Resazurin as an electron acceptor in glucose oxidase-catalyzed oxidation of glucose

Article abstract of DOI:10.1248/cpb.49.622

The behavior of resazurin (1) as an electron acceptor in glucose oxidase (GOD)-catalyzed oxidation of glucose under anaerobic conditions is described. When a mixture of 1, glucose, and GOD in phosphate buffer (pH 7.4, 0.1 M) was incubated at 25°C, the resulting solution turned purple to fluorescent pink due to the deoxygenated product, resorufin (2). On incubation of 1 with GOD alone or with H₂O₂ under essentially the same conditions, no color change was seen, indicating that generation of 2 in the enzymatic reaction is brought about through reduction of 1 by the reduced form (GOD_red) of GOD, which was also supported by the voltammetric behavior of 1. However, it was found that the enzymatic transformation of I to 2 is of no practical use as an indicator reaction for glucose determination using only GOD due to a slow reaction of 1 with GOD_red. Based on a ping-pong type mechanism with a steady-state approximation, K_M and k_cat for 1 as an electron acceptor from GOD_red were estimated to be 15±1.3 μM and (5.0±0.5)×10^-2s^-1, respectively.

Full text of DOI:10.1248/cpb.49.622

622
Notes
Chem. Pharm. Bull. 49(5) 622—625 (2001)
Vol. 49, No. 5
Resazurin as an Electron Acceptor in Glucose Oxidase-Catalyzed
Oxidation of Glucose
Hatsuo MAEDA,* Shinya MATSU-URA, Yuji YAMAUCHI, and Hidenobu OHMORI
Graduate School of Pharmaceutical Sciences, Osaka University, 1–6 Yamada-oka, Suita, Osaka 565–0871, Japan.
Received October 2, 2000; accepted February 14, 2001
The behavior of resazurin (1) as an electron acceptor in glucose oxidase (GOD)-catalyzed oxidation of glu-
cose under anaerobic conditions is described. When a mixture of 1, glucose, and GOD in phosphate buffer (pH
7.4, 0.1 M) was incubated at 25 °C, the resulting solution turned purple to fluorescent pink due to the deoxy-
genated product, resorufin (2). On incubation of 1 with GOD alone or with H2O2 under essentially the same con-
ditions, no color change was seen, indicating that generation of 2 in the enzymatic reaction is brought about
through reduction of 1 by the reduced form (GODred) of GOD, which was also supported by the voltammetric be-
havior of 1. However, it was found that the enzymatic transformation of 1 to 2 is of no practical use as an indica-
tor reaction for glucose determination using only GOD due to a slow reaction of 1 with GODred. Based on a ping-
pong type mechanism with a steady-state approximation, KM and kcat for 1 as an electron acceptor from GODred
were estimated to be 15؎1.3 mM and (5.0؎0.5)؋10؊2 s؊1, respectively.
Key words resazurin; electron acceptor; glucose oxidase; glucose; resorufin; cyclic voltammetry
Resazurin (1) (cf. Chart 1) is known to act as an electron nmol), glucose (5.0 mmol), and GOD (1.0 mg) in 3.0 ml
acceptor in diaphorase- or N-methylphenazinium methosul- phosphate buffer (pH 7.4, 0.1 M) was incubated at 25 °C
fate (PMSϩ)-catalyzed oxidation of NAD(P)H and to be re- under anaerobic conditions, the mixture turned purple to flu-
duced to resorufin (2).1—9) The reductively deoxygenated orescent pink. In the absorption spectrum of the mixture, a
product 2 exhibits strong emission (excitation maximum at new peak with lmax at 571 nm appeared as soon as the incu-
563 nm and emission maximum at 587 nm in pH 7.4 bation was started, as shown in Fig. 1. As incubation time
buffer)10,11) at wavelengthsϾ550 nm, where potential interfer- was increased, the new peak became gradually larger, while
ence in analysis of colored or turbid serum components can the peak due to 1 became smaller and had almost disap-
be avoided. Thus, the transformation of non-fluorescent 1 to peared after 15 min incubation (Fig. 1c). The new peak at
fluorescent 2 has been utilized as a fluorometric indicator re- 571 nm coincided with that of the deoxygenated product 2 in
action for determination of activity1—3) or substrates6—8) of phosphate buffer. The color of 1 was not affected by treat-
NAD(P)ϩ-specific dehydrogenases as well as NAD(P)ϩ 5): by ment with GOD alone or with H2O2 under these conditions.
enzymatic reaction with an NAD(P)ϩ-specific dehydroge- These results suggested that the color change observed in the
nase, its substrate is oxidized, while NAD(P)ϩ is transformed incubation of 1, glucose, and GOD was due to reductive for-
to NAD(P)H; 1 is reduced to 2 by thus formed NAD(P)H in mation of 2 from 1 coupled with reoxidation of the reduced
the presence of diaphorase or PMSϩ, depending on dehydro- form (GODred) of GOD. The following voltammetric behav-
genase activity, or concentration of a substrate or NAD(P)ϩ.
ior of 1 also supported the suggestion that electron transfer to
Based on the intriguing behavior of 1 in NAD(P)H-oxida- 1 from GODred yields 2. On cyclic voltammetry of 1 in phos-
tion, it is speculated that 1 functions as an electron acceptor phate buffer, a cathodic peak at Ϫ203 mV and a couple of a
in enzymatic oxidation by other oxidoreductases such as glu- redox wave (cathodic and anodic peaks at Ϫ320 and
cose oxidase (GOD), being transformed to 2. However, there Ϫ281 mV, respectively) were observed (Fig. 2a). No anodic
have been no studies of the possibility of transformation of 1 peak coupled with the first cathodic peak was observed on a
to 2 as a fluorometric indicator reaction for enzymatic analy- reverse scan even when voltammetry was performed at a
sis coupled with enzymatic redox reactions. Recently, 2 was scan rate as high as 2.0 V/s between 0.2 and Ϫ0.25 V. The
found to function as an electron acceptor with a color change redox wave was coincident with that observed for 2 (Fig. 2b).
in GOD-catalyzed oxidation of glucose under certain condi- Accordingly, it was obvious that the first cathodic peak was
tions.12) This finding prompted us to examine how derivatives due to reduction of 1, where 2 was immediately formed, to
of 2 such as 1 would behave in GOD-catalyzed oxidation of exhibit its redox wave.
glucose, and an interesting behavior of 1 in the enzymatic re-
This is the first report to show that 1 acts as an electron ac-
action was found. Here, we report that in GOD-catalyzed ox- ceptor in GOD-catalyzed oxidation of glucose, being reduced
idation of glucose, 1 is transformed to 2 similarly to the case to 2, similarly to the case of diaphorase- or PMSϩ-catalyzed
of diaphorase- or PMSϩ-catalyzed oxidation of NAD(P)H oxidation of NAD(P)H. Previously it was found that 2 reoxi-
(Chart 1), and is superior to 2 as an electron acceptor, al- dizes GODred effectively at 36 °C, but its ability becomes
though the observed behavior of 1 will not find direct appli-
cation to enzymatic analysis of glucose unless the rate for 1
to reoxidize the reduced form of GOD can be improved.
Results and Discussion
A solution of 1 in phosphate buffer (pH 7.4, 0.1 M) had
a purple color (lmax, 602 nm). When a mixture of 1 (20.0
Chart 1
To whom correspondence should be addressed. e-mail: h-maeda@phs.osaka-u.ac.jp
© 2001 Pharmaceutical Society of Japan

May 2001
623
Fig. 3. Fluorometric Traces Obtained during 10 min Incubation in 3.0 ml
Phosphate Buffer at 25 °C for a Mixture of 1 (10.0 nmol), Glucose, and
GOD (0.5 mg) (a) and for a Mixture of 2 (10.0 nmol), Glucose, and GOD
(0.5 mg) (b)
Fig. 1. Absorption Spectra Obtained Immediately (a), and after 5 min (b),
or 15 min (c) Incubation of a Mixture of 1 (20.0 nmol), Glucose (5.0 mmol),
and GOD (1.0 mg) in 3.0 ml Phosphate Buffer at 25 °C
Excitation and emission wavelengths: 568 and 582 nm, respectively.
Fig. 4. Effects of Concentration of Glucose on Fluorometric Traces Ob-
tained during 5 min Incubation of a Mixture of 1 (20.0 nmol), Glucose, and
GOD (1.0 mg) in 3.0 ml Phosphate Buffer at 25 °C
Excitation and emission wavelengths: 568 and 582 nm, respectively. Concentration of
glucose (mmol): 2.0 (a), 1.0 (b), 0.4 (c), 0.2 (d—f). Traces d—f were obtained on dif-
ferent runs under essentially the same conditions.
cose at more than 2.0 mmol resulted in fluorometric traces al-
most superimposed on that with 2.0 mmol glucose (Fig. 4a).
No progression curves were noted when glucose at concen-
trations less than 0.2 mmol was subjected to the reaction.
Those results are not included in Fig. 4 for simplicity. As
shown in Fig. 4, changes in fluorescence intensity showed
only minute dependency on glucose concentrations between
2.0 and 0.4 mmol (Figs. 4a—c). When the enzymatic reaction
with 0.2 mmol glucose was repeated under essentially the
same conditions, no reproducible traces were afforded (Figs.
4d—f): one of the runs gave a progression curve similar to
those observed for the cases with glucose at more than
0.4 mmol, and the other did not.
These results suggested the following points: 1) the rate of
reoxidation of GODred by 1 is almost saturated in the enzy-
matic reactions with glucose at more than 0.4 mmol in 3.0 ml
phosphate buffer, and 2) in the media deoxygenated as
strictly as possible, dissolved oxygen (DO) still remains in
around 0.1 mmol/3.0 ml and is preferred to 1 as an electron
acceptor from GODred. It was anticipated that totally anaero-
bic conditions could allow the present fluorometric method
to work for determination of glucose at much lower concen-
Fig. 2. Cyclic Voltammograms of 1 (0.1 mM) (a) and 2 (0.1 mM) (b) in
Phosphate Buffer at Room Temperature under an N2 Atmosphere
Scan rate: 0.1 V/s.
poor at 25 °C.12) In Fig. 3, a fluorometric trace obtained dur-
ing incubation of a mixture of 1 (10.0 nmol), glucose (5.0
mmol), and GOD (0.5 mg) in 3.0 ml phosphate buffer is com-
pared with that of a mixture of 2 (10.0 nmol), glucose (5.0
mmol), and GOD (0.5 mg) in 3.0 ml phosphate buffer at
25 °C. During incubation for 10.0 min, a total decrease in flu-
orescence intensity by reaction of 2 with GODred was less
than 10% of a total increase by reaction of 1 with GODred.
Thus, it was demonstrated that further reduction of newly
formed 2 into its dihydro derivative by GODred exerts almost
no effect on the fluorescent pink coloration by the transfor-
mation of 1 to 2 through the enzymatic reaction.
Possibility of the transformation of 1 to 2 as an indicator
reaction for glucose determination using only GOD was ex-
amined by a fluorometric method. Figure 4 compares fluoro-
metric traces obtained for incubation of a mixture of 1 (20.0
nmol), GOD (1.0 mg), and glucose (2.0, 1.0, 0.4 or 0.2 mmol)
in 3.0 ml phosphate buffer. The enzymatic reactions with glu-

624
Vol. 49, No. 5
Chart 2
trations. However, attempts to realize such conditions not
only by passing N2 gas thorough the media but also by addi-
tion of sodium sulfite or ascorbic acid and ascorbate oxidase
at various concentrations failed. This is because sodium sul-
fite and the reduced form of ascorbate oxidase rapidly re-
duced 1 to 2.
Quantitative elucidation of the efficiency of 1 as an elec-
tron acceptor in GOD-catalyzed oxidation of glucose was at-
tempted. The scheme shown in Chart 2 describes a plausible
reaction sequence based on a ping-pong type mechanism.
Fig. 5. Plot of [GOD]0/n against 1/[1] for Enzymatic Reactions of 1, Glu-
cose (5.0 mmol), and GOD (0.5 mg) in 3.0 ml Phosphate Buffer at 25 °C
During the enzymatic reaction, a progression curve with a 1.9ϫ104, and 5.5ϫ103 MϪ1 sϪ1, respectively)14) acts as elec-
larger change in fluorescence intensity was observed as the tron acceptors in GOD-catalyzed oxidation of glucose. How-
concentration of 1 or GOD was increased. Taking these facts ever, reductive bleaching of these dyes is not likely to func-
and the results in Fig. 4 into consideration, the process 1 is tion as an indicator reaction in GOD-dependent colorimetry.
much faster than the process 2 when the enzymatic reaction This is not only because the ability of these dyes to act
is carried out for glucose at a high concentration such as as electron acceptors is inferior to that of DO (kcat/KM:
10.0 mmol/3.0 ml. Accordingly, the following kinetic expres- 1.0ϫ106 MϪ1 sϪ1 at pH 7.4 and 25 °C)15) but also because re-
sion for consumption of 1 was derived with a total transit duced products of these dyes formed through reoxidation of
time inspection at a steady state.13)
[GOD]0/nϭ(kϪ3ϩk4)/k3[1]k4ϩ1/k4
GODred are generally susceptible to oxidative reverse reaction
to the original dyes by DO. Although kcat/KM for 1 is just
comparable to those for these dyes, the reduced product 2 is
quite stable in the presence of GODred as well as DO at
where [GOD]0 is the initial concentration of GOD.
When the reciprocals are taken, a Michaelis–Menten type 25 °C.
equation was derived.
In conclusion, the non-fluorescent compound 1 was shown
to act as an electron acceptor in GOD-catalyzed oxidation of
glucose, being reduced to the fluorescent compound 2. It is
nϭk4[GOD]0[1]/{[1]ϩ(kϪ3ϩk4)/k3}
where now the Michaelis constant (KM) and the catalytic con- believed that in GOD-catalyzed oxidation of glucose, 1 acts
stant (kcat) for the process 2 are (kϪ3ϩk4)/k3 and k4, respec- as an electron acceptor superior to the reported dyes from the
tively.
Substituting the first equation gives
standpoint of stability of coloration process. However, the
transformation of 1 to 2 as it is can not function as an indica-
tor reaction for enzymatic analysis of glucose using GOD. In
other words, practical application for glucose determination
[GOD]0/nϭKM/kcat[1]ϩ1/kcat
Thus, KM and kcat were estimated by plotting [GOD]0/n using GOD requires improvement of the rate of the fluoro-
against 1/[1]. Figure 5 shows the plot obtained for the enzy- metric coloration in the present methodology. The most
matic reactions for a mixture of 1, GOD (0.5 mg), glucose straightforward method for the improvement should be to in-
(5.0 mmol) in 3.0 ml phosphate buffer at 25 °C. A similar lin- crease the solubility of 1 by structural modification. Further
ear relationship between [GOD]0/n and 1/[1] was also ob- studies on this point are currently underway in our labora-
served, when the amount of GOD (0.6, 0.7, 0.8 or 1.0 mg in tory.
3.0 ml) was changed. From slopes and intercepts on the y
Experimental
axis of the obtained lines, KM and kcat were estimated to be
15.0Ϯ1.3 mM and (5.0Ϯ0.5)ϫ10Ϫ2 sϪ1. The apparent second-
order rate constant (kcat/KM) for the process 2 was then
3.3ϫ103 MϪ1 sϪ1. This value of kcat/KM is quite consistent
with 3.2ϫ103 MϪ1 sϪ1, which was independently estimated in
the same manner as used in the case of 2,12) namely, using the
assumption that the consumption of 1 obeys pseudo-first
order kinetics (nϭkcat/KM [GOD]0[1]). The estimated kcat/KM
for reaction of 2 with GODred at 36 °C was 6.6ϫ102
Reagents and Sample Solutions Resazurin sodium salt was purchased
from Wako Pure Chemical Industries, Ltd. and used as 1 without further pu-
rification. GOD from Aspergillus niger (EC 1.1.3.4, 170 units/mg) and glu-
cose were used as supplied. All solutions were prepared in phosphate buffer
(0.1 M, pH 7.4; Na2HPO4ϩNaH2PO4). Glucose solutions were stored
overnight to allow equilibration of a- and b-anomers. A solution of 2 was
prepared from acetyl resorufin recrystallized from ethyl acetate as reported
previously.12) All other chemicals were of reagent grade and were used with-
out further purification. Deionized and distilled water was used throughout
the present study.
M
Ϫ1 sϪ1.12) Thus, 1 reoxidized GODred more than 5 times
Apparatus and Procedures Cyclic voltammograms were obtained with
an ALS model 600 electrochemical analyzer under an N2 atmosphere. A
three-electrode configuration was employed: a glassy carbon electrode
(7.07 mm2), a saturated calomel electrode (SCE), and a platinum wire elec-
trode as a working, a reference, and a counter electrodes, respectively. All
spectrophotometric measurements were carried out under an N2 atmosphere.
faster than 2 (taking the reaction temperature into considera-
tion), in line with the voltammetric observation that 1 was re-
duced at a more positive potential than 2.
It was reported that some dyes such as methylene blue,
Meldola’s blue, and methylene green (kcat/KM: 1.3ϫ103, Enzymatic reactions were initiated by addition of a glucose solution (1.0 ml)

May 2001
625
to a mixture of 1 and GOD in 2.0 ml phosphate buffer in a spectrophotomet-
ric quarts cuvette (10ϫ10ϫ45 mm). Each of these solutions was deoxy-
genated by passing N2 gas through them for more than 10 min prior to spec-
trophotometric measurements. Absorption spectra were obtained with a
Hitachi U-3210 spectrophotometer without stirring at 25 °C. All fluoromet-
ric measurements were carried out using a JASCO Model FP-750 spectroflu-
orometer equipped with a JASCO ETC-272 Peltier thermostatted single cell
holder with stirring at 500 rpm at 25 °C. Measurement of the fluorescence at
excitation and emission wavelengths of 568 and 582 nm, respectively, was
started 15 s after addition of glucose solution. To obtain n in the enzymatic
reaction, changes in fluorescence intensity during 45 s after initiation were
used after normalization to the concentration of 2 based on the fact that fluo-
rescence intensity of 178.31 corresponds to 1.0 mM 2.
259 (1977).
3) Barnes S., Spenney J. G., Clin. Chim. Acta, 102, 241—245 (1980);
idem, ibid., 107, 149—154 (1980).
4) Soyama K., Ono E., Clin. Chim. Acta, 168, 259—260 (1987).
5) Castillo G. M., Thibert R. J., Microchem. J., 38, 191—205 (1988).
6) Cook D. B., Self C. H., Clin. Chem., 39, 965—971 (1993).
7) O’neill R. B., Dillon S. A., Morgan P. M., Biochem. Soc. Trans., 26,
S84 (1998).
8) Chapman J., Zhou M., Anal. Chim. Acta, 402, 47—52 (1999).
9) Candeias L. P., MacFarlane D. P. S., McWhinnie S. L. W., Maidwell N.
L., Roeschlaub C. A., Sammes P. G., Whittlesey R., J. Chem. Soc.,
Perkin Trans. 2, 1998, 2333—2334.
10) Herrmann R., Chimia, 45, 317—318 (1991).
11) Zhou M., Diwu Z., Panchuk-Voloshina N., Haugland R. P., Anal.
Biochem., 253, 162—168 (1997).
Acknowledgments This paper was supported in part by Mitsubishi
Chemical Corporation Fund and by a Grant-in-Aid for Exploratory Research 12) Maeda H., Matsu-ura S., Senba T., Yamasaki S., Takai H., Yamauchi
(12877353) from the Ministry of Education, Science, Sports, and Culture,
Japan.
Y., Ohmori H., Chem. Pharm. Bull., 48, 897—902 (2000).
13) Fersht A., “Enzyme Structure and Mechanism,” 2nd ed., W. H. Free-
man and Company, New York, 1985, pp. 98—120.
References and Notes
1) Guilbault G. G., “Methods in Enzymology,” Vol. 41, ed. by Wood W.
A., Academic Press, New York, 1975, pp. 53—56.
14) Kulys J., Buch-Ramussen T., Bechgaard K., Razumas V., Kazlauskaite
J., Marcinkeviciene J., Christensen J. B., Hansen H. E., J. Mol. Catl.,
91, 407—420 (1994).
2) De Jong D. W., Woodlief W. G., Biochim. Biophys. Acta, 484, 249—
15) Weibel M. K., Bright H. J., J. Biol. Chem., 246, 2734—2744 (1971).