CIinica C&mica Acta, 203 ( 1991J 295-304 0 1991 Elsevier Science Publishers B.V. All rights reserved 0009-8981/91/$03.50
295
CCA 0514.5
Rapid and sensitive discriminating determination of acetylcholinesterase activity in amniotic fluid with a choline sensor R.M. Morelis ‘, P.R. Coulet I, A. Simplot *, C. Boisson * and G. Guibaud * ’ Laboratoire de G&e Enzymotique, Unicersite’ Lyon, Wleurbanne and ’ Laboratoire Central, H6pital de la Croir Rouse, Lyon (France)
(Received 21 January 1991; revision received 29 August 1991; accepted 16 September 1991) Key words: Acetylcholinesterase;
Butyrylcholinesterase;
Amniotic fluid: Choline; Enzyme electrode
Summary A simple method for the separate determination of acetylcholinesterase and butyrylcholinesterase activities in amniotic fluid is reported. This determination is performed with an enzyme electrode involving an immobilized choline oxidase membrane associated with the amperometric detection of hydrogen peroxide. Acetylcholine or butyrylcholine, in the presence of samples containing acetylcholinesterase or butyrylcholinesterase are specifically hydrolyzed, the formation of choline being detected vs time by the sensor with no need for a selective inhibitor. The dynamic linear ranges for acetylcholinesterase and butyrylcholinesterase are respectively 100 PU to 10 mU and 30 PU to 3 mU per ml sample.
Introduction The detection of acetylcholinesterase (EC 3.1.1.7) activity in amniotic fluid (AF) is considered to be the best test for the diagnosis of neural-tube defects (NTD). High activities (> 5 mu/ml) of acetylcholinesterase at 14-23 weeks of pregnancy are significantly associated with fetal abnormalities [ 11. The presence of acetylcholinesterase activity in amniotic fluid can be determined by qualitative [2] and quantitative [3,41 assays, but the direct assessment of
Correspondence to: R.M. Morelis, Laboratoire de Genie Enzymatique, UMR 106 CNRS, UniversitC Lyon 1, 43, Boulevard du 11 Novembre 1918, 6%22 Villeurbanne Ckdex, France.
296
acetylcholinesterase is hampered by its low activity and by the presence of butyrylcholinesterase which overlaps in substrate specificity with acetylcholinesterase. As chromatographic and spectrophotometric methods cannot discriminate between acetylcholinesterase present in amniotic fluid and butyrylcholinesterase from contaminating fetal or maternal blood, the true acetylcholinesterase activity is characterized by the use of a specific inhibitor [5]. Attempts to use a choline enzyme electrode have not been reported for the determination of cholinesterases in the amniotic fluid: they have been used in buffer [6] or serum [7-91. The main feature of the method described in this paper is that acetylcholinesterase and butyrylcholinesterase activities can be determined separately in whole amniotic fluid without any pretreatment, using a choline enzyme electrode developed in our group [lo]. This enzyme sensor incorporates choline oxidase immobilized on a preactivated polyamide membrane associated with an amperometric electrode giving an electrical signal depending on the concentration of choline released by the hydrolysis of either acetyl or butyrylcholine added separately in the reaction medium. Materials and methods Samples
Amniotic fluid samples provided by the Laboratoire de Biochimie, Hopital de la Croix Rousse, Lyon, France, were obtained by amniocentesis of pregnant women. All samples were centrifuged at 2000 X g for 10 min at room temperature and stored at 4°C. Reagents
Choline oxidase (EC 1.1.3.17, from Alcafigenes 10 U/mg), acetylcholinesterase (EC 3.1.1.7, lyophilized powder type VS from electric eel, 1300 U/mg), acetylcholine chloride, butyrylcholine chloride and choline chloride, acetylcholine, 5,5’dithiobis (Znitrobenzoic acid) were supplied by Sigma (St Louis, MO, U.S.A.), ethopropazine was a Rhone-Poulenc product. Butyrylcholinesterase (EC 3.1.1.8) was a generous gift from Dr BCguC (Centre de Transfusion sanguine, Lyon, Beynost, France). Other reagents of the highest grade available were obtained from Prolabo (Paris). Membranes
The polyamide membranes, Immunodyne type, (120 pm thickness and 3 pm size cut off) provided in a chemically preactivated form were supplied by Pall Industrie s.a. (St Germain-en-Laye, France).
297
Enzyme immobilization
The simple wetting method for enzyme immobilization [ll] was used: 10 ~1 of choline oxidase solution (10 g/l-‘) in 0.1 mol/l potassium phosphate buffer, pH 6, were applied on each side of the membrane disk and left to react for one minute, as previously described for the design of the choline sensor [lo]. Enzyme electrode principle
The method is based on the oxidation of choline catalyzed by choline oxidase immobilized on the membrane according to reaction 1 Choline oxidast
choline + 20, + H,O
betaine
+ 2 H,O, (11
The enzymatically generated hydrogen peroxide on a platinum anode at +650 mV vs. Ag/AgCl into a polarograph PRGE (Tacussel, France). maintained in close contact with the platinum tip Pt/anode I-w,
+ 650
mV Ag/AgCI/CI
> -
is detected by anodic oxidation reference (reaction 21, plugged The enzymatic membrane was with a screw-cap.
0,+2H++2e-
Choline can be directly added to the medium for sensor calibration or released from the hydrolysis of butyrylcholine or acetylcholine when added in the medium containing acetylcholinesterase or butyrylcholinesterase according to reaction 3 or 4. acetylcholine + H,O butyrylcholine
+
H,O
Acetylcholinesterase
+ choline f acetic acid
Bwrylcholinesterase
+ choline + butyric acid
Procedure
The sensor was immersed in a thermostated cell (27°C) containing 20 ml of 0.1 mol/l KCl, 0.1 mol/l potassium phosphate buffer, pH 8. A portion of 200 ~1 of amniotic fluid was added and the variation of current was recorded. The current output increased and reached a steady-state within 2-3 min. The enzymatic reaction (hydrolysis of acetylcholine or butyrylcholine) was then initiated by adding 20 ~1 of 0.1 mol/l acetylcholine or butyrylcholine solution. The enzymatic activity was calculated from the slope of the current-time curve [12]. To determine the enzymatic avtivity, the sensor was first calibrated by addition of choline solution under the same experimental conditions.
298
Results Calibration of the choline sensor
As the measurement of acetylcholinesterase or butyrylcholinesterase activity is based on the monitoring of choline generated by the hydrolysis of acetylcholine or butyrylcholine, the response of the electrode to choline was first determined. As shown in Fig. 1, there was a linear relationship between the variation of current and choline concentration from 5.10-’ mol/l to 5.10-’ mol/l. The sensitivity obtained, expressed by the slope of the calibration curve was 2.6 mA * l/mol. Response of the sensor to cholinesterases standard activities
When butyrylcholinesterase was added into the measuring cell, the current increased linearly with time due to the release of choline and the slope was proportional to the enzyme activity. The linear dependence was demonstrated using different dilutions of a butyiylcholinesterase solution (120 U - ml- ‘) between 30 pI_J * ml-’ and 3 mU * ml-’ in the injected standard sample (Fig. 2a). In the same way, Fig. 2b shows the linear calibration graph obtained with a commercial solution of acetylcholinesterase (1230 U * ml-‘) between 100 PU * ml-’ and 10 mU * ml-‘. Measurements in biological samples
Figure 3 shows the response of the enzyme electrode for both acetylcholinesterase and butyrylcholinesterase activity determinations, when experiments were performed with amniotic’ fluid samples. After the probe was dipped into the reaction medium, a baseline was recorded (Section A). When a portion of amniotic
102 z 5 E P 2
10
1
1o-1
1o-8
1o-7
10-6
10-5
10-d
10-a
Choline (M)
Fig. 1. Calibration curve of the choline oxidase electrode (log scale). Measurements were in 0.1 mol/l potassium phosphate buffer, 0.1 mol/l KC1 at 27°C.
done at
pH 8,
299
‘* 7
.5 E =1
1
lo pJ
piJ
Ly,=( ; ‘;
r E $
*,j
1
I//
*,1
_,_‘_
0,Ol
0,l
1
10
100
0,oi
0‘1
(. 1
10
100
mu. ml-l mu. ml-t Fig. 2. Dependence of choline sensor response to cholinesterase activities in the injected standard sample. (a) Butyrylcholinesterase activity; (b) acetylcholinesterase activity.
fluid (200 ~1) was injected (arrow on the graph) electroactive substances such as urate and bili~bin present in the fluid were immediately oxidized at the set electrode potential, as shown in the first part of the curve (Section B). After completion, a plateau was reached and 20 ~1 of the chosen substrate solution containing either butyrylcholine (a) or acetylcholine (b) were injected to initiate either the butyrylcholinesterase or acetylcholinesterase reaction. Choline produced
2 min V-4
2 min Y
Fig. 3. Typical time courses obtained with the choline oxidase sensor in response to cholinesterase activities. (a) Butytylcholinesterase activity of 200 PI of a pathologic amniotic fluid fspida bifida) (1) or 200 &I of a non-pathologjc amniotic fluid (II); fb) acetylcholinesterase activity of 200 ~1 of a pathologic amniotic fluid (spida bifida) (I) or 200 ~1 of a non-pathologic amniotic fluid (If).
300 TABLE I Tests of specificity for acetylcholinesterase acetylcholine and butyrylcholine Enzyme
AChE BuChE BuChE + AChE Results (slope of current-time enzyme.
(AChE) and buty~lcholinesterase
(BuChE) with regard to
Substrate Acetyl choline
Bu&Wl choline
0.92 0.17 0.92
0 0.17 0.77
curves) are expressed in nA.min-’
for the same amount of each
in either case by the enzymatic reaction reacted with the immobilized choline oxidase and generated hydrogen peroxide yielding a current variation (Section 0. This variation was proportional to the activity of the cholinesterase present in the sample. After 2 or 3 min the measurement was stopped, the electrode was washed and the cell was refilled for the next analysis. The trace recording I of Fig. 3a corresponds to the butyrylcholinesterase activity of a NTD amniotic fluid while trace II corresponds to the enzymatic activity of a non-pathologic sample. A similar recording was shown in Fig. 3b for acetylcholinesterase activity determined in the same samples. It must be stressed that an immediate recognition of pathologic amniotic fluids is easy by simply examining the recording patterns for both bu~~lcholinesterase and ace~lcholinesterase activities. Validation of the method
As the amniotic fluid contains the two cholinesterases, it was necessary to verify the absence of interference between the two enzymatic reactions. Table I shows that ace~lcholinesterase hydrolyzed only ace~lcholine. Bu~~lcholinesterase however hydrolyzed acetylcholine and butyrylcholine, but with very different rates. This can be explained by the affinity and the maximal rate of the butyrylcholinesterase which is higher for its normal substrate than for acetylcholine. Obviously, as shown in Table I for a typical experiment, the presence of both buty~lcholinesterase and ace~lchoiinesterase in the solution does not modify the hydrolysis rate of acetylcholine by ace~lcholinesterase, since a value of 0.92 nA.min-’ was obtained in both cases. Furthermore, it was necessary to check that the amniotic fluid did not disturb the immobilized choline oxidase reaction. For this purpose, a portion of choline was injected into the measurement cell in the presence, or not, of amniotic fluid. Figure 4a shows that 20 ~1 of 10e3 mol/l choline induced the same current variation (2.8 nA) in the absence (I) or in the presence (II) of amniotic fluid. In the same way, hydrogen peroxide was injected in the measurement cell and Fig. 4b shows that 10 ~1 of 10e3 mol/l H,O, induced the same current variation (2.3 nA) in the absence (I) or in the presence (II) of amniotic fluid, thus demonstrating that
301
1
1nA
J J t
Hz02 * 2 min
T
amniotic fluid \
-
Fig. 4. Enzyme electrode response to: (al 20 ~1 of 10e3 mol/l choline injection in the measurement cell, without biological sample (I) or after an injection of 200 ~1 of amniotic fluid (II). (b) 10 ~1 of 10m3 mol/l H,O, injection in the measurement cell, without biological sample (11or after an injection of 200 ~1 of amniotic fluid (II).
the different components of the fluid did not interfere with the enzyme electrode response. The activities of acetylcholinesterase from different amniotic fluid samples are given in Table II. The results show that neural-tube defects (NTD: anencephaly and open spida bifida) are associated with a higher amniotic fluid acetylcholinesterase activity (12 and 7.3 respectively) compared to 3.7 mU * ml-’ for non pathologic samples. Furthermore, the presence of blood in the sample had no influence on the results of acetylcholinesterase activity (butyrylcholinesterase activity increased in these conditions, due to its presence in the blood). The technique was found reliable since repeated injections from the same sample (n = lo), a good precision was obtained with a CV value lower than 2.5%.
TABLE II Comparison of enzymatic activities determined with the enzyme electrode and with the spectrophotometric technique Samples
Normal amniotic fluid Blood stained amniotic fluid NTD (anencephalyl NTD (Spida bifidal
Enzymatic activities (mU/ml) Enzymatic electrode technique
Spectrophotometric technique
3.7*0.3 3.8 12 7.3
3.1 kO.3 3.4 12.4 7.3
302
As a control, acetyl~holinesterase activity was also determined by the conventional spectrophotometric method described by Ellman et al. [I31 and modified by Dale et al. [14]. Acetylthiocholine was used as substrate. 5,5’-dithiobis (Znitrobenzoic acid) was added to react with the released thiocholine to form the yellow compound Sthio-2-nitrobenzoic acid determined at 412 nm. The selective measurement of acetylcholinesterase activity is performed by the concomitant use of ethopropazine hydrochloride which inhibits the non-specific cholinesterase activities. Table II shows that both methods give similar results. Discussion The results presented here demonstrate that the potentialities of the amperometric choline sensor, (very low detection limit, 5.10-’ mol/l, and large linear range, up to 5.10-” mol/l) can be extented to the determination of cholinesterase activities. The separate determinations of butyrylcholinesterase and acetylcholinesterase can be performed rapidly and accurately after specific substrate hydrolyzis by monitoring choline appearance. The linear range was from 30 PIJ-ml-’ to 3 mU * ml - ’ for bu~~l~holinesterase and from 100 $I * ml - ’ to 10 mU . ml-’ for acetylcholinesterase. Cholinesterase activity assays are routinely performed in clinical laboratories but with procedures involving preincubation of the biological sample with an inhibitor in order to specifically identify the acetylcholinesterase activity. The procedure proposed here avoids this step and provides a distinct measure of the activity of the two enzymes in whole amniotic fluid through the use of specific substrates. On the contrary to spectrophotometric methods, enzyme electrodes are not subject to optical interferences and can be operated even in a turbid medium so it is possible to accurately determine the activities of specific and non-specific cholinesterases in amniotic fluid contaminated by maternal or fetal blood. In conclusion, this approach appears particularly promising for a rapid determination of acetyfcholinesterase activity in amniotic fluid. References 1 Smith AD, Wald NJ, Cuckle HS, Stirrat GM, Bobrow M, Lagercrantz H. Amniotic-fluid acetylcholinesterase as a possible diagnostic test for neural-tube defects in early pregnancy. Lancet 1979;i:685-688. 2 Barlow RD. Cuckle HS, Wald NJ. A simple method for amniotic fluid gel-acetylchoiinesterase determination, suitable for routine use in the antenatal diagnosis of open neural tube defects. Clin Chim Acta 1982;119:137-142. 3 Hay DL. lbrahim GF, Horacek I. Rapid acetylcholinesterase screening test for neural-tube defect. Clin Chem 1983;29:1065-1069. 4 Hullin DA, Laurence KM, Elder GH et al. Amniotic fluid cholinesterase measurement as a rapid method for the exclusion of fetal neural-tube defects. Lancet 1981;i:325-326. 5 King ME. Cholinesterase. In: Pesce AS, Kaplan LA eds. Methods in Clinical Chemistry. The C.V. Mosby Company. St Louis, Washington DC, Toronto. 1987;161-168. 6 Mizutani F, Tsuda K. Amperometric determination of cholinesterase with use of an immobilized enzyme electrode. Anal Chim Acta 1982;139:359-362.
303 7 Yao T. Flow injection analysis for cholinesterase in blood serum by use of a choline-sensitive electrode as an amperometric detector. Anal Chim Acta 1983;153:169-174. 8 Gruss R, Scheller F. Electrochemical determination of cholinesterase activity and indication of its inhibitors using a thick-film metallized platinum electrode. Anal Lett 1989;22:1159-1169. 9 Palleschi G, Lavagnini MG, Moscone D, Pilloton R, D’Ottavio D, Evangelisti ME. Determination of serum cholinesterase activity and dibucaine numbers by an amperometric choline sensor. Biosens Bioelec 1990;5:27-35. 10 Morelis RM, Coulet PR. Sensitive biosensor for choline and acetylcholine involving fast immobilization of a bienzyme system on a disposable membrane. Anal Chim Acta 1990;231:27-32. 11 Assolant-Vinet CH, Coulet PR. New immobilized enzyme membranes for tailor-made biosensors. Anal Lett 1986;19:875-885. 12 Blum LJ, Bertrand C, Coulet PR. Amperometric sensor and immobilized enzyme electrode for the determination of enzymatic activities. Anal Lett 1983;16:525-540. 13 Ellman GL, Courtney KD, Andres V, Featherstone RM. A new and rapid calorimetric determination of acetylcholinesterase activity. Biochem Pharmacol 1961;7:88-95. 14 Dale G, Archibald A, Bonham JR, Lowdon P. Diagnosis of neural tube defects by estimation of amniotic fluid acetylcholinesterase. Br J Obstet Gynaecol 1981;88:120-125.
295
CCA 0514.5
Rapid and sensitive discriminating determination of acetylcholinesterase activity in amniotic fluid with a choline sensor R.M. Morelis ‘, P.R. Coulet I, A. Simplot *, C. Boisson * and G. Guibaud * ’ Laboratoire de G&e Enzymotique, Unicersite’ Lyon, Wleurbanne and ’ Laboratoire Central, H6pital de la Croir Rouse, Lyon (France)
(Received 21 January 1991; revision received 29 August 1991; accepted 16 September 1991) Key words: Acetylcholinesterase;
Butyrylcholinesterase;
Amniotic fluid: Choline; Enzyme electrode
Summary A simple method for the separate determination of acetylcholinesterase and butyrylcholinesterase activities in amniotic fluid is reported. This determination is performed with an enzyme electrode involving an immobilized choline oxidase membrane associated with the amperometric detection of hydrogen peroxide. Acetylcholine or butyrylcholine, in the presence of samples containing acetylcholinesterase or butyrylcholinesterase are specifically hydrolyzed, the formation of choline being detected vs time by the sensor with no need for a selective inhibitor. The dynamic linear ranges for acetylcholinesterase and butyrylcholinesterase are respectively 100 PU to 10 mU and 30 PU to 3 mU per ml sample.
Introduction The detection of acetylcholinesterase (EC 3.1.1.7) activity in amniotic fluid (AF) is considered to be the best test for the diagnosis of neural-tube defects (NTD). High activities (> 5 mu/ml) of acetylcholinesterase at 14-23 weeks of pregnancy are significantly associated with fetal abnormalities [ 11. The presence of acetylcholinesterase activity in amniotic fluid can be determined by qualitative [2] and quantitative [3,41 assays, but the direct assessment of
Correspondence to: R.M. Morelis, Laboratoire de Genie Enzymatique, UMR 106 CNRS, UniversitC Lyon 1, 43, Boulevard du 11 Novembre 1918, 6%22 Villeurbanne Ckdex, France.
296
acetylcholinesterase is hampered by its low activity and by the presence of butyrylcholinesterase which overlaps in substrate specificity with acetylcholinesterase. As chromatographic and spectrophotometric methods cannot discriminate between acetylcholinesterase present in amniotic fluid and butyrylcholinesterase from contaminating fetal or maternal blood, the true acetylcholinesterase activity is characterized by the use of a specific inhibitor [5]. Attempts to use a choline enzyme electrode have not been reported for the determination of cholinesterases in the amniotic fluid: they have been used in buffer [6] or serum [7-91. The main feature of the method described in this paper is that acetylcholinesterase and butyrylcholinesterase activities can be determined separately in whole amniotic fluid without any pretreatment, using a choline enzyme electrode developed in our group [lo]. This enzyme sensor incorporates choline oxidase immobilized on a preactivated polyamide membrane associated with an amperometric electrode giving an electrical signal depending on the concentration of choline released by the hydrolysis of either acetyl or butyrylcholine added separately in the reaction medium. Materials and methods Samples
Amniotic fluid samples provided by the Laboratoire de Biochimie, Hopital de la Croix Rousse, Lyon, France, were obtained by amniocentesis of pregnant women. All samples were centrifuged at 2000 X g for 10 min at room temperature and stored at 4°C. Reagents
Choline oxidase (EC 1.1.3.17, from Alcafigenes 10 U/mg), acetylcholinesterase (EC 3.1.1.7, lyophilized powder type VS from electric eel, 1300 U/mg), acetylcholine chloride, butyrylcholine chloride and choline chloride, acetylcholine, 5,5’dithiobis (Znitrobenzoic acid) were supplied by Sigma (St Louis, MO, U.S.A.), ethopropazine was a Rhone-Poulenc product. Butyrylcholinesterase (EC 3.1.1.8) was a generous gift from Dr BCguC (Centre de Transfusion sanguine, Lyon, Beynost, France). Other reagents of the highest grade available were obtained from Prolabo (Paris). Membranes
The polyamide membranes, Immunodyne type, (120 pm thickness and 3 pm size cut off) provided in a chemically preactivated form were supplied by Pall Industrie s.a. (St Germain-en-Laye, France).
297
Enzyme immobilization
The simple wetting method for enzyme immobilization [ll] was used: 10 ~1 of choline oxidase solution (10 g/l-‘) in 0.1 mol/l potassium phosphate buffer, pH 6, were applied on each side of the membrane disk and left to react for one minute, as previously described for the design of the choline sensor [lo]. Enzyme electrode principle
The method is based on the oxidation of choline catalyzed by choline oxidase immobilized on the membrane according to reaction 1 Choline oxidast
choline + 20, + H,O
betaine
+ 2 H,O, (11
The enzymatically generated hydrogen peroxide on a platinum anode at +650 mV vs. Ag/AgCl into a polarograph PRGE (Tacussel, France). maintained in close contact with the platinum tip Pt/anode I-w,
+ 650
mV Ag/AgCI/CI
> -
is detected by anodic oxidation reference (reaction 21, plugged The enzymatic membrane was with a screw-cap.
0,+2H++2e-
Choline can be directly added to the medium for sensor calibration or released from the hydrolysis of butyrylcholine or acetylcholine when added in the medium containing acetylcholinesterase or butyrylcholinesterase according to reaction 3 or 4. acetylcholine + H,O butyrylcholine
+
H,O
Acetylcholinesterase
+ choline f acetic acid
Bwrylcholinesterase
+ choline + butyric acid
Procedure
The sensor was immersed in a thermostated cell (27°C) containing 20 ml of 0.1 mol/l KCl, 0.1 mol/l potassium phosphate buffer, pH 8. A portion of 200 ~1 of amniotic fluid was added and the variation of current was recorded. The current output increased and reached a steady-state within 2-3 min. The enzymatic reaction (hydrolysis of acetylcholine or butyrylcholine) was then initiated by adding 20 ~1 of 0.1 mol/l acetylcholine or butyrylcholine solution. The enzymatic activity was calculated from the slope of the current-time curve [12]. To determine the enzymatic avtivity, the sensor was first calibrated by addition of choline solution under the same experimental conditions.
298
Results Calibration of the choline sensor
As the measurement of acetylcholinesterase or butyrylcholinesterase activity is based on the monitoring of choline generated by the hydrolysis of acetylcholine or butyrylcholine, the response of the electrode to choline was first determined. As shown in Fig. 1, there was a linear relationship between the variation of current and choline concentration from 5.10-’ mol/l to 5.10-’ mol/l. The sensitivity obtained, expressed by the slope of the calibration curve was 2.6 mA * l/mol. Response of the sensor to cholinesterases standard activities
When butyrylcholinesterase was added into the measuring cell, the current increased linearly with time due to the release of choline and the slope was proportional to the enzyme activity. The linear dependence was demonstrated using different dilutions of a butyiylcholinesterase solution (120 U - ml- ‘) between 30 pI_J * ml-’ and 3 mU * ml-’ in the injected standard sample (Fig. 2a). In the same way, Fig. 2b shows the linear calibration graph obtained with a commercial solution of acetylcholinesterase (1230 U * ml-‘) between 100 PU * ml-’ and 10 mU * ml-‘. Measurements in biological samples
Figure 3 shows the response of the enzyme electrode for both acetylcholinesterase and butyrylcholinesterase activity determinations, when experiments were performed with amniotic’ fluid samples. After the probe was dipped into the reaction medium, a baseline was recorded (Section A). When a portion of amniotic
102 z 5 E P 2
10
1
1o-1
1o-8
1o-7
10-6
10-5
10-d
10-a
Choline (M)
Fig. 1. Calibration curve of the choline oxidase electrode (log scale). Measurements were in 0.1 mol/l potassium phosphate buffer, 0.1 mol/l KC1 at 27°C.
done at
pH 8,
299
‘* 7
.5 E =1
1
lo pJ
piJ
Ly,=( ; ‘;
r E $
*,j
1
I//
*,1
_,_‘_
0,Ol
0,l
1
10
100
0,oi
0‘1
(. 1
10
100
mu. ml-l mu. ml-t Fig. 2. Dependence of choline sensor response to cholinesterase activities in the injected standard sample. (a) Butyrylcholinesterase activity; (b) acetylcholinesterase activity.
fluid (200 ~1) was injected (arrow on the graph) electroactive substances such as urate and bili~bin present in the fluid were immediately oxidized at the set electrode potential, as shown in the first part of the curve (Section B). After completion, a plateau was reached and 20 ~1 of the chosen substrate solution containing either butyrylcholine (a) or acetylcholine (b) were injected to initiate either the butyrylcholinesterase or acetylcholinesterase reaction. Choline produced
2 min V-4
2 min Y
Fig. 3. Typical time courses obtained with the choline oxidase sensor in response to cholinesterase activities. (a) Butytylcholinesterase activity of 200 PI of a pathologic amniotic fluid fspida bifida) (1) or 200 &I of a non-pathologjc amniotic fluid (II); fb) acetylcholinesterase activity of 200 ~1 of a pathologic amniotic fluid (spida bifida) (I) or 200 ~1 of a non-pathologic amniotic fluid (If).
300 TABLE I Tests of specificity for acetylcholinesterase acetylcholine and butyrylcholine Enzyme
AChE BuChE BuChE + AChE Results (slope of current-time enzyme.
(AChE) and buty~lcholinesterase
(BuChE) with regard to
Substrate Acetyl choline
Bu&Wl choline
0.92 0.17 0.92
0 0.17 0.77
curves) are expressed in nA.min-’
for the same amount of each
in either case by the enzymatic reaction reacted with the immobilized choline oxidase and generated hydrogen peroxide yielding a current variation (Section 0. This variation was proportional to the activity of the cholinesterase present in the sample. After 2 or 3 min the measurement was stopped, the electrode was washed and the cell was refilled for the next analysis. The trace recording I of Fig. 3a corresponds to the butyrylcholinesterase activity of a NTD amniotic fluid while trace II corresponds to the enzymatic activity of a non-pathologic sample. A similar recording was shown in Fig. 3b for acetylcholinesterase activity determined in the same samples. It must be stressed that an immediate recognition of pathologic amniotic fluids is easy by simply examining the recording patterns for both bu~~lcholinesterase and ace~lcholinesterase activities. Validation of the method
As the amniotic fluid contains the two cholinesterases, it was necessary to verify the absence of interference between the two enzymatic reactions. Table I shows that ace~lcholinesterase hydrolyzed only ace~lcholine. Bu~~lcholinesterase however hydrolyzed acetylcholine and butyrylcholine, but with very different rates. This can be explained by the affinity and the maximal rate of the butyrylcholinesterase which is higher for its normal substrate than for acetylcholine. Obviously, as shown in Table I for a typical experiment, the presence of both buty~lcholinesterase and ace~lchoiinesterase in the solution does not modify the hydrolysis rate of acetylcholine by ace~lcholinesterase, since a value of 0.92 nA.min-’ was obtained in both cases. Furthermore, it was necessary to check that the amniotic fluid did not disturb the immobilized choline oxidase reaction. For this purpose, a portion of choline was injected into the measurement cell in the presence, or not, of amniotic fluid. Figure 4a shows that 20 ~1 of 10e3 mol/l choline induced the same current variation (2.8 nA) in the absence (I) or in the presence (II) of amniotic fluid. In the same way, hydrogen peroxide was injected in the measurement cell and Fig. 4b shows that 10 ~1 of 10e3 mol/l H,O, induced the same current variation (2.3 nA) in the absence (I) or in the presence (II) of amniotic fluid, thus demonstrating that
301
1
1nA
J J t
Hz02 * 2 min
T
amniotic fluid \
-
Fig. 4. Enzyme electrode response to: (al 20 ~1 of 10e3 mol/l choline injection in the measurement cell, without biological sample (I) or after an injection of 200 ~1 of amniotic fluid (II). (b) 10 ~1 of 10m3 mol/l H,O, injection in the measurement cell, without biological sample (11or after an injection of 200 ~1 of amniotic fluid (II).
the different components of the fluid did not interfere with the enzyme electrode response. The activities of acetylcholinesterase from different amniotic fluid samples are given in Table II. The results show that neural-tube defects (NTD: anencephaly and open spida bifida) are associated with a higher amniotic fluid acetylcholinesterase activity (12 and 7.3 respectively) compared to 3.7 mU * ml-’ for non pathologic samples. Furthermore, the presence of blood in the sample had no influence on the results of acetylcholinesterase activity (butyrylcholinesterase activity increased in these conditions, due to its presence in the blood). The technique was found reliable since repeated injections from the same sample (n = lo), a good precision was obtained with a CV value lower than 2.5%.
TABLE II Comparison of enzymatic activities determined with the enzyme electrode and with the spectrophotometric technique Samples
Normal amniotic fluid Blood stained amniotic fluid NTD (anencephalyl NTD (Spida bifidal
Enzymatic activities (mU/ml) Enzymatic electrode technique
Spectrophotometric technique
3.7*0.3 3.8 12 7.3
3.1 kO.3 3.4 12.4 7.3
302
As a control, acetyl~holinesterase activity was also determined by the conventional spectrophotometric method described by Ellman et al. [I31 and modified by Dale et al. [14]. Acetylthiocholine was used as substrate. 5,5’-dithiobis (Znitrobenzoic acid) was added to react with the released thiocholine to form the yellow compound Sthio-2-nitrobenzoic acid determined at 412 nm. The selective measurement of acetylcholinesterase activity is performed by the concomitant use of ethopropazine hydrochloride which inhibits the non-specific cholinesterase activities. Table II shows that both methods give similar results. Discussion The results presented here demonstrate that the potentialities of the amperometric choline sensor, (very low detection limit, 5.10-’ mol/l, and large linear range, up to 5.10-” mol/l) can be extented to the determination of cholinesterase activities. The separate determinations of butyrylcholinesterase and acetylcholinesterase can be performed rapidly and accurately after specific substrate hydrolyzis by monitoring choline appearance. The linear range was from 30 PIJ-ml-’ to 3 mU * ml - ’ for bu~~l~holinesterase and from 100 $I * ml - ’ to 10 mU . ml-’ for acetylcholinesterase. Cholinesterase activity assays are routinely performed in clinical laboratories but with procedures involving preincubation of the biological sample with an inhibitor in order to specifically identify the acetylcholinesterase activity. The procedure proposed here avoids this step and provides a distinct measure of the activity of the two enzymes in whole amniotic fluid through the use of specific substrates. On the contrary to spectrophotometric methods, enzyme electrodes are not subject to optical interferences and can be operated even in a turbid medium so it is possible to accurately determine the activities of specific and non-specific cholinesterases in amniotic fluid contaminated by maternal or fetal blood. In conclusion, this approach appears particularly promising for a rapid determination of acetyfcholinesterase activity in amniotic fluid. References 1 Smith AD, Wald NJ, Cuckle HS, Stirrat GM, Bobrow M, Lagercrantz H. Amniotic-fluid acetylcholinesterase as a possible diagnostic test for neural-tube defects in early pregnancy. Lancet 1979;i:685-688. 2 Barlow RD. Cuckle HS, Wald NJ. A simple method for amniotic fluid gel-acetylchoiinesterase determination, suitable for routine use in the antenatal diagnosis of open neural tube defects. Clin Chim Acta 1982;119:137-142. 3 Hay DL. lbrahim GF, Horacek I. Rapid acetylcholinesterase screening test for neural-tube defect. Clin Chem 1983;29:1065-1069. 4 Hullin DA, Laurence KM, Elder GH et al. Amniotic fluid cholinesterase measurement as a rapid method for the exclusion of fetal neural-tube defects. Lancet 1981;i:325-326. 5 King ME. Cholinesterase. In: Pesce AS, Kaplan LA eds. Methods in Clinical Chemistry. The C.V. Mosby Company. St Louis, Washington DC, Toronto. 1987;161-168. 6 Mizutani F, Tsuda K. Amperometric determination of cholinesterase with use of an immobilized enzyme electrode. Anal Chim Acta 1982;139:359-362.
303 7 Yao T. Flow injection analysis for cholinesterase in blood serum by use of a choline-sensitive electrode as an amperometric detector. Anal Chim Acta 1983;153:169-174. 8 Gruss R, Scheller F. Electrochemical determination of cholinesterase activity and indication of its inhibitors using a thick-film metallized platinum electrode. Anal Lett 1989;22:1159-1169. 9 Palleschi G, Lavagnini MG, Moscone D, Pilloton R, D’Ottavio D, Evangelisti ME. Determination of serum cholinesterase activity and dibucaine numbers by an amperometric choline sensor. Biosens Bioelec 1990;5:27-35. 10 Morelis RM, Coulet PR. Sensitive biosensor for choline and acetylcholine involving fast immobilization of a bienzyme system on a disposable membrane. Anal Chim Acta 1990;231:27-32. 11 Assolant-Vinet CH, Coulet PR. New immobilized enzyme membranes for tailor-made biosensors. Anal Lett 1986;19:875-885. 12 Blum LJ, Bertrand C, Coulet PR. Amperometric sensor and immobilized enzyme electrode for the determination of enzymatic activities. Anal Lett 1983;16:525-540. 13 Ellman GL, Courtney KD, Andres V, Featherstone RM. A new and rapid calorimetric determination of acetylcholinesterase activity. Biochem Pharmacol 1961;7:88-95. 14 Dale G, Archibald A, Bonham JR, Lowdon P. Diagnosis of neural tube defects by estimation of amniotic fluid acetylcholinesterase. Br J Obstet Gynaecol 1981;88:120-125.
