Awfq’ticu Ckfrnica Acttc. 78 ( 1975)69-80 t’t Elsevicr Scientific Publishing Company.

A PHOSPHATE-SELECTIVE ALKALINE PHOSPHATASE

G. G. GUILBAULT Depnrtnrertt

(Rcccived

CJf

Chen~istry,

69 Amsterdam

- Printed

in The Ncthcrlanda

ELECTRODE BASED ON AND GLUCOSE OXIDASE

and M, NANJO ~wIwr*sit~ (JJ’New Orleu~rs. New Orleu~w. Louisia,ro

27th January

IMMOBILIZED

70122 (U.S.A.)

1975)

Ion-selective electrodes are now available or can be constructed for a wide variety of ions. Many investigators have attempted to construct electrodes for phosphate ion f --I ; however. all these electrodes have suffered serious interferences from common ions, such as chloride, nitrate and sulfate. and there is no report on a phosphate electrode which could be used for a practical assay of environmental and biological samples. Another approach for a phosphate-selective electrode is the use of selective enzymes, that is. an enzyme electrode in which the enzyme catalyzes a reaction involving phosphate ion. and the reaction is followed electrochemically. Weetall and Jacobson8 have reported the spectrophotornetric assay of phosphate ion based on a chromogenic substrate for alkaline phosphatase as follows: alkolinc

p-nitrophenyl (colorless)

phosphate

+ Hz0

!!!!!!YFllul;lrc p_nitrophenol (yellow)

+ HP02

-

(1)

The presence of added inorganic phosphate in this reaction mixture causes a shift in the equilibrium between the substrate and the products, with a reduction in color production which is reciprocally proportional to the added inorganic phosphate,, The assay of sulfate has also been reported by the same investigators. with alkyl sulfatase instead of alkaline phosphatasea. It is difficult to design an electrode based on this reaction, since there is no electrochemical difference between the substrate and products. The same’is true of other chromogenic substrates. such as o-carboxyphenyl phosphate’ and sodium thymolphathalin monophosphatelO. However, it was found possible to use a dual enzyme system for electroanalytical measurement of the alkaline phosphatase catalytic reaction as follows: ;rlk:rllnc glucose-G-phosphate + H 2O “!“lB glucose -i- HP02 (2) ylucclvc acid + H,02 glucose i- O2 a.gl.uconic (3) When the glucose formed from glucose-6-phdsphate is sensed electrochemically, it is possible to follow reaction (2), in which phosphate inhibition decreases glucose formation. thus lowering the response to the dual enzyme electrode. This dual enzyme electrode was also found to be useful for the assay of

G. G. GUILBAULT.

70 other oxyacids, such as arsenate. tungstate. for the substrate itself, glucose-6-phosphate.

molybdate

and borate

M. NANJO

ions as well as

EXPERIMENTAL

Immobilization of the dual entynres Alkaline phosphatase (100 mg; EC., 3.1.3.1.. Sigma, Type V. from chicken intestinal mucosa. 5 units/mg of solid) and glucose oxidase ( 100 mg; EC., I. I .3.4., Sigma, Type II, from aspergillus rtipr, 18 units/mg of solid) were dissolved in 5.0 ml of phosphate buffer (pH 7.8. 0.1 M). and then 10 drops of glutaraldehyde (Sigma, grade IV, aqueous 25% solution) were added to the mixture and stirred well. Next, the mixture was frozen by a dry ice-acetone coolant and kept in a refrigerator for one night. The sponge-like copolymer was washed with phosphate buffer. and finally washed out by glycine buffer (pH 8.4, 0.1 M) to eliminate phosphate, and stored in a refrigerator until used. Other combinations of the amounts of two enzymes, i.e., glucose oxidase and alkaline phosphatase, with albumin as a matrix, did not give higher activity than that without a matrix. Construction of the dual enzyme electrode A platinum disc electrode(Beckman Model 39273) was used as the solid base electrode to sense the oxygen change. The procedure for the preparation of the enzyme electrode was as follows, The immobilized dual enzyme (dry weight, about 5 mg; 45 units of glucose oxidase and 13 units of alkaline phosphatase if there is no loss of activity during the immobilization process) was mounted on the surface of the platinum electrode and secured with a nylon net (Pharmacia Fine Chemicals, Inc., Piscatway, N. J., Nylon Screen No. 0670) and “0” rings. A cellophane sheet could be instead of nylon net, as in the case of a previous glucose electrode’ I. Then the electrode was stored in the buffer solution at room temperature (23°C). Substrates P-D-glucose-6-phosphate(monosodium salt; Sigma) was used without further purification. This substrate was confirmed not to be contaminated by glucose, by means of a glucose enzyme electrode’ 2. A P-D-glucose (Sigma) was also used to test the dual enzyme function. Buffers Glycine buffer was prepared by adjusting the pH of a 0.1 M solution of glycine powder (Sigma) to pH 8.4. Tris-HCI buffer (pH 8.4. 0.1 M) was prepared from tris( hydroxymethyl)aminomethane (Sigma). Barbital buffer (pH 7.4. 0.05 M. with 0.1 Jfd KCl) was made from 5,5’-diethylbarbituric acid (Sigma), the pH being adjusted with sodium hydroxide. All buffers contained 10 mM magnesium chloride as a cofactor for alkaline phosphatase. Apparatus ad procedures A Heath polarograph

system (Model

EUA-19-2) was used as reported

for ear-

AN ENZYME

PHOSPHATE-SELECTIVE

ELECTRODE

71

her electrodes’ 2 I’. The dual enzyme electrode together with a calomel electrode, was placed into a stirred buffer solution. and a potential of -0.6 V us. SCE (where the current is proportional to the dissolved oxygen content) was applied. When the current reached a constant level. the phosphate solutions (0.01-I .O ml) to be assayed were pipctted into the buffer solution ( 10.0 ml). About 1 min later. a certain amount of glucose-dphosphate (CU. 35 mM. 0.5 ml) was in.jected into the sample-buffer mixture to initiate the reaction, and the initial rate of change in the dissolved oxygen limiting current was recorded. as well as the steady-state current. After the current had reached a steady state (l-2 min), the dual enzyme electrode was rinsed well with distilled water, and was dipped into a fresh buffer solution for I or 2 min to recover the dissolved oxygen level around the dual enzyme layer and to eliminate the remaining reactant, products and inhibitor, i.e., phosphate ion. All experiments were conducted at 30°C. The dual enzyme electrode could be used several hundred times for more than three months before it lost activity. RESULTS

AND DISCUSSION

Enzynw reaction-alkuline phosphatase Alkaline phosphatasc is a broad term associated with non-specific phomonoesterases with an optimal activity at alkaline pH as follows: orthophosphoric

monoester

+ H 2O

:nlkillinv ~h~~urilwl;wc

-

phos-

alcohol + H J PO4

(4) Assays of alkaline phosphatase in serum and milk are one of the major subjects in clinical chemistry l4 . This enzyme has been reported as a zinc metalenzyme and reacts with only orthophosphate to form a thermodynamically stable l s-1 ‘I. The equilibria between inorganic orthophosphate (Pi) and phosphoprotein alkaline phosphatase (E) can be represented as followsi6: . ”

E+Pi

* E-Pi.

kk’,*

k I

The hydrolysis E+S

of phosphate

- kyi * I

E*S -%

E-Pi + Hz0 ester (S) is expressed

(5) as follows:

E-P+Pi

The relationship at steady state has the Michaelis-Menten hibition as follows: u = k;[ES]

=

(6) form for competitive

in-

1~[Ed Ktll /_PiI Km 1 + [S] [sl ’ F

where K, and Ki are the Michaelis-Menten and inhibition constants, In the presence of an excess Pi, eqn. (7) can be simplified as follows:

respectively.

(8) At constant [S] and [E,], the rate of reaction, u, is proportional reciprocal of Pi, and Pi can be estimated by the decrease of the original the uninhibited reaction.

to the rate of

G. G. GUILBAULT.

72

M. NANJO

Enzynw reaction-glucose oxidase The second step of enzymatic reaction is the oxidation of glucose to gluconic acid. As reported previously11*13, glucose can be easily measured by an amperometric method, by means of a platinum electrode coupled with immobilized glucose oxidase at +0.6 V US.SCE. This second reaction was found to have no interference from phosphate ion, and glucose oxidase did not oxidize the glucose ester, that is, glucose-6-phosphate. Since the formation of glucose in the alkaline phosphatase reaction builds up glucose inside the dual enzyme layer, and glucose oxidase has enough activity to oxidize glucose faster than its rate of formation, the total reaction rate will be determined by the first step, and phosphate ion can be measured by means of the oxidation of glucose. This assumption was confirmed by the results that follow. Electrode response to glucose-6-phosphate and glucose Figure 1 shows the dual enzyme electrode response to substrate, glucose-6phosphate (added at A). The current at -0.6 V US. SCE, which is proportional to the amount of dissolved oxygen in the enzyme layer, decreased rapidly, and reached a steady state when the consumption of dissolved oxygen by the enzymatic reaction matched the supply of oxygen from the bulk solution to the enzyme layer by diffusion. This dissolved oxygen decrease could not be seen by the addition of

o

I 0

I

I 4

2 Time,

I

I

6

6

min.

Fig. 1. Changes in the dissolved oxygen current during enzymatic reactions: 30°C: glycinc bulk. 0.1 M: V IT. SCE. Curve I. 0.5 ml of glucose-6-phosphate (32 mM) :tdded to 10.0 ml of glycine pH 8.4; -0.60 bulfcr (10 mM MgClJ nt point A: 0.5 ml of 0.1 M phosphate solution added to this mixture at point B. Curve II shows the rcsponsc when glucose-6-phosphate and phosphate arc added at the same time (ut A). Fig. 2. Calibration buffer, 0.1 M; pH

curves for glucose-6-phosphutc (I) without itnd (II) 8.4: -0.60 V us. SCE. I. no phosphate. II, 4.6 mM

with phosphate ion: 30°C; phosphate added.

glycinc

AN ENZYME

PHOSPHATE-SELECTIVE

73

ELECTRODE

glucose-dphosphate to the glucose oxidase electrode, which indicates that glucose oxidase in the dual enzyme reacted only with glucose which was produced by the alkaline phosphatase catalytic reaction (eqn. (2)). At steady state, when phosphate ion, which was adjusted to the same pH as the buffer solution, was added at Ijoint B, an increase in the dissolved oxygen level was observed. indicating an inhibition15 of the enzyme reaction by phosphate ion, with a consumption of less oxygen. The addition of substrate and inhibitor at the same time resulted in a slower decrease of dissolved oxygen as shown in Fig. 1, curve II. In the case of a glucose oxidase electrode, the recovery of the dissolved oxygen level could not be obtained by the addition of phosphate ion. The same oxygen recovery in the dual enzyme electrode was observed in the presence of arsenate, molybdate, tungstate and borate solutions also. One might question why the phosphate formed from the substrate during reaction (eqn. 2) did not affect the rate of recovery of the dissolved oxygen, by product inhibition of reaction (2), as added phosphate does in Fig. 1B. However, placement of the electrode in a glucose-G-phosphate solution for a long time (5 h) did not show self-inhibition by such buildup of phosphate in the enzyme layer, because of rapid diffusion of the small amount of phosphate produced to the bulk solution. Figure 2 shows the calibration curves for glucose-6-phosphate, with and without’ phosphate ion added before injection of glucose-6-phosphate. In both cases, liiiear relationships were obtained and the same ratio of depression of the rate by

Fig.

3. Comparison of the rcsponsc of the dual enzyme clcctrodc glycinc buflkr. 0.1 M: pH 8.4; -0.60 V us. SCE. of Fig. 2.

glucose: 30°C: that

to (I) glucose-6-phosphate, A difkrent elcctrodc was

and (II) used from

CLG.

74

GUILBAULT.

M. NANJO

phosphate ion was observed at each glucose-6-phosphate concentration. Relative standard deviations were 5.9’;” without phosphate and 3.6% with 1.0 mM phosphate added. Glucose also reacts with glucose oxidase in the dual enzyme electrode and gave almost twice as high a rate as glucose-6-phosphate, as shown in Fig. 3. Thus, this dual enzyme electrode can be used for the assay of both substrates. in concentrations as low as 1 *10s4 M. Figure 3 also shows that the amount of glucose that diffuses into the enzyme layer from bulk solution is larger than that of glucose formed from the same amount of glucose-6-phosphate in the dual enzyme layer. Alkaline phosphatase has been known to require magnesium ion as a cofactor”. Figure 4 shows the effect of magnesium ion concentration on the rate. At the optimal magnesium concentration (10 mM) the reaction rate is 1.7 times faster than that without magnesium ion. Hence. all experiments were conducted, and all electrodes were stored. in a buffer solution which contained IO mM magnesium( II). ,A___

0.6

_.

t-

.A3 _ .__

-

.__..

___._.__

_ _._.

_

.

..~

_._

-

h_ _

.j

-I&J[Mg-] Fig. 4. Effixt of mugncsium phosphate was kept constant TABLE

--------_-.-

8.4:

-0.60

V IX. SCE.

Glucose-6-

I

EFFECT OF ELECTRODE

---_[Zit”]

ion: 3o’C: plycinc buffer. 0.1 M: pH at 0.85 mM. (0) No mugncsium ion.

(M)

2.4. Io-J 9.4*10-A 2.4.10-Z 8.7. lo- 2 -----._-

ZINC

AND

CALCIUM

--_Idtibiriotr

0 18 66 loo _----..-__

IONS

ON

THE

RESPONSE

OF

THE

--

- --.--

[Ca”] ----..-_-

- ----

2.3. 2.3. 4.6. 6.8.

(M) Idrihiliotr -.-.--.---_ ---.__

IO-J lo-” IO- * IO-’

. ._--

ENZYME

__

_..----_-_----_______ (Y,;) ------..----.

DUAL

(!f;;)

.___ _,

0 0 7 8

. .--. ..---- _ . -.. __.-_._-_._.____.___

The effects of other metal ions were also tested in barbital buffer solution. The results, shown in Table I, indicate a serious effect by an excess of zinc ion and slight effect by an excess of calcium ion. The normal levels of these ions in water would not pose an interference in phosphate assay. Solutions ( 1 mM) of copper, mercury,

AN ENZYME

PHOSPHATE-SELECTIVE

75

ELECTRODE

lend and cobalt ions were also examined by the oxygen (Fig, 1B): no interference on enzyme function was found,

level recovery

method

Stnce the enzyme electrode contains two enzymes. and the response rate is aflecled by the enzyme activity of both, the response of this dual enzyme electrode to glucose-&phosphate should show a mixed pH profile. Alkaline phosphatase from chicken intestine has been reported’” to show an optimal pH at 8-9, and glucose oxidase has an optimal %pHat 5.5 (ref. 19). _-._ .._ .._ .

_

,.L.LL_ ._____~ . ..l.___. 6 3

. _

_

..I. 6

.

-!

a

.” - 10 7

_ _.

_

L,

4

Fig:. 5. Efliict or pH on the dual enzyme %ctrodc: 30 C : -0.60 V US. SCE. I, Glycinc hufkr. 0.1 M: glucosu-6-phosphntu. 0.93 mM. II. Borbititl bulk, 0.05 M, with 0.1 M KCI: glucose-6ghosphatc. OS7 m&f. Q = Borate buff&, 0. I M: gl~tcos~-6-phnsphi~tc. 0.93 mN.

Figure 5 shows the pH profile for the dual electrode, indicating an optimal pH around 9 in glycine and barbital buffers. However, the pH profile was not a typically shaped curve. At pH values lower than 9, the depression of activity was not as steep as on the alkaline side in glycine buffer. In the case of barbital buffer, which contained 0.1 M KCI, a different profile was observed on the alkaline side, owing to the stabilization of magnesium ion by chloride ion. At lower pH values, even at 5, the dual enzyme was stilt found to work well. The activity profile can be explained by a compensation in the activity. of each enzyme and immobilization effects. Borate buffer could not be used. since borate ion was found to inhibit the enzyme in the same wily as phosphate ion. The pH effect of phosphate inhibition must also be considered; pH 7.4 guve more sensitive results than pH 8.4, because of the greater K; value at lower 17 20 PH The depression of the rate with increase in phosphate concentration is shown in Fig. 6. Also, the reciprocal of the rate was found to be proportionul to phosphate ion concentration added (in verification of eqn. (8)); measurement of phos-* phate ion by this method was found possible as low as 1 9 lo-* M. In Fig. 7, another

G. G. GUILBAULT.

76

M. NANJO

calibration curve for phosphate is shown, based on the recovery of oxygen level by added phosphate.after the attainment of a steady state (Fig. 1B). However, the oxygen recovery method, as can be seen from Fig. 1B. took a longer time than the decrease in oxygen level method (Fig. IA) and also the current change was smaller.

..+[PO,

_... -t-._.

_..+__- ____L_____ 16

-2k-

It ( lO%l

Fig. 6. Calibration curve for phosphate: 30°C: barbital us. SCE: glucoscd-phosphntc I .78 m M.

bulk.

0.05 M with 0. I M KCI, pH 7.4; -0.60

Fig. 7. Calibration curve for phosphate using the oxygen recovery method. 0.05 M with 0.1 M KCl. (pH 7.4). E= -0.60 V us. WE: glucose-6-phosphate

T=30”C.

barbital

V

bulk

I .78 mM.

A change in slope of the calibration curve was observed above 2 mM phosphate(equivalent to the glucose-6-phosphate concentration). so that the method was not so sensitive to low phosphate concentration. of tris bufytr Tris buffer enhances the activity of alkaline phosphatase because of acceptance of phosphate ion by tris forming tris phosphate ester2’. Figure 8 shows the effect of tris buffer on the phosphate inhibition of alkaline phosphatase activity at the same pH. As can be seen, about a one decade shift to lower sensitivity for phosphate was obtained because of this tris-phosphate reaction pathway, which enhanced the formation of glucose-6-phosphate. Ejjiict

A more powerful phosphate acceptor than tris buffer is ethanol. Figure 9 shows the serious effect from the presence of ethanol. Curve I is the response of the dual enzyme electrode to glucose-6-phosphate as in Fig. 1; when ethanol was

AN ENZYME

OI

PHOSPHATE-SELECTIVE

-_f

-----..-_

-..

---

77

ELECTRODE

A--___-_._-

-_

-lo+oLJt Fig. 8. Effect of tris-HCI bullkr on the dual cnzymc elcctrodc rcsponsc: 30°C; -0.60 V vs. SCE; G-phosphutc. 0.93 mM. 1. Glycinc bulk. 0.1 M, pH 8.4. II, ‘I’ris/HCI bulk. 0.1 M. pH 8.4.

glucose-

added at B, a rapid decrease of dissolved oxygen resulted. Curve II shows the response of the dual enzyme electrode to a mixture of substrate and ethanol. The rate was increased and also the steady-state current change caused by consumption of dissolved oxygen increased. This enhancement of the reaction reached up to a 180% increase of the initial rate (Table II). In the absence of glucose-6-phosphate, ethanol did not give any response to the dual enzyme electrode at first; however, several treatments of the dual enzyme electrode with ethanol solution caused a change of theelectrode characteristics: namely, the dual enzyme electrode responded to ethanol faster than to glucose or glucose-6-phosphate (Table III). This unexpected characteristic was found to be due to an alcohol oxidase impurity in glucose oxidase: this alcohol oxidase has been found during research 6

,i...

___& _._.____

._

4.

__;

..-.. $.-.

.

-I

Tim*,mln

Fig. 9. Effect of cthnnol on the dual cnzymc elcctrodc: 30’C: -0.60 V us. SCE; barbital bukr. 0.05 M with 0.1 M KCI. 1. Without ethanol. glucose-6-phosphate is added at A (I I.8 m&I. 0.5 ml) to 10.0 ml of buil’cr; at B nbsolutc cthunol (I.0 ml) is added. II. The same umounts of glucose-&phosphate and ethanol are uddcd at A.

78

G. G. GUILBAULT.

TABLE

M. NANJO

II

EFFECT

OF

ETHANOL

(M)

[EtOH]

DUAL

ENZYME

ELECYRODE

RESPONSE

Incretrse in rute (I%,)

x2*10-5 1.1 * 10-Z 5.5~ 10-Z TABLE

ON THE

159 171 176 III

RESPONSE ETHANOL

OF

THE

DUAL

ENZYME

TO

GLUCOSE-6-PHOSPHATE.

GLUCOSE

-

-

Slrhstrute (M)

Rutc

l.O* 10-J

2.0. 10-4 4.0. 10-o

--

AND

mill-‘)

(pA

.

Glucose-6-pho.s/)lrale ~__

Glucose

Bt/rutd

0.65 I .33 2.60

1.10 2.05 3.41

3.85 6.43 9.9 1

---

on glucose oxidase 22. The results in Table III indicate that this dual enzyme electrode could also be used for the assay of glucose-6-phosphate, glucose and ethanol. Inhibitim of alkalirw phosphatase by oxyacitls One of the inhibitors of alkaline phosphatase is an oxyacid which can combine with the serine hydroxyl group, at the active center of the enzyme, to form esters; this results in an inhibition of the hydrolysis of substrate ester”. Figure 10 shows the effect of various oxyacids on the response of the dual enzyme electrode. The order of inhibition at pH 7.4 was found to be as follows: WOi- >AsOi- >, 1.2

7

x

1.0-.

‘.

0.e :-

‘\

\

‘\

‘\

0.6~-

1 t

04

0.2 0

L

_

.

_

1

-----. ._.

I..

‘\

‘\

‘\

‘\

.1

.-

I

-

-_ -Y.-’

‘\

.

.

-

‘\

\

\

\

\

_.__..

---

.I__ __-

\

Pi- ”

‘.._ .-II

____’

\

I\0

‘L, \ ‘.

-1.

--_______‘_Y’l

2.

*\

--

VII

-I i

q

I

-----k--

_---. .-+-A ---- _;-._

1

-WllXl Fig. 10. Efkct of vurious anions on the dual cnzymc clectrodc. These inhibitors wcrc added 2 min before the addition of glucose-6-phosphate. 3O’C: -0.60 V us. SCE: barbital bullix. 0.05 M. pH 7.4. with 0.1 M KCI. (I) NazWO4, (II) NaHrA~04. (III) NilH,PO,. (IV) NatM004. (V) NazB40,. (VI) EDTA. (VII) NH&I. (VIII) NuOAc.

AN ENZYME

PHOSPHATE-SELECTIVE

ELECTRODE

79

PO2 - >MoO:->BO;-. Arsenite, had no effect on the enzyme activity up to 0.1 M, in contrast to arsenate. As can be seen from Fig. 10, arsenate, tungstate, molybdate and even borate ions can be assayed as well as phosphate ion with the dual enzyme electrode. However, these mixtures could be separated and identified before measurement. In an assay for phosphate in rivers, streams or lakes, it is unlikely that any of these ions would be present to interfere. The other common ions that exist in water solutions, such as chloride, nitrate and sulfate, which gave serious interferences on non-enzymatic ion-selective electrodes. did not show any effect up to 0.1 M. The buffer solution used has a relatively high concentration of potassium chloride (0.1 M) and also magnesium chloride (0.01 M). Bromide and iodide ions have no effect up to 0.1 M. The other possible inhibitors are reagents which can form complexes with the zinc in the enzyme protein resulting in the extraction ofzinc from the enzyme or with magnesium ion used as a cofactor. As can seen in Fig. 10 (curve VI), EDTA interfered seriously above 15 mM, because of chelation of all the magnesium ion. Ammonium ion (curve VII) also affected the electrode response, but the effect was less than that of EDTA. Other inhibitors, like cyanide and high concentrations of heavy metal ions, were not tested. However. the addition of Mg-EDTA or Zn-EDTA could be helpful to protect the enzyme from serious denaturation by heavy metal ions. The main disadvantage of the dual enzyme electrode is the difl‘lculty in use in biological fluids which already contain glucose. In this case, the glucose in blood or biological fluid should be removed or measured before phosphate measurements. Another difficulty of this dual enzyme electrode could be possible inhibition and/or activation by other unknown chemicals in river water or biological fluids. The dual enzyme electrode at present is not sensitive enough for assay of low concentrations of phosphate ion in river water; but the sensitivity for phosphate was found to depend on enzyme activity; namely, when the dual enzyme gave less response to glucose-6-phosphate, the sensitivity for phosphate was also less. It might be necessary to use more active enzyme or a more stable form and more substrate and possibly other oxygen-sensing devices. which could be used to make a more practical enzyme electrode for phosphate, and even sulfate. based on the same idea of a dual enzyme system. Electrode stcthility The daily response of the dual enzyme electrode to the substrate. glucose-6phosphate, was observed during storage in buffer solution. Daily calibration was necessary, and the non-inhibition rate was compared with that of phosphate samples. After 3 months, the dual enzyme could still be used to measure phosphate over 10e4 M; even in solutions as low as pH 2.5 the enzymes remained active. After acidification, the enzymes recovered their activity reversibly. Even after EDTA treatment of the dual enzyme, the addition of zinc and magnesium ion could bring back some activity to the enzyme electrode. The authors gratefully acknowledge the financial support of the Environmental Protection Agency (Grant No. EPA 800359) and the National Institutes of Health (Grant No. GM 17268). We would also like to thank Dr. S. S. Kuan for his helpful advice and suggestions.

80

G. G. GUILBAULT.

M. NANJO

SUMMARY

An enzyme electrode which senses oxygen consumption for the assay phosphate ion ( 10-3-10-4 M), was constructed by using two enzymes together:

of

glucose-6-phosphate glucose

*“k”‘incphOu”“““‘*~ glucose + phosphate ylUcouu ~lxidauc + gluconic acid-t- HzOz 02 The competitive inhibition by phosphate ion added caused a smaller and slower oxygen consumption which could be detected by a platinum disc electrode at -0.6 V us. SCE amperometrically. Thig dual enzyme electrode was also found useful for theassay ofoxyacidsother than phosphate, such as arsenate. tungstate, molybdute and borate. REFERENCES

1 E. Pungor. 2 3 4 5

6 7 8 9 IO 11 12 13 14

15 16

17 18 19 20 21 22

G. G. 1. F.

K. Toth and J. Havns. Mlkroclliw. Acru. 4 (196G) 689. Rcchnitz, 2. F. Lin and S. B. Zamochnick. And. Lcrr.. 1 (1967) 29. G. Guilbault and P. J. Brignuc. Jr.. Artul. Clrirn. Actu. 56 (1967) 139; Nagclberg, L. Braddock and G. Barbcro. Scler~ce. 166 (1969) 1403. R. Shu and G. G. Guilbuult. Am~l. Lerr., 5 (1972) 559.

Awl.

C/XVII.. 41 (1969)

C. J. Coctzee and H. Frciscr. Awl. Chem.. 40 (1968) 2071: 41 (1969) 1128. M. Nunjo and G. G. Guilbault. AmI. Clrlnt. Aurtr, in press. H. H. Wcctull und M. A. Jucobson, IFS: Fertwwf. Tdrrwl. Tuduy, Proc. 1 V. (1972) H. Brondcrbcrgcr and R. Hanson. Ifelv. Chlm. Acru. 36 (1953) 900. A. V. Rou. C/h Chm.. 16 ( 1970) 43 1. G. G. Guilbault und G. J. Lubrano. Awl. Chiln. Acru. 64 (1973) 439. M. Nanjo and G. G. Guilbu’ult, Anal. Chim. Acra, 73 (1974) 367; 75 (1975) 169.

M. Nanjo and G. G. Guilbault.

AMJ/. Chcnr.. 46 (1974)

1136.

361.

1769.

Sigma Chemical Co.. St. Louis. Missouri. (ISA. L. Engstrom and G. Agrcn. Acru Chm. Scuml.. 12 (1958) 357. A. Garen and C. Levinthal. Blocltlrn. Biopkys. Acfcr. 38 (1960) 470. T. W. Reid and 1. B. Wilson in P. D. Boycr (Ed.). The E~I~JwI~~. Vol. IV. Academic Press. New York. 3rd edn.. 1973. p. 373. G. Y. Shinowara. L. M. Jones and H. L. Rcinhart. J. Bid/. C/rem., 164 (1964) 32. H. J. Bright und M. Applcby. J. Bid. Chcnr.. 244 (1969) 3625. T. W. Reid, M. Paulic. D. J. Sullivan and I. B. Wilson, Eiochwisfrp, 8 (1969) 3184. J. Dnyan and 1. B. Wilson, Biochiw. f3ioplrys. Actu. 8 1 ( 1964) 620. F. W. Janssen and H. W. Ruclius. Biochh. Elophys. Acra, 151 ( 1968) 330. ,Si,/rrw

T~L

L3~rllr~irr.s. No.

85.

No. 104. mrtl No. 104 M: