaport, S.,ed.)pp.85-216, University of Tokyo Press, 4 Comprehensive Treatise, pp. 3) Kigse , M. (1974) Methemoglobinemia: -., Biol.

-,244 6049-6055. Chem., -,246 6072-

6078.

Passon, P.G., and Hultquist, D.E. (1972) Biochim. Kuma, F., and Inomata, H. (1972) L. Biol. m., 78; Kitao, T., Sugita, Y., Yoneyama, Y., and Hattori,

6)

Biophys. m, 275, 62-73. -,247 556-560. K. (1974) Blood, Q, 879-

884. 9)

Lowly,

O.H.,

x.3

-,193 265-275.

12) 13) 14) 15)

Rosebrough,

N.J.,

Farr,

A.L.,

and Randall,

R.J.

(1951) L. w.

K., and Osborn, M. (1976) J. Biol. Chem., 244, 4406-4412. Puget, K., and Michelson, A.M. (1572) Biochimie, =1197-1204. 57, 461-467. J. (1975) Eur. 2. Biochem., Gerlo, E., and Charlier, 6 Duane, W., and Hasting, J.W. (1975) Mol. Cell. Biochem., -, 53-64. Watanabe, H., Miura, N., Takimoto, A., and Naksmura, T. (1975) J. Biochem., 7-J, 1147-1155. R.W. (1972) --The Red Cell (Harris, J.W., and Harris, J.W., and Kellermeyer, Kellermeyer, R.W. ea.) pp.281-333, Harvard University Press, Cambridge, Massachusetts.

10) Weber, 11)

14-25.

182