Nghiên cứu mô phỏng giảm methemoglobin trong hồng cầu: Đóng góp khác biệt của hai con đường vào khả năng chịu đựng stress oxy hóa

Simulation study of methemoglobin reduction

in erythrocytes

Differential contributions of two pathways to tolerance to

oxidative stress

Ayako Kinoshita

, Yoichi Nakayama

1,2

, Tomoya Kitayama

and Masaru Tomita

1 Institute for Advanced Biosciences, Keio University, Kanagawa, Japan

2 Network Biology Research Centre, Articell Systems Corporation, Keio Fujisawa Innovation Village, Fujisawa, Japan

In an integrative approach to gaining a better under-

standing of cell-wide molecular networks at the sys-

tems level (systems biology), many attempts at

computer simulations of metabolic networks have been

made. Because of the simplicity of its structure and

components, and the availability of kinetic informa-

tion, erythrocyte metabolism has been the focus of

mathematical modeling for over three decades [1], and

is a leading example of the use of mathematical mode-

ling not only for understanding biochemical regulation,

but also as the basis for the construction of other

mathematical frameworks [2–5]. For example, models

of erythrocyte metabolism were able to predict the

importance of the de novo synthesis of glutathione and

its mechanism of action in glucose-6-phosphate dehy-

drogenase (G6PDH)-deﬁcient cells [6], and to analyze

Keywords

erythrocyte; mathematical modeling;

metabolism; methemoglobin; oxidative

stress

Correspondence

Y. Nakayama, Institute for Advanced

Biosciences, Keio University, Fujisawa,

Kanagawa, 252–8520, Japan

Fax/Tel: +81 466 57 5099

E-mail: ynakayam@sfc.keio.ac.jp

Note

The mathematical model described here has

been submitted to JWS Online database

and can be accessed free of charge at

http://www.jjj.bio.vu.nl/database

(Received 25 August 2006, revised 30

November 2006, accepted 10 January 2007)

doi:10.1111/j.1742-4658.2007.05685.x

Methemoglobin (metHb), an oxidized form of hemoglobin, is unable to

bind and carry oxygen. Erythrocytes are continuously subjected to oxida-

tive stress and nitrite exposure, which results in the spontaneous formation

of metHb. To avoid the accumulation of metHb, reductive pathways medi-

ated by cytochrome b5 or ﬂavin, coupled with NADH-dependent or

NADPH-dependent metHb reductases, respectively, keep the level of

metHb in erythrocytes at less than 1% of the total hemoglobin under nor-

mal conditions. In this work, a mathematical model has been developed to

quantitatively assess the relative contributions of the two major metHb-

reducing pathways, taking into consideration the supply of NADH and

NADPH from central energy metabolism. The results of the simulation

experiments suggest that these pathways have different roles in the reduc-

tion of metHb; one has a high response rate to hemoglobin oxidation with

a limited reducing ﬂux, and the other has a low response rate with a high

capacity ﬂux. On the basis of the results of our model, under normal oxi-

dative conditions, the NADPH-dependent system, the physiological role of

which to date has been unclear, is predicted to be responsible for most of

the reduction of metHb. In contrast, the cytochrome b5–NADH pathway

becomes dominant under conditions of excess metHb accumulation, only

after the capacity of the ﬂavin–NADPH pathway has reached its limit. We

discuss the potential implications of a system designed with two metHb-

reducing pathways in human erythrocytes.

Abbreviations

cytb5, cytochrome b5; cytb5R, (NADH-dependent) cytochrome b5 reductase; FR, (NADPH-dependent) flavin reductase; GAPDH,

glyceraldehyde-3-phosphate dehydrogenase; G6PDH, glucose-6-phosphate dehydrogenase; GSSG, oxidized glutathione; GSH, glutathione

(reduced form); LDH, lactate dehydrogenase; metHb, methemoglobin.

FEBS Journal 274 (2007) 1449–1458 ª2007 The Authors Journal compilation ª2007 FEBS 1449

the effects of deﬁned single nucleotide polymorphisms

in G6PDH and pyruvate kinase on cellular function,

through which it was possible to demonstrate the dif-

ferences between patients with the two enzyme deﬁci-

encies with chronic and nonchronic anemia [7]. A

mathematical model focusing on the methemoglobin

(metHb)-reducing pathways in erythrocytes, which

might be essential for the function of normal erythro-

cytes, has not yet been developed.

In circulating erythrocytes, hemoglobin oxidation to

metHb occurs continuously, not only via intracellular

and extracellular reactive oxygen species, but also via

exogenous and endogenous nitrites and nitric oxide [8].

Methemoglobinemia occurs when the level of metHb is

greater than 1% of the total hemoglobin content of

the cell [9,10], although no clinical symptoms are

observed until the level of metHb reaches 10% of the

total. Cyanosis occurs at levels of 10% or greater. Lev-

els of metHb above 20% are associated with headache,

dizziness, fatigue, and tachycardia. Coma and death

may occur at levels of 60–70% metHb. The accumula-

tion of metHb in erythrocytes reduces their ability to

supply oxygen, and, moreover, cycling of hemoglobin

and metHb causes an associated persistent production

of superoxide anions, potentially resulting in additional

oxidative stress [11,12].

In normal erythrocytes, metHb is maintained at a

level of less than 1% of total hemoglobin by two

major enzymatic systems for reducing metHb as sche-

matically shown in Fig. 1 [13,14]. One of these path-

ways is a redox cycle consisting of cytochrome b5

(cytb5) and cytochrome b5 reductase (cytb5R), which

transfers electrons from NADH to cytb5. It is estima-

ted that this pathway is responsible for more than

95% of erythrocyte-reducing capacity under experi-

mental conditions [15–17]. Hereditary methemoglo-

binemia is a congenital deﬁciency of cytb5R and ⁄or

cytb5 [18]. The second pathway uses ﬂavin as an elec-

tron carrier for the reduction of metHb, coupled with

NADPH oxidation, catalyzed by NADPH-dependent

ﬂavin reductase (FR). The ﬂavin–NADPH pathway

has generally been considered to be a minor contribu-

tor to the reduction of metHb, relative to the cytb5–

NADH system, as malfunction of the ﬂavin–NADPH

system in erythrocytes is not associated with a metHb-

reduction-deﬁcient phenotype [19]. On the basis of

these lines of evidence, the role of the ﬂavin–NADPH

pathway remains unclear. NADPH-dependent FR is

Fig. 1. Schematic representation of the pathways included in the model: metHb-reduction pathways and the relevant central carbon path-

ways in human erythrocytes. The central carbon pathways were modeled using an existing model developed by Mulquiney & Kuchel

[22,23], in which the 108 differential equations were constructed from precise kinetic equations. R

b5R

is the reaction catalyzed by NADH-

dependent cytb5R. R

b5-metHb

and R

FMN-metHb

represent the nonenzymatic metHb-reducing reactions mediated by reduced cytb5 and reduced

flavin mononucleotide, respectively. Oxidized flavin mononucleotide is reduced by the NADPH-dependent flavin reductase, described by R

GLC, glucose; G6P, glucose 6-phosphate; F6P, fructose 6-phosphate; FDP, fructose 1,6-diphosphate; DHAP, dihydroxyacetone phosphate;

GA3P, glyceraldehyde 3-phosphate; GL6P, gluconolactone 6-phosphate; GO6P, gluconate 6-phosphate; RU5P, ribulose 5-phosphate; R5P,

ribose 5-phosphate; X5P, xylulose 5-phosphate; S7P, sedoheptulose 7-phosphate; E4P, erythrose 4-phosphate; PYR, pyruvate; LAC, lactate.

Simulation of methemoglobin reduction A. Kinoshita et al.

1450 FEBS Journal 274 (2007) 1449–1458 ª2007 The Authors Journal compilation ª2007 FEBS

reported to be widely distributed in human tissues, but

is most abundant in erythrocytes [20,21]. This appears

to be an inefﬁcient distribution pattern, given the

potentially minor role of FR in metHb reduction, and

raises the question of why erythrocytes would primar-

ily use an NADH-dependent reduction process for the

reduction of metHb, as it runs contrary to the accep-

ted belief that NAD

⁄NADH is used for oxidative

processes, whereas NADP

⁄NADPH is used for

reductive processes.

In this paper, a mathematical model of the two

metHb-reduction pathways in erythrocytes has been

developed, which incorporates the previously published

model of major carbon metabolic pathways by Mul-

quiney & Kuchel [22,23]. Using this model to make

predictions about the two pathways, in terms of their

metHb-reducing behavior, we demonstrate a switching

of the dominant source of metHb reduction from the

NADPH–ﬂavin pathway to the NADH–cytb5 path-

way, depending on the level of hemoglobin oxidation.

These results suggest that the ﬂavin–NADPH path-

way is the one mainly responsible for metHb reduction

under physiological conditions of metHb generation,

and that this system can keep the steady-state concen-

tration of metHb at a level that is much lower than

that attainable by the cytb5–NADH pathway alone.

The mathematical model described here has been

submitted to JWS Online database and can be accessed

free of charge at http://www.jjj.bio.vu.nl/database.

Results

Simulation experiment of biological conditions

To assess the quantitative validity of our model of

metHb reduction, we compared the reducing ability of

the two pathways in simulation experiments and in

biochemical experiments of the in vitro reconstructed

pathway [15]. The relationship between the reduction

rate of the ﬂavin–NADPH pathway and ﬂavin concen-

tration is shown in Fig. 2. The experimental conditions

were taken from the literature [15] and used in the

simulation experiments as described in the ﬁgure

legend. The results of the simulation experiment were

quantitatively similar to the experimentally determined

reduction rate of the ﬂavin–NADPH pathway in the

low ﬂavin concentration range of less than 2.5 lm.

Because the highest ﬂavin concentration (2 lm)is

much lower than the K

of FR for the oxidized form

of ﬂavin (52.76 lm, see the section on the mathemat-

ical model) and because the rate constant for the non-

enzymatic metHb-reducing process driven by the

reduced form of ﬂavin is sufﬁciently high, most of the

ﬂavin was kept in the oxidized form. On the other

hand, the NADPH concentration (50 lm) was sufﬁ-

ciently high to ensure complete saturation of FR.

Therefore the reduction rate of the ﬂavin–NADPH

pathway was highly relevant only with the K

of FR

for ﬂavin, maximum velocity of FR, and total ﬂavin

concentration (supplementary Fig. S1A).

The relative contributions to metHb reduction made

by the ﬂavin–NADPH and cytb5–NADH pathways at

a ﬁxed concentration of metHb (63 lm), as determined

in the simulation experiments, were compared with the

results of the in vitro reconstitution assays [15]. For the

cytb5–NADH pathway, the reaction mixture contained

63 lmmetHb, 300 lmNADH, 1.04 lmcytb5 and

0.141 lmpuriﬁed cytb5R (9.3 lg in the 2-mL reaction

mixture if the molecular mass of the enzyme is assumed

to be 33 kDa). For the ﬂavin–NADPH pathway, the

experimental conditions used were the same as those

described in the legend of Fig. 2 where the ﬂavin con-

centration was 1 lm. The ﬂavin–NADPH pathway

handled 9.2% of the metHb-reducing activity through

the cytb5–NADH pathway in the in vitro experiment,

whereas the model predicted 4.3% (data not shown). A

possible reason for the slight overestimation of the

reduction rate through the cytb5–NADH pathway may

be the difference in temperature between the in vivo

conditions (or the setting of the simulation) and the

Fig. 2. Comparison of experimental and predicted reduction rates

of metHb through the flavin–NADPH system. Data for the in vitro

experiments were taken from the literature (closed square [15]).

The in vitro experiment was a reconstituted flavin–NADPH pathway

containing various amounts of FMN, 63–75 lMmetHb, 8.73 lM

purified NADPH–flavin reductase (0.384 mg in the 2-mL reaction

mixture where the molecular mass of the enzyme was assumed to

be 22 kDa), 50 lMNADPH, 2.2 lMG6PDH, 3.85 lMglucose

6-phosphate in the presence of oxygen at 24 C and pH 7.0.

Parameters similar to those used in the in vivo experiments were

used in the simulation experiments (solid line).

A. Kinoshita et al.Simulation of methemoglobin reduction

FEBS Journal 274 (2007) 1449–1458 ª2007 The Authors Journal compilation ª2007 FEBS 1451

experimental conditions (24 C). Under these condi-

tions, the rate constant of the nonenzymatic metHb-

reducing process realized by the reduced form of cytb5

is the most relevant parameter determining the rate of

reduction of metHb (supplementary Fig. S1B). A differ-

ence in temperature between the experimental condition

(24 C) and physiological condition (37 C) may also

affect the parameter of the nonenzymatic process, and

consequently result in lower overall rate than the

simulation result.

Relative contribution of each pathway to metHb

reduction as a function of the rate of metHb

formation

The steady-state accumulation of metHb in response to

increased hemoglobin oxidation is shown in Fig. 3 for

the two-pathway model, which includes both the ﬂavin–

NADPH and cytb5–NADH pathways, as well as each

pathway alone. Hemoglobin oxidation was represented

as a ﬁrst-order rate process with the rate constant k

The accumulation of metHb in the ﬂavin–NADPH-

only pathway model was lower than that in the cytb5–

NADH-only pathway model when the pseudo-ﬁrst-

order rate constant of hemoglobin (k

) was below

3.5 ·10

. However, when this rate was exceeded,

an abrupt increase in metHb occurred in the ﬂavin–

NADPH-only pathway model. The accumulation of

metHb in the two-pathway model was similar to the

ﬂavin–NADPH-only model under the condition of

slow hemoglobin oxidation; however, under conditions

of fast hemoglobin oxidation, it was similar to the

cytb5–NADH-only model. This result suggested that a

switch from the ﬂavin–NADPH pathway to the cytb5–

NADH pathway occurs upon an increase in the oxida-

tion rate of hemoglobin.

Pathway switching was also observed in the analysis

of ﬂux contribution. The overall ﬂux of metHb reduc-

tion and the ﬂux contributions of the ﬂavin–NADPH

and cytb5–NADH pathways are shown in Fig. 4. When

was below 3.5 ·10

, most of the metHb-reduc-

tion ﬂux was due to the ﬂavin–NADPH pathway, and

the concentration of metHb was much lower than 1 lm.

In in vitro reconstitution experiments, it was also

observed that the ﬂavin–NADPH pathway had already

reached its rate limit with respect to metHb concen-

tration before metHb reached 1 lm, at near-physio-

logical concentrations of enzyme and ﬂavin [17].

When k

was over 3.5 ·10

, an abrupt switch

in ﬂux contribution from the ﬂavin–NADPH pathway

to the cytb5–NADH pathway occurred, similar to that

seen in Fig. 3. The relative ﬂux contribution of the

ﬂavin–NADPH pathway under conditions of high

metHb concentration ultimately reached 2% of the

total ﬂux. These results suggest that the two reduction

pathways are active under different conditions of

Fig. 3. Steady-state concentration of metHb with various k

val-

ues. Steady-state concentrations of metHb were plotted as a func-

tion of oxidative load for the reference model which included both

pathways (gray solid line), the model with the cytb5–NADH path-

way only (broken line), and the model with the flavin–NADPH path-

way only (black solid line).

Fig. 4. Steady-state flux of metHb reduction and the percentage

contribution of each pathway. The black broken and the black solid

lines represent percentage contributions to total flux of metHb

reduction for the cytb5–NADH pathway and the flavin–NADPH

pathway, respectively. The gray solid line represents the total

metHb reduction rate with respect to changes in oxidative load.

Line A indicates the hemoglobin oxidation rate where the total

metHb-reducing flux equals the reported rate of spontaneous

hemoglobin oxidation (2.78 ·10

MÆs

). Lines B and C represent

the hemoglobin-oxidation rates where the concentration of metHb

reaches 1% and 10% of total hemoglobin, respectively.

Simulation of methemoglobin reduction A. Kinoshita et al.

1452 FEBS Journal 274 (2007) 1449–1458 ª2007 The Authors Journal compilation ª2007 FEBS

hemoglobin oxidation, and make different contribu-

tions to the tolerance of oxidative stress. The ﬂavin–

NADPH pathway may function mainly to provide

reduction potential under normal conditions, whereas

the cytb5–NADH pathway functions as the main

driving force for reduction of metHb under conditions

of excess oxidation, such as during the intake of oxi-

dant drugs.

Furthermore, according to our model, under condi-

tions where the cytb5–NADH pathway was responsible

for more than 95% of metHb reduction, the ﬂux

through the cytb5–NADH pathway could potentially

reach almost 1.2 ·10

mÆs

. This rate is much higher

than the rate reported for nonglycolytic NADH con-

sumption in human erythrocytes, which is approxi-

mately 2.78 ·10

mÆs

[24]. The source of this vast

amount of NADH was an interesting reaction in our

simulation experiments. NADH is supplied primarily

through glycolysis under normal conditions in erythro-

cytes; however, in our simulation, the production of

NADH was also coupled to the lactate ⁄pyruvate shut-

tle under conditions of excess oxidation, and the

increased demand for NADH was met by the reverse

reaction of lactate dehydrogenase (LDH), rather than

by the increased glycolytic ﬂux (supplementary

Fig. S2A). It is possible to hypothesize that an

increased hemoglobin-oxidation rate leads to a switch

in the major glycolytic product from lactate to pyru-

vate, which could then be released into the plasma. In

fact, the accumulation of pyruvate in plasma resulting

from the pyruvate–lactate shuttle in erythrocytes under

the condition of excess oxidation and increasing

metHb is observed experimentally [25,26].

Discussion

In this paper, we developed a mathematical model of

the NADPH-dependent and NADH-dependent metHb

reduction pathways in erythrocytes, combining them

with Mulquiney and Kuchel’s precise model of central

carbon metabolism pathways in human erythrocytes

[22,23]. Simulation experiments predicted that, under

low oxidative stress, the reduction of metHb occurred

primarily through the NADPH-dependent pathway,

and that switching of pathways from the ﬂavin–

NADPH pathway to the cytb5–NADH pathway

occurred upon reaching a maximal permissible rate of

NADPH-dependent ﬂavin reductase.

As mentioned above, to date, the role of the ﬂavin–

NADPH pathway is unclear. However, our simulation

results indicate that the ﬂavin–NADPH pathway may

have a major role, or at least a non-negligible role, in

reducing metHb under physiological conditions. One

reason why the role of this pathway in the reduction of

metHb has not been uncovered may be the difﬁculties

in measuring the trace levels of oxidation occurring

normally in vivo, whereas it is much easier to observe

conditions of excess oxidative stress, in which the

cytb5–NADH pathway may play a major role. The oxi-

dative rate under physiological conditions is estimated

to be much lower than that of experimental and abnor-

mal conditions. The rate of hemoglobin oxidation

occurring constitutively in normal erythrocytes has

been reported to be 3% of the total amount of hemo-

globin in the cell per day [27], which converts into

a spontaneous rate of hemoglobin oxidation of

2.78 ·10

mÆs

. This rate is signiﬁcantly lower than

the value of 2.31 ·10

mÆs

for the ﬂux capacity of

the ﬂavin–NADPH pathway. In addition, the rate

constants for auto-oxidation of oxyhemoglobin, the

oxidation rate of which is much higher than that of

deoxyhemoglobin, is quite low, 2.75 ·10

[28,29]. Under this low level of oxidative stress, the ﬂa-

vin–NADPH pathway would be responsible for most

of the metHb reduction, because the rate constant of

the nonenzymatic process which directly reduces

metHb is 1000-fold higher than in the cytb5–NADH

pathway. However, as the ﬂavin–NADPH pathway has

low reductase activity, under excess oxidative stress, the

reduction of ﬂavin would be insufﬁcient to meet the

demand for the reduction of metHb. In fact, the pri-

mary use of the source of reducing equivalents of

NADPH in erythrocytes is to convert the oxidized form

of glutathione (GSSG) into its reduced form (GSH) in

a reaction catalyzed by glutathione reductase [24]. GSH

is necessary for avoiding the irreversible oxidation of

intracellular proteins, including membrane proteins and

enzymes, while the accumulation of GSSG causes pro-

tein dysfunction, by creating disulﬁde bonds between

the –SH groups of cysteine and methionine residues

[30]. For this reason, high activities of G6PDH and

glutathione reductase in human erythrocytes are evolu-

tionarily maintained because they are necessary to

avoid strong NADPH depletion (not GSH depletion)

and GSSG accumulation under oxidative stress,

respectively [31,32]. One explanation for the use of

NADH as a source of reducing equivalents for metHb

may be to avoid competition for NADPH by the two

reducing pathways for glutathione and hemoglobin.

In contrast, the cytb5–NADH pathway has a high

pseudo-ﬁrst-order rate constant for reducing metHb

with cytb5, which was predicted to be 4000-fold higher

than that of the ﬂavin–NADPH pathway, but has a

low rate constant for the nonenzymatic process, and a

low concentration of cytb5. These properties could

make this pathway efﬁcient at high levels of oxidative

A. Kinoshita et al.Simulation of methemoglobin reduction

FEBS Journal 274 (2007) 1449–1458 ª2007 The Authors Journal compilation ª2007 FEBS 1453

Báo cáo khoa học: Simulation study of methemoglobin reduction in erythrocytes Differential contributions of two pathways to tolerance to oxidative stress

Chủ đề:

Tài liệu liên quan

Tài liêu mới

AI tóm tắt

Giới thiệu tài liệu

Đối tượng sử dụng

Từ khoá chính

Nội dung tóm tắt

Hỗ trợ

Phương thức thanh toán

Theo dõi chúng tôi